Mirko D. Grmek - Diseases in The Ancient Greek World-Johns Hopkins University Press (1991)

MirkoD.
Grmek
DISEASES IN THE
ANCIENT
GREE!( WORLD
Translated byMireille Muellner

and Leonard Muellner
THE JOHNS HOPI{INS UNIVERSITY PRESS
Baltimore and London

This book has been brought to publication with the generous assistance of
the French Ministry of Culmre and the David M. Robinson Publication
Fund.
Originally published as Les Maladies a l)aube de la civilisationoccidentale.

Copyright © Payor, Paris, 1983
English translation © 1989The Johns Hopkins University Press

All rights reserved
Printed in the United States of America
The Johns Hopkins University Press, 701 West 40th Street, Baltimore,
Maryland 212n
The Johns Hopkins Press Ltd., London
The paper used in this publication meets the minimum requirements of

American National Standard for Information Sciences-Permanence of Paper
for Printed Library Materials, ANSI Z39.48-1984.
LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA
Grmek, Mirko Drazen.

Diseases in the ancient Greek world.
Translation of: Les maladies a l'aube de la
civilisation occidentale.
Includes bibliographies and index.
I. Diseases-Greece-History. 2. Diseases
-Middle East-History. 3. Medicine, Ancient.
I. Title. [DNLM: 1. Disease-history-Greece.
2. History of Medicine, Ancient. oz n G86 G8m]
RA650.6.G8G76131989 6161 .00938 88-45412
ISBN 0-8018-2798-1(alk. paper)
Translators)Note ix
Author)s R--eface x1
Introduction: THE CONCEPTUALIZATION OF PATHOLOGICAL EVENTS I
The Conceptof Pathocoenosis 2

Pandora)sJar 4
Semantic Constantsand the Difficulties of Re-trospective
Diagnosis 6
The Constants of Human Geography 8
The GreekPathocoenosisin the ClassicalPeriod 14
The HistoricalImpact of Disease 16
Chapter One: LITERARY REFLECTIONS OF PATHOLOGICAL REALITY 17

Language as HistoricalEvidenceand the Linear B Tablets 18
Life and Death in the Homeric World 19
Health and Sicknessin Heroic Times 23
Trojan War Wounds 27
True Diseasesin Homer 33
From Hesiod to the R--esocratics 38
Lyric Poetsof the Seventh and Sixth Centuries 42
Chapter Two: PALEOPATHOLOGY: Evidencefrom Ancient Bones

on Diseasesin Greece 4 7
The HistoricalDevelopmentof Paleopathology 47
Earliest Osteoarchaeological
Researchin Greece 52
Traumas 57
Cranial Trepanning 63
The «skull of Sophocles))
and the Por"traitof Menander 65
CongenitalMalfonnations) Tumors) and Metabolic Disorders 69
Vitamin Deficiencies 75
Rheumatic Ailments 77
Notes on InfectiousDiseases 85
V
.
Vl CONTENTS
Chapter Three: PALEO DEMOGRAPHY: Evidencefrom Ancient Bones

on the Conditionsof Daily Life in Greece 87
The Dynamics of the Ancient GreekPathocoenosis 88
GeographicUni-tyand Genetic Varie-tyamong the Inhabitants of Greece 92
DemographicDensi-ty 95
Osteoarchaeological
Data on Lifespan 99
Determination of AverageLifespanfrom InscriptionalEvidence 105
Old Peopleand Maximum Longevi-ty 107
Variationsin Height and Some Consequencesof Malnu"trition 109
Paleo-odontology 113
Chapter Iour: COMMON PURULENT INFLAMMATIONS 119

The Paleopathologyof NonspecificPurulent Infections 119
Ancient Explanations of the Rlrmation of Pus 123
The First Descriptionsof Traumatic Purulent Inflammations 126
The OldestMedical Evidenceon Purulent Inflammation of Internal Organs 129
Chapter Five: THE ORIGIN AND SPREAD OF SYPHILIS 133

The B-oblemof the Uni-tyof the Treponematoses 134
HudsonJsHypothesis:A SocioculturalExplanation 135
HackettJsHypothesis:A MicrobiologicalExplanation 137
CockburnJs Views 138
The Osteoarchaeology of the Treponematoses 139
Tr. Pallidum: An American Mutant
The Mythology of VenerealPlague 142
Gonorrhea)Spermatorrhea)and Leukorrhea 144
Other Diseasesof the External Genitals 149
Chapter Six: LEPROSY: The Gradual Spreadof an Endemic Disease 152

Artistic Representationsof Leprosy 155
The OldestLiteraryEvidenceof Leprosy 157
Leprosyin Palestineand Persia 160
Lepra, Lenke, Alph6s, and the PhoenicianDiseasein the HippocraticCorpus 165
Chapter Seven:TUBERCULOSIS: A Great I(iller 177

The Clinical RJrmsand Paleopathologyof Tuberculosis 177
The OldestLiteraryEvidenceof Tuberculosis 180
B-otohistoryof Tuberculosisin America and Greece 181
The Ravagesof Consumption in the ClassicalAge 183
An EpidemiologicalObservationand Two Clinical Cases 189
..
CONTENTS vu
Notes on Consumption in the Hellenistic and &man Eras 192
Extrapulmonary Tuberculoses 194
Chapter Eight: LEPROSY AND TUBERCULOSIS: Their Biological

Relationship 198
The Epidemiology of Leprosy 200
The Epidemiology of Tuberculosis 204
Diagram of the Evolution of Mycobacteria 206
Chapter Nine: THE HARM IN BROAD BEANS: Legendand Reali-ty 210

A Pythagorean Prohibition and Its Justifications in Antiqui-ty 211
Blinkered Physicians and the Discoveryof Lathyrism 220
The Reali-ty of Favism 224
Modern Opinions on the Pythagorean Prohibition 233
Concluding Questions 240
Chapter Ten: POROTIC HYPEROSTOSIS, HEREDITARY ANEMIAS,

AND MALARIA 245
Clinical Picture and Physiopathology of the Thalassemias 249
Other Hereditary Anemias 251
Distribution of Thalassemias and Hypotheses on Their Origin 254
Thalassemia: Greek or Mongolian? 259
Distribution and Origin of Sickle Cell Anemia 264
The &le of Malaria 265
The Distribution and Paleopathological Significance of Porotic Hyperostosis 266
The Osteoarchaeologyof Mediten-anean Porotic Hyperostosis 268
Henwglobinoses in the Art and Literature of Ancient Greece 274
The Evolution of Malaria in the Eastern Mediterranean 275
Chapter Eleven: THE HIPPOCRATIC CONCEPTION OF DISEASE:

An ExemplaryClinical Report 284
The Case of Philiscus 284
The Place and Date of the Disease of Philiscus 288
The Meaning of Kausos in Hippocratic Medicine 289
Hippocratic Prognosisand Its Relationship to Diagnosis 292
Blackwater Fever: The First Case in History 295
Modern- Diagnosis of the Disease of Philiscus 300
Vlll CONTENTS
Chapter Twelve: THE CONSTITUTION OF A WINTER IN THRACE:

The cccough of Perinthus)) 305
Commentary on the Hippocratic Text and Its Translation 306
The Author) Date) and Locale of this Epidemiological &port 314
A RetrospectiveDiagnosis of the cccough of Perinthus)) 319
Some Individual Case Histories from Perinthus 328
Epilogue: Diseases ltesent in Perinthus around 400 B. c. 333
Chapter Thirteen: A DIALOGUE BETWEEN A PHILOLOGIST

AND A PHYSICIAN 340
A Case of Chronic Tetanus 341
A Case of Typhoid Fever 346
lour Concise Case Histories 350
Cases of Epilepsx Pneumonia) and Amebic Dysentery 352
Notes 357
Index 441
Translators)Note
Our chief aim has been to produce a version of the original that is idiomatic
and technically correct. The problems of translating a sophisticated inter-
disciplinary text from French into English are not slight, and we have
sought-and received-assistance from many quarters. A secondary aim has
been to make this book accessible to an audience beyond physicians and
historians of science, who perhaps constitute its primary readership, and
we hope that nonspecialists equipped with a standard medical dictionary
can read the English-language edition without difficulty.
We are truly grateful to those who have helped us in this effort-first of
all, to the author, who took great pains to fix and chasten our version as
well as to revise and correct his original. Two expert readers, Caroline
Hannaway and Dr. Gert Brieger, were chosen by the Press to review the
manuscript, and they, too, have saved us from error. The manuscript has
benefitted much in addition from the attentiveness and skill of its copy
editor, Mary Yates, as well as from the care of Eric Halpern and Barbara
Lamb. We thank the following persons for their technical advice: Dr.
Christian Baecker, Pierre-Yves Jacopin, John Kirsch, Dr. Anne-Marie
Moulin, and Dr. Robert Torchin.
-M. M. and L. M.
IX
Author)s Preface
How much uncertainty and obscurity does the passage of time cast over history,
if false displaces true even for recent events, which took place right before our
eyes?
-Plutarch, The Banquet of the SevenSages
This book is a history of diseases, not a history of medicine. It is not,

nor does it try to be, a history of professional and popular knowledge
about nosological entities, nor is it even a history of the means of combat-
ing diseases. Its purpose is to discover the pathological reality hidden
behind ancient medical and paramedical texts. The ideas people have held
in the past about diseases and ancient efforts to effect cures for them will
detain us only insofar as we can gain enlightenment about the existence of
pathogenic agents and their consequences on the individual or social level.
In fact, I will devote the greatest possible attention to certain ancient
conceptualizations of disease precisely because they can shed light on path-
ological realities. And yet I will dispense with the details of medical treat-
ment in antiquity because until the beginning of the nineteenth century,
therapeutics and prophylaxis had practically no effect whatever on the
nature and frequency of the diseases present in any society.
My initial aim was to encompass the whole of the ancient world, but
the complexity of the problems presented by such a global approach forced
me to modify my ambitions and to limit my investigations in this first
phase to the eastern Mediterranean, the so-called Greek world. I hew to
this convenient term knowing full well that, in probing prehistoric civili-
zations for· evidence or by including places such as Macedonia and the
territories of Magna Graecia, the purview of 1ny research is not, strictly
speaking, limited to the Greek populations. My choice of the Aegean as a
bridgehead for exploring the diseases of the European past is in no way
arbitrary. At least two powerful reasons dictate it: first, the antiquity and
relative abundance of its written sources and archaeological sites, and,
second, its unique geographical position, at the crossroads of three conti-
nents, which resulted in the region's playing a decisive historical role.
Texts in ancient Greek are in fact our oldest testimony from physicians
on diseases of the West. But deciphering them is not always as easy as is
often believed. Translation into a modern language, while it may be ade-
quate when it comes to an ancient author's theoretical teaching, is less so
Xl
..
Xll AUTHOR'S PREFACE
for the concrete experience on which the theory is based. In this respect,
the medical texts are even notably obscure. For instance, understanding
technical terms, particularly the names of diseases themselves, presents
grave difficulties even if we are content to accept an approximation of their
sense, that is, their translation into popular medical language as against the
terminology of modern medicine. The problem stems from the abstract
nature of the concept of disease and the partly conventional character of
medical discourse.
In this work, the mode of presentation, the investigative method, and
even the degree of historical analysis vary from one chapter to the next.
The heterogeneity is intended. It results from my wish to open new hori-
zons by using different approaches and to place in the service of history
certain recent developments in the biological and medical sciences. Despite
the diversity of its methodologies and an ampleness that surpasses my
expectations and intentions, this book cannot pretend to exhaust its sub-
ject, even within the geographical and chronological confines imposed
upon it. Not all the diseases of the past have been studied; for instance,
mental illnesses have been left to one side. It is my intention to devote
another book to plagues in antiquity and to write a series of articles on the
descriptions of pathological states in Greek tragedy and on the relationship
between myths and diseases.
Several chapters of this book have already been presented at colloquia
or published, in whole or in part, as articles. All such texts have been
reviewed, revised, elaborated, and updated. I thank the editors and direc-
tors of the journals in question; their kind permission has made possible
this reprinting and recasting.
This book would never have seen the light of day without the initiative
of Fernand Braudel at a time now distant, or without the encouragement
which that famous historian of Mediterranean civilization was willing to
lavish upon me. The lengthy gestation of this work was linked to my
teaching in the Fourth Section of the Ecole Pratique des Hautes Etudes,
and its final form is shaped by the original rules of an institution that ties
research to teaching and prefers knowledge in the making to the ready-
made kind. I am indebted to my colleagues, especially.Pierre Huard, Guy
Beaujouan, and Marc Ferro, as well as to the audience of the seminar on
the history of medicine and the biological sciences. I could hardly overes-
timate the stimulus from exchanging ideas with such interlocutors, nor
that of discussions with my philologist friends, Fernand Robert, Jacques
J ouanna, Jackie Pigeaud, and Danielle Gourevich, all so willing to share
with me their critical observations. I express to them my deepest gratitude.
For the American edition of this work, I have corrected some errors that
crept into the French version, filled out the bibliographical references with
special attention to English-language publications, and added some new
thoughts and information. These contributions are strictly lin1ited to ma-
terials published before the appearance of the original version (1983).
DISEASES IN THE
ANCIENT
GREEI( WORLD
Introduction
THE CONCEPTUALIZATION
OF PATHOLOGICAL EVENTS
It is impossible to apprehend correctly the significance of an ancient text

concerning a pathological event unless we rid ourselves as completely as
possible of the ontological notion of disease embedded in our everyday
language. Notions of disease and even of particular diseases do not flow
directly from our experience. They are explanatory models of reality, not
its constitutive elements. To put it simply, diseases exist only in the realm
of ideas. They interpret a complex empirical reality and presuppose acer-
tain medical philosophy or pathological system of reference. So, for ex-
ample, one can say that there exists a person who is sickly, coughs, spits
blood, and grows thin, and one can say that the bacillus that pervades his
organism and produces characteristic lesions on his lungs and other organs
also exists in the strict sense of the word, but one cannot say the same for
tuberculosis. Its existence is bound up with a well-articulated conceptual
structure and a particular medical ideology. 1
The history of Western medicine, as well as the comparative study of
medicine in diverse societies, shows clearly that diseases are not inevitably
conceptualized as they are nowadays. 2 How they are conceptualized de-
pends as much on the scientific sophistication of a society as on the patho-
logical realities of a given historical moment in a specific geographical area.
The Hippocratic medical teachings are rich in instances of this. The theory
of humors is at once the logical consequence of Ionian philosophy and a
faithful reflection of the pathological and clinical features of the ills actually
suffered by Mediterranean populations. If the Hippocratic doctrine of
critical days can reasonably be interpreted as the result of a desire to intro-
duce number into the explanation of nature, it is no less true that such a
notion was well supported in a region where the majority of patients suffer
I
2 DISEASES IN THE ANCIENT GREEK WORLD
from malaria or pneumonia. Bouts of tertian or quartan fever succeed each

other with perfect regularity due to the biological cycle of the parasite in
question; similarly, patients with simple pneumonia undergo a crisis right
at the end of a week of fever. A medical practitioner in Scandinavia would
never have devised a theory of acute fevers comparable to the Hippocratic
one.
Establishing a link in specific cases between observed reality and nosol-
ogy is the essence of medical diagnosis. It used to be a relatively simple
activity that never really went beyond the difficulties any process of classi-
fication gives rise to. But all that changed when the definition of disease
was transformed from a clinical to an anatomical notion. Instead of de-
scribing and classifying a disease according to its symptoms, its apparent
characteristics, one must now discover the fundamental lesion, the hidden
injury. As a result, the historian of diseases is continually troubled by the
inevitable uncertainties that arise when passing from a diagnostic system of
the first kind to a more modern one.
We have no choice but to express ourselves in the medical idiom-using
the terms and, more significantly, the concepts-of our own time. Yet
between us and Hippocrates the pathological systems of reference have
changed, not just the names of the diseases (actually, the names themselves
are remarkably static, even at the expense of a continual change in their
meaning). To put it simply, a delimitation of pathological reality based on
clinical symptoms and on a theory of disease that attributed a preponder-
ant role to the fluid parts of the organism has now been replaced by
anatomical and etiological diagnosis that sometimes even includes a molec-
ular definition of lesion-which is to say, by a set of criteria that, except
for certain surgical syndromes, was completely unavailable to ancient prac-
titioners. In order to make ancient medical texts understandable to mod-
ern readers, then, we have to resort to interpretations that take us beyond
a purely philological approach. Understanding the real pathology under-
lying a Hippocratic text can significantly elucidate its content; hence the
interest in completing our understanding of that reality by access to non-
literary sources. Unavoidably, much space in what follows will be given to
the results of anthropological and medical examination of ancient human
remains. Useful information on the diseases of the past can also be gleaned
from the study of bas-reliefs, paintings, statuettes, and votive offerings in
the shape of bodily organs. 3 Such sources offer the advantage of being
nonlinguistic and thus skirting certain interpretative pitfalls.
The Conceptof Pathocoenosis

Until very recently, the history of diseases has almost always been un-
dertaken analytically, examining separately the history of each disease or of
groups of related diseases in isolation from one another. 4 At least in the
INTRODUCTION 3
first phase of historical exploration such a simplification is necessary and
justifiable in terms of its results. Nevertheless, it obscures numerous facts
of capital significance regarding the interconnections among diseases. To
facilitate a more synthetic approach I introduce the concept of pathocoe,wsis. 5
Three propositions can serve to define this neologism: (1)Pathocoenosis

is the ensemble of pathological states present in a specific population at a
given moment in time. It consists of a system with precise structural
properties that should be studied so as to determine its nosological param-
eters in qualitative and quantitative terms. (2) The frequency and overall
distribution of each disease depends on the frequency and distribution of
all the other diseases within a given population (in addition to various
endogenous and ecological factors). (3) A pathocoenosis tends toward a
state of equilibrium expressible in relatively simple mathematical expres-
sions; that state is especially perceptible under stable ecological conditions.
Two diseases belonging to a single pathocoenosis can be in a state of
symbiosis, antagonism, or indifference to each other. Cases of symbiosis
are nu·merous. They arise either from the fact that the same environmental
conditions favor two or more diseases (for instance, the diseases resulting
from stress in a technologically advanced society), or from an etiological
link at the level of the individual (for instance, the link between angina,
rheumatic fever, and endocarditis), or to genetic synergy, or to a complex
interplay of factors operating at the level of the society and the individual
(for instance, avitaminosis, anemia of various types, typhus, or in general
the coexistence of malnutrition with certain serious infectious diseases).
In investigating pathocoenoses of the past, the results are more illumi-
nating when the pathological states in question are antagonistic. Antago-
nism between two diseases can stem from a conflict between man's genetic
state and the germ causing a specific disease (as in the relationship between
malaria and thalassemia discussed later in this book), or-this is more often
the case-conflict can appear as the last consequence of a complex chain of
heterogeneous causes. Infectious diseases due to poor hygiene and bad
water (typhoid fever, dysentery, etc.) are antagonists of so-called degener-
ative diseases (arteriosclerosis, cancer, etc.) for the simple reason that peo-
ple die of the former before they can run the risk of growing old and
contracting the latter.
In reality there is no such thing as a synergistic or antagonistic relation-
ship between two isolated diseases, only a complex interdependence of all
the diseases present in a given population. The mass of reciprocal depen-
dencies can be studied globally by looking at the distribution of diseases
by frequency in a given population during a relatively short time-frame.
The problem closely resembles that of determining the distribution of
animal and vegetable species as a function of the number of individuals
living in a biotope. Modern investigations show that the distribution of
living species due to their quantitative importance corresponds to the log
normal series, that is, the series x) x½, x 3 /3 ... , which expresses the
probability of the distribution of variations whose classes are expressed by
a geometric progression (as the Gaussian curve expresses the probability of
an arithmetic progression). Indeed, it seems that a pathocoenosis, when it
is in the state of equilibrium that is only really noticeable in a relatively
closed, ecologically stable population, presents a mathematically regular
structure. That structure corresponds to the conjunction of several kinds
of distribution but with a preponderance of the log normal type, which
therefore gives its character to the whole. 6
How can the mathematical regularity in the distribution of diseases be
accounted for if they are actually only a logical construct? In my opinion,
the regularity remains even if it is admitted that the conceptualization of
diseases is not entirely determined by direct experience, by objects in the
strictest sense of the word. However, the nosological categorization proc-
ess must be logically consistent and conform to pathological reality. The
changeover to a different, and, in its own way, consistent, conceptualiza-
tion will change the numerical data and to a certain extent the general
shape of the curve of frequencies, but the structured character of the
pathocoenosis will not disappear.
To be sure, it is impossible to apply true mathematical analysis to the
diseases of peoples as ancient as those that are the subject of this book.
The data on such distant pathological events are quantifiable only in a
uselessly hypothetical and approximate way. Nevertheless, our research
can be and should be inspired by one consequence in particular of such
analysis. It is clear, from logarithmic and log normal distributions, that
each pathocoenosis must have a small number of common diseases and a
great number of rare ones. This corollary permits-or, rather, forces-us to
discover and explore the dominant diseases in a given historical setting.
The study of a pathocoenosis consists primarily in establishing the syn-
chronic relationship among diseases during a given historical period. Yet a
static description is not enough: it is also necessary to observe the way
diseases evolve over time. Accordingly, I have in traduced the notion of
pathocoenotic dynamism. Future historians will surely scrutinize the two
most distinct phases of these diachronic processes, the times when a patho-
coenosis is in equilibrium and the times when it is in disarray, for example,
the Neolithic agricultural revolution, the beginnings of urbanism, the times
of the great migrations, colonial expansion, the industrial revolution, and
so on. But the concept of pathocoenosis is still in its infancy, and no one
can yet predict its practical value or its limitations.
Pandora)sJar
A sense of justice prevented men of old from believing that diseases had
affiicted mankind from the start. Such a terrible perquisite of the human
INTRODUCTION 5
condition could only, be due to a moral error committed by man himself.
Myths of the Golden Age or of paradise evoke a primordial time without
blame or disease. As Hesiod puts it, ''First of all the deathless gods who
dwell on Olympus made a golden race of mortal men ... they lived like
gods without sorrow of heart, remote and free from toil and grief: misera-
ble age rested not on them; but with legs and arms never failing they made
merry with feasting beyond the reach of all evils (kakon)." 7 The silver race,
of inferior biological quality, replaced the golden race, only to be replaced
in turn by the race of iron. Men of the iron race, the Boeotian poet's
contemporaries, "never rest fro1n labor and sorrow" but still "have some
good mingled with their evils." In the end they, too, will be supplanted,
by men who "have grey hair on their temples at their birth. " 8 For the
mythographer, all humanity is like a man climbing a staircase, his fading
vigor marked from one landing to the next.
Although the myths may suppose diseases were absent at the moment
of creation, no time is wasted introducing them into the world, often as
the result of an original sin: without disease the true human adventure is
unimaginable. In Greek mythology the artificial woman Pandora is pre-
sented by all the Olympian gods as a gift to Epimetheus. Actually she is a
gift to the whole human race, and she brings as her deadly dowry diseases
shut up in a jar. 9 As Hesiod says,
For ere this the tribes of men lived on earth remote and free from ills and hard toil
and heavy sicknesses (nousoi)which bring the Fates upon men; for in misery men
grow old quickly. But the woman took off the great lid of the jar with her hands
and scattered all these and her thought caused sorrow and mischief to men. Only
Hope remained there in an unbreakable home within under the rim of the great
jar and did not fly out at the door; for ere that, the lid of the jar stopped her, by
the will of Aegis-holding Zeus. 10
As I shall show in the first chapter of this book, Pandora's diseases have
the peculiar quality of affecting men without being brought on by personal
error or direct divine intervention. 11 So the myth sanctions a general feel-
ing that, mqre often than not, disease has no apparent ethical justification
at the level of the individual. Plato later rationalized the myth of the
Golden Age, asserting that in a natural state man has no need of doctors
and that most of the dis~ases in his day were due to luxury, by which he
meant laziness and too much fancy food. 12 Plutarch and a host of l~ter
moralists repeat Plato's argument that diseases are the result of the blan-
dishments of civilization. Even the medical treatises echo it. In the famous
preface to De medicina) Celsus voices an opinion that must have been
widely held in his time:
It is probable that with no aids against bad health, none the less health was
generally good because of good habits, which neither indolence (desidia)nor luxury
(luxuria) had vitiated: since it is these two which have afflicted the bodies of men,
first in Greece, and later amongst us; and hence this complex Art of Medicine, not
needed in former times, nor among other nations even now, scarcely protracts the
lives of a few of us to the verge of old age. 13
The myth of the Golden Age was founded on no true historical inquiry,
nor does it reflect any real memory of an era of well-being. It is just a faulty
notion with a psychological origin: the myth externalizes and unduly gen-
eralizes individual recollections of a strong and happy youth. But now-
adays this mythical notion of historical degradation has given way to an
idea of progress that, in its absolute form, is no less deluded.
Greek and Roman attempts to shed light on the origin of diseases in
their world never went beyond legendary tales or moralizing. Being as far
as we are from the events under study, can we moderns really do better?
The answer to this rhetorical question should be affirmative, and the pres-
ent book intends to provide some proof that it is. Despite the scarcity of
sources, the occasional opacity of written testimony, and the hypothetical
nature of most retrospective diagnoses, I believe in the possibility of recon-
structing the nosological reality of the ancient societies of the eastern Med-
iterranean. What encourages me is the certainty that, despite the historical
vicissitudes of both things and words, there must be verifiable constants in
language and ideas as well as in external reality.
Semantic Constants and the

Difficulties of RetrospectiveDiagnosis
I begin with what seem to be the most straightforward of constants,
those in the meaning of words. It is not risky to assert that the term
peripneumon(aalways designated a respiratory ailment. Moreover, it is al-
most certainly a disease of the lungs-I say "almost" because the persis-
tence of meaning in the names of diseases derived from organs is not
absolute. Hippocratic peripneumon(a may well be a disease of the same
localization as modern pneumonia, but inferences of this kind are not
always true. For instance, the Hippocatic terms nephritis and splinitis refer
to various pathological states some of which, as we now know, are not
localized in the kidneys or spleen. Roughly speaking, kark(nosor karkfnlnna
designate what we call "cancer," and etymological considerations are ad-
equate clues to the generic sense of terms like arthritis) ophthalm(a) hudrbps)
apeps(a)and so on. The meaning of terms tends to persist particularly for
symptoms or simple syndromes, like jaundice, hemorrhage, diarrhea, uri-
nary or anal incontinence, loss of hair, blindness, toothache, and so on.
Unhappily, there are also several traps. Most dangerous are the shifts in
meaning-the word remains, the concept changes. Sciatica (iskhias) is in-
deed the same disease now as in the ti1ne of Hippocrates, but the term
refers to two entirely different concepts: for them, hip pain, for us, neu-
ralgia of the sciatic nerve. Frequently such shifts in meaning are so subtle
INTRODUCTION 7
that only the most nuanced exegesis reveals their extent. For instance,
terms like phthfsiYypleuritis) apoplex(a)erus(pelas)or eile6sare used by the
ancient Greek physicians in senses that in some ways correspond to their
modern meaning but in others differ profoundly from it. The ancient
Greek word kholira designates, as it does in English, a disease marked by
vomiting, severe but blood-free diarrhea, and colics. But for the physicians
of antiquity cholera was an acute, nonspecific, gastrointestinal syndrome,
while for us it is an infection caused by Vibrio cholerae)a distinct and
particularly dangerous disease. The word has existed at least since Hippoc-
rates, yet the ailment it now designates is an endemo-epidemic disease
from India unknown in Europe until the nineteenth century. So themed-
ical historian could almost be more at ease with terms whose meaning has
changed completely (e.g., lipra and kardialg(a)or those whose usage and
significance are now corn pletely lost_(e. g., kausos)phrenztis)and kardiogm6s).
The manuscript tradition of the classical medical texts and its critical
restoration by the painstaking, tireless effort of philologists have made it
possible for modern doctors to make a retrospective diagnosis of some
diseases from symptoms described by ancient clinicians. Such a procedure
assumes that the semantic context and the biological processes in question
are unchanged, but it also assumes a certain constancy in Western medical
thinking. 14At times we can arrive at solid results without much difficulty.
There are some diseases whose retrospective diagnosis, based on the clinical
descriptions of the classical era, can be considered certain: epilepsy, mumps
(with associated inflammation of the testicles), lobar pneumonia, pulmo-
nary tuberculosis, pleural empyema, peritonitis, tetanus, dysentery, hem-
orrhoids, cancer of the breast and of the uterus, gout, acute intestinal
obstruction, ordinary fractures and dislocations, and so on. Even so, we
cannot stress too much the difficulty and frailty of retrospective diagnosis.
Always hypothetical, it is often dubious and rarely exclusive of other dis-
eases.15Where we wish to recognize in Hippocratic accounts rabies, men-
ingitis, rheumatic heart disease, anthrax, gas gangrene, adrenocortical
virilization, Beh<_;et'sdisease, gastric carcinoma, or peptic ulcers, the iden-
tification is in fact only plausible and uncertain.
The paradox is that the progress of medicine in the past few decades has
made our picture of the past even more questionable than before. Our
knowledge of the great diversity of rare pathologies makes us suspicious of
any diagnosis based on a few clinical symptoms. Such diagnostic excogita-
tions are usually dependent on an evaluation of statistical probabilities.
But for historical periods and societies whose pathocoenosis is poorly un-
derstood, we run the risk of being circular: our diagnoses only prove what
we have previously supposed was true. A patient, for instance, is said to
cough up red blood. Without hesitating we would pronounce the person
tuberculous, since the last few centuries' clinical experience has taught us
that pulmonary tuberculosis is the main cause of this sympton1. Yet it is
8 DISEASES IN THE ANCIENT GREE!{ WORLD
not the only imaginable pathogenic reality. It is not impossible that in

certain pathocoenoses of the past, the rupture of varicose veins in the
esophagus (e.g., as a function of a high incidence of cirrhosis of the liver)
was a more common pathological complication than tuberculous blood-
spitting. In the future, computers will be capable of producing almost
exhaustive lists of the diagnoses possible for a goodly number of ancient
clinical accounts, but they will not be able to match each possibility with
a statistical probability.
The Constants of Human Geqgraphy

The pathocoenosis of a given population is determined by its geograph-
ical setting, by the presence in it of pathogenic parasites and their hosts,
by its gene pool, and by its social life. Moreover, the state of affairs in a
pathocoenosis at a given historical moment depends on the previous dis-
tribution of diseases in its population. All of these factors are intercon-
nected in an extremely complex, even inextricable way.
These remarks about pathocoenosis in general can and must apply to
the study of diseases among the ancient populations of the Greek world.
Such a study presupposes that for the past few millennia biological laws,
especially those that govern pathological events, have undergone no
changes and that the properties of the human body have changed but
little. To us it seems entirely reasonable to assert the immutability of
natural laws as an epistemological postulate, at least on the scale of human
time and human space. As for the biological evolution of man, the scien-
tific evidence currently available points strongly to its slowness. Accord-
ingly, we infer that the regulatory mechanisms and pathological reactions
of the human body on all the levels of its organization are the same now
as in the past, particularly from the Neolithic period to the end of the
classical age. So an epileptic fit, an attack of apoplexy with its aphasic and
paralytic aftereffects, the spread and ulceration of a cancer, functional dis-
orders as a result of physical trauma, and even allergic reactions of the skin
must have taken place in the human of the past as they do in the human
of the present. The Hippocratic texts confirm this biological constancy.
Where differences exist between the diseases of ancient Greece and those
of today, they must stem from changes in external factors (such as food,
habitat, work and leisure, therapeutic procedures, and so on) or in the
relations between man, pathogenic germs, and their hosts. With regard to
the latter, there have been considerable changes, but it is worth noting
that they obey certain rules. Recent discoveries in immunology and epi-
demiology allow us to make some fairly plausible hypotheses about the
historical evolution of infectious diseases and to use the results in various
parts of the present book.
Certain diseases are surpassingly ubiquitous. They are so linked to essen-
INTRODUCTION 9
tial vital processes that they seem "programmed" ahead of time. Either
they derive from inevitable flaws owing to the constraints of the phyloge-
netic past, 16 or their manifestations, so harmful to the individual, provide
biological advantages to a population or even the species as a whole. In
particular I have in mind the so-called degenerative diseases, most of which
are unavoidable consequences of the aging of the human body. Plainly,
the inhabitants of Greece, just as those of the rest of the world, suffered
since time immemorial from senility and arteriosclerosis and the host of
local complications they bring.
Other, relatively less frequent diseases also look programmed, since they
affect man because of inherent flaws in the functioning of his genetic
equipment. The root of several complex clinical pictures can actually be
traced to relatively simple abnormalities in the karyotype, in either the
structure or the distribution of chromosomes. Structural anomalies are
transmitted as hereditary diseases in the strict sense. They can be localized
to the synthesis of a particular molecule necessary for the proper function-
ing of the organism ("inborn errors of metabolism"), and their frequency
is determined by a kind of equilibrium between the mutation rate and the
selection pressure of the environment. In studying hereditary anemias in
Greece, especially thalassemia, we shall see that selection can have a posi-
tive as well as a negative effect on the genetic trait. The history of favism
will serve to illustrate the persistence of such traits. When the abnormality
lies in the number of chromosomes, the resultant disease is innate but
not necessarily hereditary. The standard example is Down's syndrome
(mongolism), a deformity accompanied by mental retardation that is due
to one supernumerary autosomal chromosome in the fertilized ovum (tri-
somy 21).
Biological considerations lead us to believe that the risk of an abnormal-
ity due to the absence of disjunction of a chromosomic pair during meiosis
did not change between antiquity and the beginning of the modern dem-
ographic explosion. In other words, the frequency of Down's syndrome
and sex chromosome abnormalities was about the same in the classical age
and the eighteenth century. The only factor known to change the fre-
quency of Down's syndrome is the age of the mother at conception. This
consequence of maternal aging was uncovered in statistical analyses by
L. S. Penrose in 1934, and subsequent research has indeed confirmed his
finding. Among the present populations of Europe, the overall frequency
of Down's syndrome is 1 in about 600 live births. If the mother is younger
than 30 years old, the rate drops to 1 in 2,500. Accordingly, one can
suppose that in archaic and classical Greece at least 1 newborn in 2,000-
2,500 had trisomy 21. As for trisomies affecting the sex chromosomes, the
frequency of Klinefelter's syndrome in boys could not have been far from
1 in 500-1,000 male births; likewise, the frequency of Turner's syndrome
in girls should have been 1 case in about 2,000-3,000 female births.
I hesitate to supply figures for diseases caused by an inborn error of

metabolism. Their frequency is determined by the genetic load, something
that varies from population to population, and the mutation rate, which
depends, in turn, on environmental factors like radiation and chemical
pollution. However, certain of these molecular diseases now occur with a
distinctly different frequency than the mutation rate, a fact that forces us
to assume their existence in the remote past of the human race. Phenyl-
ketonuria is a good example of this. It is an autosomal, recessive disease
clinically manifested as serious mental disorders sometimes accompanied
by epileptic fits. First described by I. A. Foiling in 1934, this hereditary
disease strikes approximately I newborn in 10,000. Its mutation rate has
been calculated at 25 per million loci. It must be a very old inheritance,
since its frequency is at once too high and too consistent among all current
Caucasian populations to be explained as the result of recent mutations.
The same inference can be made about mucoviscidosis or cystic fibrosis of
the pancreas, a disease now occurring in Europe at the rate of I newborn
per 2,000-2,500 or about hereditary deaf-mutism and albinism, even
though the frequencies of the latter vary considerably among different
populations.
In a paper presented in 1980 to the Paleopathology Congress at Caen,
V alfredo Capecchi maintained that congenital clubfoot is due to ''a genetic
mutation at least 100,000 years old, that is, pre-existing the glaciation of
Wurm." He reached this conclusion from two premises, namely, that (1)
"this deformity occurs throughout humanity, irrespective of race, at a rate
of approximately 0.12 percent," and (2) "its recessive hereditary transmis-
sion is considered proven." Its selective disadvantage, especially for a pri-
mordial society of hunters, would have been offset by the attribution of
magical powers to its carriers. In this regard one can cite the mythological
deformity of Hephaestus. 17
Regrettably, Capecchi's argument is not probative. If we leave aside his
incomplete and too approximate statistical data on the worldwide distri-
bution of clubfoot, his inferences concerning its past are very weak since
its etiology is not simple and unique. Long ago, Hippocratic physicians
saw that congenital clubfoot could derive from a lesioµ developed during
intrauterine life. 18 Not all occurring cases are of chromosomic origin, and
even when they are, the defect is polygenic and can be attributed not to
the antiquity of the genes responsible but to some relatively recent muta-
tions. So the trait's advantage would not consist in some link between its
external, phenotypic expression and the environment, such as social pro-
tection justified by magical beliefs, but in some particular, internal feature
of the human genome itself, whose structural stability might demand
weakness in specific genetic loci.
It seems to me that a similar argument could be made about some
hereditary forms of diabetes and other endocrine disorders. Such a hypoth-
INTRODUCTION II
esis would not allow us to judge the real impact of these diseases within
the pathocoenos<!s of the past, nor could we assert that their frequency is
unchanged, but it would suffice to establish their existence at the dawn of
human history. Although the Hippocratic writings make mention of club-
foot and know certain diseases are hereditary (generalized epilepsy, pul-
monary consumption, strabismus, among others), 19 the truth is that most
congenital diseases were unknown to ancient physicians. The causes of this
were, above all, the faint interest in medical treatment of very young
children and the extremely high childbirth and infant mortality rate; to-
gether, they simply prevented doctors from seeing such diseases. 20
Nevertheless, certain of them have such typical symptoms lasting on
into adulthood that the silence accorded them by the physicians of antiq-
uity can be considered meaningful. Hemophilia is a good example. It is a
hereditary enzyme disorder transmitted recessively and linked to gender.
Disorders in the blood coagulation system inevitably result in serious hem-
orrhages. The two classic forms of this disease appear only in males, not
necessarily from birth but generally soon after, for instance, when the child
learns to walk. In the past, hemophilic boys usually died of accidents
before reaching puberty. But the transmission of the disease was main-
tained by females as heterozygous carriers of the gene that their own bodies
did not express.
The literature of the Greco-Roman world seems to contain no mention
of the kind of hemorrhagic episode that suggests hemophilia, and the
silence persists in medieval writings from Byzantium and western Europe.
By contrast, Jewish and Arabic writers of this period offer several unim-
peachable descriptions of the disease. The Talmud notes not only that it
was hereditary but also that it was transmitted in the fen1ale line. Further-
more, nineteenth-century doctors remark on the absence, or at least the
extreme rarity, of family-linked hemophilic syndrome among native inhab-
itants of Greece, Italy, and Turkey, and at the same time they note its
frequency among Jews and descendants of the northern barbarians. The
inescapable conclusion seems to be that hemophilia did not exist in ancient
Greece and Rome. 21
But things are not so simple. Current research suggests that hemophilia
has a particularly high mutation rate. The trait must have appeared rela-
tively often without familial antecedents, and in ancient populations it
could only have produced short-lived hereditary lines. Among Jews and
Islamic peoples the practice of circumcision facilitated its early discovery.
However, none of these considerations explain the important distinctions
that exist in the geographic distribution of hemophilia.
In fact there is no way to understand actual pathological events until we
thoroughly grasp the idea that most diseases result from the conjunction
of an innate weakness with a wide variety of environmental factors. In the
great majority of cases, diseases occur at the intersection of two causal
12 DISEASES IN THE ANCIENT GREE!( WORLD
chains, one genetic and the other external. Consider hypertension as an

example: the frequency of this disease-an important one in a pathocoe-
nosis because of its multiple complications-increases with the quantity of
salt- in food and with stress, but it is also genetically determined by a
particular permeability of the cellular membranes to sodium and potas-
sium. The qisease is an organism's response to attack, so its occurrence
depends as much on the force of the attackers as on the vulnerability of
the attacked. This kind of twofold causality is apparent as well in certain
forms of cancer. In this regard Jacques Ruffie has stressed the difference in
the frequency of carcinoma of the nasopharynx in Asia where it is com-
mon, and in Europe, where it is rare. In Macao it occurs among the
Chinese but never the Portuguese. It is thought that the difference is due
to racial differences, particularly as concerns tissular antigens, but even so
environment is not blameless in the etiology of this disease (distinctive
diet, viral infections, and so forth).
The factor of race is sometimes adduced for diseases whose causes are
still unclear. It has been said, for instance, that hemorrhoids are especially
common among inhabitants of the Mediterranean basin, but no serious
study has been able to prove it. Certain societal habits, such as sitting for
long periods of time on high, hard chairs, favor this disease in Europe
nowadays. But that did not prevent the inhabitants of classical and proba-
bly also archaic Greece from having it too although they did not lead the
sedentary life of a modern civil servant. We recall that a whole treatise of
the Hippocratic corpus is devoted to hemorrhoids, and the disease is often
mentioned in other classical texts. 22
Diabetes, a disease whose early history needs further study, offers a fine
example of double causality. As an inherited trait, diabetes is indeed an-
cient, doubtless as old as the human race. Yet it is possible that for a long
time this inborn error of metabolism was advantageous or at least had no
practical consequences in most cases because there was no excess of sugar.
Beekeeping had to be invented for diabetes to become a true disease. Once
there was regular honey production, man had the possibility of putting
the functional capacity of his pancreas to the test. But changes of a social
nature had to take place, along with modifications in the eating habits of
several levels of society, before diabetes could appear to be a common
disease. Unknown in the time of Hippocrates, probably because of its
rarity, diabetes mellitus becomes an important disease for medical writers
of the Imperial period (Aretaeus, Celsus, Galen). 23
I note in passing that, according to modern knowledge of its pathogeny,
another disease of the endocrine glands, exophthalmic gaiter (Graves or
Basedow's disease), must have occurred sporadically. The silence of ancient
Greek writings in regard to it is surprising to us. As for endemic (hypothy-
roid) gaiter, today we know it is linked with iodine deficiency and proba-
bly associated with nutritional and genetic factors. According to V. M.
INTRODUCTION 13
Goldschmidt, the decline in iodine was due to the expansion of the gla-
ciers. In fact, F. 'Merke has shown that the confines of the worldwide
distribution of endemic goiter and of its derivative, cretinism, correspond
to the furthest extension of the glaciers during the most recent stage of
glaciation. 24 So it must have always been rare in Greece. And indeed, this
kind of goiter was described in antiquity only by Latin authors well in-
formed on the Alpine regions.
Despite the constancies we have just instanced, diseases do change not
only in their frequency but also in their clinical aspect and in certain
epidemiological peculiarities. It is worth stressing that these changes are
especially common for two groups of diseases: those caused by germs
(viruses, bacteria, protozoa) and those stemming from environmental
changes introduced by man. To be sure, the first of these groups is far
more important than all other diseases combined. In the past, infectious
diseases completely dominated the pathocoenosis. In today's world, the
second group has had an unequaled impact on human health. 25
The history of infectious diseases could be told in epic style. 26 One could
speak of legions of corpses and ineffable suffering, of an incredible variety
of microbes and their biological plasticity, of population density and the
progressive acquisition of genetic resistance to common germs, of violent
responses to relatively new or rare germs, of struggles between diseases, of
the presence of animal carriers and germ reservoirs, and so on. As for killer
diseases, theirs is a terrifying drama with a varied cast: humanity, the
parasite, the parasite's enemies, the carriers. After the initial invasion, there
is a long period of mutual adaptation between host and germ. Bubonic
plague, for instance, is only secondarily a human disease, and that only for
short periods of time, since it destroys its own biological props too radi-
cally. A human can only be a subsidiary host ofYersin's bacillus. Actually,
plague is a disease of rodents, among which it has achieved biological
equilibrium between parasite and host.
Syphilis today is not what it was in the recent past, but still less is it like
sixteenth-ce.ntury syphilis, a disease whose violent clinical manifestations
resemble smallpox. In this book we will study the immunological phe-
nomena that typify humanity's adaptation to germs as well as some of the
germs' strategies against biological defenses. Sometimes such study yields
results about past epidemiological states. For instance, yellow fever _is a
relatively benign disease for Africans. When introduced in modern times
in Greece, it took a terrible toll on the inhabitants. One can only conclude
that their ancestors had no opportunity to build a biological defense.
I insist on the fact that changes have taken place for just two particular
groups of diseases in order to make it clear that the spectacular aspect of
those changes is only very recent. Until the nineteenth century, humanity
did not know how to combat infectious agents directly. As for the way
Greek peasants live, that has changed more in the past fifty years than it
did in the previous three millennia. 27 A modern medical inquiry, such as

the one recorded by Richard and Eva Blum during their fieldwork in
Greece from 1957 to 1962, can produce useful information and unexpected
insights even for a specialist in classical medicine. The inhabitants of three
Greek villages were interviewed and examined by this American medical
team and were found to be suffering from the following important ail-
ments: malaria, typhoid fever, amebic dysentery, pulmonary tuberculosis,
scrofula, diabetes, jaundice and hepatitis, pneumonia, meningitis, diph-
theria, undulant fever, scarlet fever, cataracts, and eye diseases. Acute
respiratory diseases, gastrointestinal disorders, and rheumatic pains consti-
tuted the overwhelming majority of ills seen by the team in this population
of peasants and shepherds. If we include the relatively frequent diagnosis
of nutritional dystrophy, high blood pressure, tonsillitis, peptic ulcers,
hernia, gout, and sciatica, it is striking how closely this pathocoenosis
resembles the one suggested to us by reading the Hippocratic texts. Aside
from a decline in malaria, the villages of Dhadhi, Panorio, and Saratzani
confronted these American researchers with a nosological reality essentially
the same as the one confronting an itinerant physician in the fifth cen-
tury B.C. 28
The GreekPathocoenosisin the ClassicalPeriod

Texts offering direct evidence of the relative frequency of diseases in
antiquity are very rare. I cite three of them here in order to give the reader
a general orientation, since the unavoidably analytic discussions in most of
the subsequent chapters of this book can obscure a wider perspective.
According to the Regimen of Acute Diseases)a Hippocratic treatise from
the last third of the fifth century B.c., acute diseases "kill the greatest
number of people. " 29 The rest of the passage makes clear what diseases it
concerns: '' Acute diseases are those which men of old have called pleurttis)
peripneumonia)lftha,;gos) 30
kausos)and all others dependent on them whose
fevers are generally unremittent. In the absence of an epidemic of some
kind of pestilence (loim~clisnousos),when diseases are sporadic, as many or
more people die of acute diseases as of the rest. " 31
As for the frequency of nonfatal diseases, a valuable piece of evidence is
provided by Plato, who puts the following words in Socrates' mouth when
he wishes to show that the whole of a concept must not be confused with
its most common elements: "Does it seem to you necessary that every sick
person be gouty or feverish or afflicted with ophthalmia? Can't one be sick
without having one of these diseases, since there are many others and they
are not the only ones?" 32
The Hippocratic writings offer profuse details, notably in the description
of the seasonal morbidity on the island of Thasos 33 and in the following,
more general medical-geographic evaluation:
INTRODUCTION 15
A city that lies exposed to the hot winds . . The heads of the inhabitants are
moist and full of p]llegm that runs down into them from the head . . . The
endemic diseases are these. In the first place, the women are unhealthy and subject
to excessive fluxes. Then many are barren through disease and not by nature, while
abortions are frequent. Children are liable to convulsions and asthma, and to what
they think causes the disease of childhood, and to be a sacred disease. Men suffer
from dysentery, diarrhoea, ague, chronic fevers in winter, many attacks of eczema
[epinuctfs) a particular kind of dark pustule], and from hemorrhoids. Cases of
pleurisy, pneumonia, ardent fever [kausos], and of diseases considered acute, rarely
occur. These diseases cannot prevail where the bowels are loose. Inflammations of
the eyes occur with running, but are not serious; they are of short duration, unless
a general epidemic takes place after a violent change. When they are more than
fifty years old, they are paralyzed by catarrhs supervening from the brain, when
the sun suddenly strikes their head or they are chilled. These are their endemic
diseases, but besides, they are liable to any epidemic disease that prevails through
the change of seasons.
But the following is the condition of cities with the opposite situation, facing
the cold winds that blow from between the summer setting and the summer rising
of the sun, being habitually exposed to these winds, but sheltered from the hot
winds and from the south ... The natives must be sinewy and spare ... Their
endemic diseases are as follow. Pleurisies are common, likewise those diseases
which are accounted acute. It must be so, since their digestive organs are hard,
and the slightest cause inevitably produces in many patients abscesses, the result of
the stiff body and hard digestive organs. For their dryness, combined with the
coldness of the water, makes them liable to internal lacerations. Inflammations of
the eyes occur at last; they are hard and violent, and rapidly cause rupture of the
eyes ... Instances of the diseases called "sacred" are rare but violent ... As to
the women, ... their menstrual discharges are not healthy, but scanty and bad.
Then childbirth is difficult, though abortion is rare. After bearing children they
cannot rear them, for their milk is dried up ... cases of phthisis are frequent after
parturition, for the violence of it causes ruptures and tears. Children suffer from
dropsies in the testicles while they are little, which disappear as they grow older. 34
If we are not distracted by the essentially false etiological discussions of the

Hippocratic writer, we can look at his text as a deposition of exceptional
value on the nosological reality of long ago. It is a faithful account of what
itinerant Greek physicians of the fifth century B.C. really came across in
the regular practice of their profession. This is not to say it is a faithful
account of what actually took place in the bodies of their patients. At least
two factors come into play to deform this testimony. The first is dogmatic
prejudice, and the second is what one might call the invisibility of the
familiar. An anecdote can illustrate what we mean here. In an ancient
anthology of funny stories, we find the following joke: "We were talking
about indigestion (apepsfa),and the backslapper said he never got it. But
when we asked him if he'd ever a sour or nasty-tasting burp, he said, 'That
I get every day.' " 35 As for the doctrinal obstacles, it is enough to note, as
an example, that their medical theories prevented Greek and Roman prac-
titioners from perceiving the true frequency of heart disease.
As we do, Greek doctors distinguished endemic, epidemic, and sporadic
r6 DISEASES IN THE ANCIENT GREEI( WORLD
diseases. The most awful endemic diseases were, in modern terminology,

malaria, tuberculosis, acute forms of gastroenteritis (particularly typhoid
fever and other diseases caused by salmonellae and enteroviruses, bacillary
dysentery, and amebiasis), ophthalmias, and skin diseases, especially pu-
rulent ones. These diseases were so frequent that the ancient doctors di-
vided them into a great number of distinct nosological entities, some
epidemic, others sporadic. Diseases from contaminated water must have
been the principal cause of the elevated infant mortality rate.
The HistoricalImpact of Disease

Disease is one of the factors forging human destiny. Its importance in
the footnotes of history, in the private life of everyone, is a truism, yet
biographers inhibited by exaggerated modesty have not always been equal
to its scope. By studying the effects of disease on the public activity of
some famous persons, an attempt has also been made to introduce pathol-
ogy· into history proper. 36 Couldn't the diseases of leaders determine the
lot of their peoples? Greek historians, from Herodotus to Plutarch, were
convinced of it, and the idea is not inappropriate to an era when the
personal genius of a few great men was granted a preponderant role on the
stage of history. What would the destiny of the ancient world have been if
Nikias had not been sick during the Athenian expedition to Sicily, or if
Alexander had not abused alcohol and gone to an early grave from malaria?
These are truly naive questions, on the order of Pascal's famous joke about
Cleopatra's nose and Cromwell's kidney stone. 37
Traditional history aims to describe the most extraordinary political,
military, and cultural events and to give homage to "great men," but its
proper opposite is a more complete vision of the past of humanity, a
history of daily life, a sociological approach especially interested in the fate
of the "common man." In this global historical vision, diseases take on
prime significance as mass phenomena affecting economics, demography,
and social behavior. 38 There are epidemic diseases, plagues whose terrible
consequences are so stunning and vast that their role in history escapes no
one. I note that the best accounts of exceptionally serious epidemics in
antiquity are not provided by professional physicians but by historians or
other men of letters. It is enough to mention Thucydides' account of the
"Great Plague," which was an event of decisive importance for the out-
come of the Peloponnesian War and the future of Athenian imperialism. 39
The significance of such disasters needs no proving, but the same is not
true of endemic diseases, with their slow evolution. Yet they affect the
biological potential of a society much more profoundly and in more en-
during ways.
Chapter One
LITERARY REFLECTIONS OF
PATHOLOGICAL REALITY
The Hippocratic corpus is a collection of about sixty fairly heterogeneous

medical treatises traditionally attributed to Hippocrates (ea. 460-377 B.c.),
the most prestigious member of the guild of Asclepiadae on the island of
Cos. Together they make up the oldest professional account of diseases in
the Greek world. The dates of composition vary from one treatise to the
next, but no one of them is older than the fifth century B.c. Since my
purpose is to go as far back in time as possible, I am obliged to appeal to
other written sources that are more venerable but much less relevant to
my actual subject. This means scouring archaic inscriptions and, above all,
what survives in Greek from the oldest poetic, historical, religious, philo-
sophical, and legal works.
In the Introduction I stressed that the modern reader must confront a
special obstacle to understanding ancient names of various diseases, the
fact that they conceptualize morbid states in a way that is not our own.
This change in the system of reference for the terminology of disease is a
hindrance even in the highly technical texts from antiquity in which it is
possible to glimpse a link between concept and clinical reality. In other
chapters of this book there are several concrete instances of the difficulties
to be faced in the modern interpretation of diseases described by Greek
medical treatises. Clearly, the task of the historian of disease is that much
more arduous when he must track down the pathological states of the past
with nothing more to guide him than literary passages that refer to them
incidentally, as subjects of casual concern, and usually without mention of
symptoms. The project that awaits is a kind of "linguistic archaeology,"
with all that such a locution implies with respect to both significance and
frailty. It would be hard to overstate the pitfalls in this field of study.
17
Language as Historical Evidenceand the Linear B Tablets

In itself language is a source of information about realities that can be
much older than the moment when certain words are attested for the first
time in surviving documents. The Greek language as we know it through
inscriptions and the oldest literary works appears and develops in the
Mediterranean world after invasions of the Balkan peninsula by people
speaking an Inda-European language. The arrival of the tribes who made
the decisive contribution to the birth of the "Greeks" took place in the
Bronze Age, more precisely toward the end of Old Helladic II, in the last
centuries of the third millennium B.C. The Greek language developed from
the idiom of the newcomers and of the indigenous peoples. The Linear B
inscriptions prove that this language was dominant no later than the time
of the expansion of Mycenaean civilization, that is, by the middle of the
second millennium B.c. It is interesting to note that, according to special-
ists, the majority of Greek medical terms are of Inda-European origin. 1
This is especially the case for the ancient vocabulary of pathology. 2 In fact
there is no Greek word for a disease, disability, or symptom that is securely
identified as of proto-Balkan, Semitic, or Egyptian origin. 3 The richness of
Greek terms inherited from Inda-European is particularly striking for visi-
ble disabilities and ailments on the surface of the human body: deformities,
paralyses, injuries, tumors, hernias, a very wide variety of skin and eye
afllictions. The same is true of the main intestinal and cutaneous parasites.
The number of words referring to hunchback and other chronic trunk and
limb deformities goes way beyond the normal need for synonyms. There
was a refined differentiation among such states that reflects their social
significance. 4
Since it contains no identifiable proto-Balkan medical terms, the Greek
language offers no information on the diseases found by the Inda-Euro-
pean tribes on arriving in their new homeland. Instead, it preserves the
memory of those that the invaders brought with them. Even so, care is
necessary in applying this sort of reasoning to concrete philological analy-
sis: we must remember that the Greek language is hospitable to the crea-
tion of neologisms formed from a stock of old lex~mes whose meanings
modulate with changed circumstances. The fact that a Greek word has an
Inda-European root does not always guarantee its antiquity or, a fortiori)
that of the concept it is thought to signify.
The decipherment of Linear B by M. Ventris and J. Chadwick in 1952
has made available Mycenaean epigraphy, a treasure-house of Greek words
from the Late Bronze Age, in particular the period from the fifteenth to
the twelfth century B.c. It is now possible to read the inscriptions found
in great numbers at Mycenae, Knossos, Pylos, and, more recently, Thebes.
Most of the texts deciphered are inventories that relate almost exclusively
LITERARY REFLECTIONS OF PATHOLOGICAL REALITY 19
to the administration of the palaces; unfortunately, none concerns medi-

cine directly. 5 From reading these documents the medical historian can
recover a few simple terms of external anatomy and glean some informa-
tion concerning diet and various medicinal plants. 6 Among the occupa-
tions mentioned, he will find an unguent-boiler, a female bath-keeper,
and, to be sure, a doctor. 7 But he will search in vain for the names of
diseases or disabilities. Nevertheless, it may well be that a terminology of
pathology is hidden in the onomastic evidence. In most languages, there
are hypocoristic personal names with a negative tinge that allude to an
individual's (or his ancestor's) physical idiosyncrasy. According to Oscar
Landau, author of a meticulous study of Mycenaean personal names, the
proper name I(u-ra-no is derived from the same root as kulMs 'crippled,
deformed,' while I(u-jo and I(u-ja-ro mean 'lame,' A-ra-i-jo 'thin,' A-pa-
u-ro 'weak,' Pa-ra-ro 'bald,' No-da-ro 'toothless,' and Nu-o 'dumb. ' 8 The
name No-sa-ro attested on a tablet from Knossos is etymologically related
to n6sosand would have once designated a sickly person; it also constitutes
a proof of the great antiquity of the generic term in Greek for disease. 9
Life and Death in the Homeric World

In the second half of the eighth century B.c. the oldest work of Greek
literature, the Iliad) assumed the form it has today, with the exception of
a few interpolations. The Odyssexprobably but not necessarily composed
by the same poet, appeared a little later. The events related in the two
epics 10 are for the most part imaginary, but it is generally accepted, espe-
cially since Schliemann' s excavations, that the historical background is not
wholly fictitious. Achaean warrior kings did rule over peasant populations
in the geographic sites mentioned by the poetic narrative. They did indeed
wage a war against Troy, and they destroyed that city toward the last third
of the twelfth century B.C. Using a general framework transmitted orally
from generation to generation, a genius poet was able to create a dazzling
world in which gods, heroes, and plain mortals undertake the most varied
activities and in which the ordinary and the marvelous, the mundane and
the sublime, seamlessly commingle. 11
In this epic world, medicine has a privileged place, but only one of its
aspects will preoccupy me here: the way pathological states are concep~u-
alized and the historical reality -that underlies them. I pass over in silence
the Homeric understanding of biological phenomena in general as well as
pharmacology, the treatment of wounds, hygiene, the identity and social
status of doctors and healers, and so on. 12 Thorny questions about the real
existence of Homer as well as the genesis and unitary or composite char-
acter of the poems ascribed to him are not my concern. Nevertheless,
insofar as I accept the Iliad and the Odysseyas sources of information on
20 DISEASES IN THE ANCIENT GREEK: WORLD
some aspects of daily life in Greece, it is incumbent on me to make plain

my opinion regarding the historical era pertinent to the Homeric evidence
that does interest me.
The Homeric world is formidably ambiguous from a chronological point
of view. Properly speaking, the story takes place in the grand days of the
Mycenaean kingdom. But what is actually the case? To be sure, the repre-
sentation of material and social conditions, of customs and beliefs, is delib-
erately archaizing. The bard wished to convey to his hearers a vision of an
old-fashioned civilization, of a lost heroic age. But after four dark centuries
in which even the memory of Mycenaean writing was lost, what was still
known about the actual way of life of Achaean society before the Dorian
invasion? Aside from a few broad strokes of history, the poet evoking this
glorious past was free to invent what he wished. And he could not do
otherwise, since the object of his enterprise was essentially aesthetic and
pedagogical. Even so, as far as daily life is concerned, for the material details
of existence he was dependent on the state of knowledge of his own time.
The believability of his narrative was based on its everyday realism. The
poet could introduce the extraordinary and revel in fantastic events, but
only if he carefully respected part of the daily experience of his public.
Such veracity was particularly important with regard to my object of study,
namely, the sufferings of the human body, diseases, and medical practices.
When touching upon such subjects, which are frankly secondary in the
epic purview, the Ionian bard could not deviate perceptibly from his own
historical and geographic milieu. 13
The Iliad begins by invoking the funeral pyres on which the corpses of
countless warriors slain by the arrows of Apollo are burning, and it con-
cludes with a description of the funeral of Hector. Death is present in it at
every moment. 14 Nor does its author tire of relating, in detail and without
fear of repeating the most significant
,.
expressions, how a sort of vital prin-
ciple (thum6s) and the soul (psukhe) 15 quit the bodies of those who succumb
to their wounds: a dark cloud covers the dying man, his nostrils cease to
inhale air, his vision swirls and black night veils his eyes, his knees bend
and give way, he sleeps a sleep of bronze and breathes out life through his
mouth or a wound. 16 Sarpedon, wounded by Patroclus ''where the phrenes
encircle the stout heart," falls "like an oak," calls out to his companion,
Glaucus, and "the end of death (thanatos) covers his eyes and nostrils."
Then Patroclus places his foot on his victim's chest, pulls on the bronze
spear, and "draws out at once Sarpedon's soul and the point of his
spear. " 17
On the physical side, Homer is aware not only of simple facts like the
fatal results of a massive hemorrhage but also of the rare and striking
phenomenon called cataleptic rigidity: Mydon, the squire and charioteer
of a Paphlagonian chieftain, dies almost instantly after a sword blow to the
temple. His body falls head first from the chariot, sticks well into the sand,
and "stays upright" for some time" until his horses knock it down on the
ground. 18 As for psychological phenomena, Homer describes men's para-
lyzing terror in the face of imminent death and stresses the fading sense
perceptions and progressive loss of consciousness in the dying.
When Hector is killed by a bronze lance through his neck, "his soul
leaves his limbs and goes off, flitting to Hades', bemoaning its fate, aban-
doning its strength and youth. '' 19 The psukh~ that escapes from the body
persists after death and leads a pitiful existence in the underworld, like a
shadow or faded image of the living person. That is where Odysseus learns
from the shade of his mother, Anticleia, what the profound reality of
death is in the Homeric scheme of things:
That is the law of mortal men, when they die:

the tendons (tnes) no longer hold together the bones and the flesh;
the mighty fury of 1/re destroys all, once life (thum6s) has left the calcined bones
and the soul (psukhe) has flown away like a dream. 20
The psukh~ of a man, declares Achilles in his passionate speech to Odysseus,

can neither return nor be seized or captured "once it has crossed the
barrier of one's teeth. " 21 Once breathed out, life can never be regained.
But despite death's definitiveness, there are states comparable to it that are
entirely reversible: Homer describes several cases of syncope. 22 The poet
conceives of it as a provisional loss of breath, in other words, a temporary
departure of the vital principle manifested by a sudden dimming of con-
sciousness and an overall weakness in the limbs-serious symptoms, but
brief ones. Such syncopes result either from a physical injury, for example,
the spear-shot that wounds Sarpedon in the thigh or the boulder-throw
that crushes Aeneas's hip or smites Rector's breastbone, 23 or a psycholog-
ical shock, as in Andromache's emotional response to Rector's corpse or
Laertes recognizing Odysseus. 24
To return to the subject of death: the Homeric world clearly distin-
guishes four _types-namely, death by overt violence (combat, accident,
sacrifice), death as the result of debilitating disease, sudden death without
apparent external cause, and death from grief. The first type, violent death,
predominates throughou~ ancient poetic and historical texts. It fascinates
by virtue of its tragic dimension. Its frequency is extremely high in war
narratives (as in the Iliad), but that does not mean it is considered the
most common way to die in peacetime conditions. Accidents while hunt-
ing or working, drownings, and fatal falls could not have been rare. They
are only rarely spoken of because they are humdrum. I recall the story of
Elpenor, one of Odysseus's companions, who drank too much and went
out on the roof to get some air: falling head first, he broke his cervical
vertebrae and died immediately. 25
Whom the gods love dies young: such is the sentiment of a poet who
abhors death caused by chronic disease. But the third type, sudden death,
stands between violent death and natural death from disease. Sudden death
without visible external cause (i.e., from the modern point of view, natural
death from acute internal disease) is, for ~omer, simply violent death by
divine intervention. When Menelaus returns home, at the moment his
ship touches Sunium, the sacred promontory off Athens, the pilot Phron-
tis dies holding in his hands the steering oar of the running ship. Though
modern medicine is hesitant in the face of such an event and can only
cautiously suggest the possibility of a vascular mishap affecting heart or
brain, Homer is sure of the facts and expresses his etiological diagnosis
clearly through the mouth of Nestor: the valiant pilot of Menelaus's ship
succumbed to the gentle arrows of Apollo. 26 In wishing death on some-
one, the convention was to cry out, "May Apollo strike him today!" 27
But not wanting to make the heavenly bowman a murderer of women,
the Greeks attributed to Artemis the power, not to call it the job, of
destroying with her arrows persons of her own sex. In the myth of Niobe
as Homer tells it, that proud mother lost her twelve children in one day,
Apollo killing the boys and Artemis the girls. 28 If pestilence (loim6s) rages
in the Achaean army before Troy, the reason is, according to Homer, the
wrath of Apollo who fires his arrows at their mules, their dogs, and finally
at the men. 29 So sudden death is a sign of divine anger. Yet the epic
tradition is not unequivocal on this point. The usual, accepted epithet of
these murderous arrows is "gentle." The fact is that in this archaic world,
long suffering is more to be feared than sudden death.
For simple mortals, the interventions of Apollo and Artemis were of the
order of the invisible, but they also knew of a kind of divine punishment
that put before their very eyes the reality of such phenomena, namely,
thunderbolts. Zeus, master of lightning, so slew Asclepius and, above all,
Capaneus, one of the seven Argive chieftains to attack Thebes. Such tales
are known only from relatively late texts, but they assuredly belong to the
distant past of the Greeks.
As for the fourth type, death provoked by grief, it is enough to cite the
answer that Anticleia's shade provides to her son's, question about the
reason for her demise:
It was not the unerring archer, Artemis, who slew me at home with
her gentle arrows; neither was it a disease that drives out the
thum6s by the hideous consumption of flesh. It was my
regret, my worries, noble Odysseus, my longing for you that stole
the honey-sweet thum6s from me. 30
I have not dared translate the term thum6s in this passage, since its poly-
semy is deployed advisedly by the poet. Suffice it to say that besides its
more abstract sense, it connotes the spark of life.
Health and Sickness in Heroic Times

Each individual human being is the result of inborn and acquired fac-
tors, the product of both heredity and environment. A human is like any
living organism, carrying within itself the combined history of its ancestry
and of its own formation. The human organism's biological quality and
state of health depend on them. Early on, the Greeks understood fully
these two aspects of man's rootedness. According to the oldest epic and
dramatic poets, a person's health and predisposition to disease and death
are determined not only by the unforeseeable will of the gods but also by
blood and by the climate of the place where he or she is born and lives.
Blood must be understood to mean the stock or biological link between
the members of a family conceived as a succession of generations. When
Glaucus, the son of Hippolochus, presents himself to Diomedes, he lists
his ancestors and proudly declares, ''That is the stock, the blood from
which I claim to be descended. " 31 And Menelaus says of Telemachus, the
son of Odysseus, that his blood is noble. 32 The Homeric epics describe an
aristocratic world in which gallantry, passion, and strength properly belong
to the members of the dominant class. So it is socially obligatory to believe
that the warrior virtues as well as physical force and good health are hered-
itary. That is why genealogical considerations are customarily used to ac-
count for and rank the physical traits of the various heroes. In the Iliad)
only one man of low birth dares to speak during an assembly, and it is no
accident that this representative of the lower classes, Thersites (of whom
more later), is described by the poet as an inferior being, ugly and de-
formed, disabled from birth. Homer refrains from giving eugenic advice, 33
but an echo of such precautions is audible in Hesiod:
Bring home a wife to your house when you are of the right age,
while you are not far short of thirty years but not much above;
this is the right age for marriage. Let your wife have been grown
up four years, and marry her in the fifth. Marry a maiden, so that
you can teach her careful ways, and especially marry one who lives
near you, but look well about you and see that your marriage will
not be a joke to your neighbors. For a man wins nothing better than a
good wife, and, again, nothing worse than a bad one, a greedy soul
who roasts her man without fire, strong though he may be, and
brings him to a raw old age. 34
The overt purpose of most of these recommendations is the happiness,
health, and longevity of the patriarch, not the qualitative and quantitative
preservation of his race. But closer observation reveals that, for Hesiod
and moralists of his stripe, the welfare of the progeny is basically at stake.
Taken as a whole, this advice seeks to bring about a state of affairs held to
be especially favorable for biological reproduction and the preservation of
the patrimony.
The farmer's experience teaches that the use of first-rate seed is a favor-
able but not sufficient condition for the production of a good harvest. The
crop's worth also depends on the soil and the weather. Should it not be
the same for human beings? In the land of the Phaeacians, for instance,
the climate is very mild: the crops never fail there, since the west wind's
blowing fosters the sprouts and ripens the full-grown fruit. The poet of
the Odysseyremarks that the character of the Phaeacians reflects their land's:
not remarkable as boxers or wrestlers, they are fast on their feet and excel-
lent sailors, with a love for feasting, the lyre, singing, changing clothes,
warm baths, and bed. 35 Still more important for my purposes is the neat-
herd Eumaeus's description of his homeland, the island Syria:
Not so much a populous island, but a good one, good for
cattle and good for sheep, full of vineyards, and wheat raising.
No hunger ever comes on these people, nor any other
hateful sickness, of such as befall wretched humanity;
but when the generations of men grow old in the city,
Apollo of the silver bow, and Artemis with him,
comes with a visitation of painless arrows, and kills them. 36
Even if Syria truly exists and the soil of this little island of the Cyclades
has simply been depleted over the centuries, this passage is still nothing
other than an evocation of a fatherland forever lost, an imaginary paradise.
But that is not the issue. What interests me is the implicit correlation of
climate and substance with long life and the absence of chronic disease. It
is not said that the land where life is especially pleasant is the cradle of the
best men. The fantastic tale of the Phaeacians expresses the same thought
as the proverb Herodotus puts in the mouth of Cyrus:
"Soft countries," he said, "breed soft men. It is not the property of any one
soil to produce fine fruits and good soldiers, too. " 37
Ideas like these, already outlined in archaic times, reach their acme in the
great treatises of the Hippocratic corpus. 38
As a rule, the Homeric heroes are well-built, of unfailing health, and, at
least in the Iliad) durable to an extent that defies experience. Even old age
does not stop these stout-hearts: Nestor, wise king of Pylos and eloquent
orator, accomplishes deeds of valor even though "he had already seen two
generations of mortal men pass. " 39 Binges of slaughter leave them un-
harmed: the Iliad knows nothing of disabled veterans, despite its horren-
dous battles. But there is nothing surprising in that: the epic disdains the
other side of the coin.
Yet there is an exception, an anti-hero par excellencenamed Thersites.
This is how he is caricatured:
This was the ugliest man of those who came beneath Ilion.
He was bandy-legged and went lame of one foot,
with shoulders stooped and drawn together over his chest,
and above this his"skull went up to a point with the wool grown sparsely
upon 1t.
Beyond all others Achilles hated him, and Odysseus. 40
Thersites the Insolent is turricephalous, almost bald, stooped, bow-legged,

and lame. According to Christos Bartsocas, this Homeric passage and the
iconography it gave rise to are evidence for ancient knowledge of a congen-
ital dysplasia called dysostosis cleidocranialis by modern medicine. 41 That
is possible but unproven. Whatever the case may be, Thersites' deformity
is above all a dramatic t6pos:this personage is the court buffoon, the show-
off, and the whipping-boy of Odysseus, who does not flinch from smiting
him with the scepter and raising, as though it were needed, a "bloody
welt. " 42
Skipping over the hump of Eurybates, the senile kyphosis of Aegyptius,
the ugliness of Dolon (whose appearance is disturbing but whose feet are
swift), and the twisted foot of the god Hephaestus, 43 I focus my attention
instead on the most frequently mentioned infirmity in epic and in classical
tragedy: blindness. Enraged by his boasts, the Muses punished the Thra-
cian bard, Thamyris, with it, 44 and the gods themselves saw fit to blind
the Theban seer Tiresias. 45 In ancient belief, loss of sight was linked by a
kind of compensation magic to clairvoyance and the gift of poetic creation,
song, and enchantment. Thus the bard Demodocus, who sang so well the
events of the Trojan War, was blind as a result of the Muse's special
affection: the Muse "gave him both good and evil. / She reft him of his
eyes, but she gave him the sweet singing / art. " 46 Some have seen in
this bard the poet's self-portrait. 47 Was Homer blind? If we accept the
attribution to him by Thucydides and Aristophanes of the majestic Hymn
to Apol/,o)he must have been blind, ,, since the author of that poem describes
himself as "a blind man (tuphlosaner) living in harsh Chius. " 48 But modern
criticism is not inclined to date the composition of this hymn earlier than
the beginning of the seventh century B.C. or to consider it a work by the
author of the Trojan epic.
Several of the Lives of Homer mention the great poet's blindness, but all
of them are late. According to the biography of Homer transmitted in the
chrestomathy of Proclus, "Some claim that he received his name as a result
of his blindness; according to them, the Aeolic peoples call blind f!len
homeroi.JJ 49 That is an old opinion. There is a trace of it in Ephorus, the
fourth-century Greek historian. According to him, the poet's original name
was Melesigenes. He changed it to Homer because of his blindness, since
that was what blind people were called as a result of their using guides (ton
homereuonton).50 The etymology is patently false, but the need for such a
philological digression in Ephorus proves that learned men of his time
thought of the father of epic poetry as a sage and venerable personage who
had lost the use of his eyes. In Plato there is a strange parallel between
Homer and the great lyric poet Stesichorus, who had been "deprived of
his eyesight for having cursed Helen. " 51 The iconographic tradition, which
goes back at least to the classical period, provides us with a few magnificent
busts of Homer as a blind old man. At ~imes they can barely be distin-
guished from busts of Stesichorus. 52 But it is important to note that in
some ancient representations, for example, a fourth-century B.C. coin from
Ios, Homer is a sighted figure with a piercing gaze.
Ancient writers were not unaware of the contradiction between a tradi-
tion that wished the bard to be blind and the fact that visual experience of
the world is necessary to describe it well in poetry. On the one hand, says
Cicero, the tradition portrays Homer as blind; on the other, the text he
left is so rich in descriptions of places and events that it more nearly
resembles painting than poetry. 53 In short, to cite the Homeric biography
handed on by Proclus, ''Those who have called him blind are themselves
poor observers, since no man has seen as much. " 54 Lucian is similarly
ironic about Homer's supposed blindness and pokes fun at everything
thought known about the poet's life. 55 But such critiques could not gain
headway against a notion so deeply rooted in psychological, as against
historical, compulsions. According to Dio Chrysostom, poets catch blind-
ness from Homer as though it were an infectious ophthalmia. 56 To be sure
there were a hardy few who claimed, out of a taste for paradox or to
underline the marvelous nature of divine inspiration, that Homer was
blind from birth or youth. 57 But generally the perfection of Homeric
descriptions was a major reason for asserting a relatively late date in the
poet's life for the onset of his blindness.
Pausanias says explicitly that Homer lost his eyesight exactly like Tha-
myris before him. Although in both cases the ultimate cause of the blind-
ness was on the divine level, for Pausanias it is still true that it was actually
produced through the agency of an eye disease. 58 According to Heraclides
Ponticus, ''There is evidence that Homer traveled from the Tyrrhenian
Sea to Cephalonia and Ithaca, where it is said that he lost his sight as the
result of an eye ailment. " 59 According to a biography falsely attributed to
Herodotus, Homer contracted ophthalmia in Ithaca and went blind in
Colophon. 60 Such accounts are of some interest as historical evidence con-
cerning eye ailments current at the time of their invention, but they have
no value for the diagnosis of the disease of the historical Homer or for the
nosological realities of the archaic period. 61
A variety of causes can be invoked to account for the cases of acquired
blindness that are spoken of in the epic, including the possible but un-
proven case of Homer himself. Infectious ophthalmias were most likely
the cause of bilateral blindness in those days. Despite the absence of direct
proof, for paleo-epidemiological reasons it must be admitted that tra-
choma, a disease present in Egypt since Pharaonic times, occurs around
the eighth century B.c. along all the other shores of the eastern Mediter-
ranean. 62 The multiplicity and richness of Homeric expressions for the
functions of the eye and for visual perception are remarkable 63 and make it
difficult to imagine that their author hadn't the benefit of personal expe-
rience. So saying, we rejoin the argument in Proclus. It does not exclude
the possibility of acquired blindness. On the contrary, the point can be
turned around and used as a proof of the poet's sensitization to visual
phenomena after sight loss. In my opinion, this kind of reasoning is too
weak to be taken seriously. I cannot consider it a scientifically sound idea
that the special structure of Homeric dreams is a proof of their author's
blindness, 64 or that the imprecise color terms in Greek epic diction prove
that the Hellenes were color-blind or "immature" in their sense
65
perception.
Trojan War TV<Junds

The Iliad and the Odysseypresent us with the circumstances surrounding
the death of about 200 named individuals. In the majority of cases death
results from combat wounds. In accord with the differences in their subject
matter, martial violence has a greater role in the Iliad than the Odyssey.The
tale of Odysseus's wanderings contains a score of woundings, most of
which occur in book 22 and are part of the massacre of the suitors. From a
medical point of view the descriptions are relatively summary and stereo-
typed. By contrast, the author of the Iliad describes with remarkable ana-
tomical precision, and not without some delight, a great number of highly
varied wounds visited upon the warriors fighting before the walls of Troy.
In some ways these descriptions constitute the oldest surgical report of
losses in a military campaign.
What could be more tempting for a modern physician than to take the
epic literally and consider these descriptions surgical evidence from which
to compile statistics? In 1865 Charles Daremberg was the first to scour the
text of the Iliad to establish, with prudent restraint, just the numeric
distribution of wounds over various regions of the human body. 66 A Ger-
man army doctor, Hermann Frolich, revised Daremberg's statistics to
complete them 67 and to include both the result of each wound and the
weapon used. 68 Frolich's tableJs reproduced in Table 1.
Other, more recent writers have contributed to Homeric wound statis-
tics by comparing them with modern evidence and by remarks on some
questions of detail, but they have not succeeded in changing in any essen-
tial aspect the opinions formulated long ago by Daremberg and Frolich. 69
The statistical method brings out correlations hidden in the Homeric nar-
rative, but it is important not to get carried away by numbers and lose
sight of the fact that the Iliad is not an exhaustive or even a representative
report on the surgical state of affairs during a specific war. Statistical analysis
28 DISEASES IN THE ANCIENT GREEI{ WORLD
TABLE I. Statistical Tableof vVtJundsin the Iliad

Weapon
Body part Result Stone Sword Spear Arrow Total
Head Fatal 4 8 17 2 31
Not fatal 0 0 0 0 0
Unclear 0 0 0 0 0
Neck Fatal 1 4 8 0 13
Not fatal 0 0 1 0 1
Unclear 1 0 0 1 2
Trunk Fatal 1 4 59 3 67
Not fatal 1 0 5 3 9
Unclear 0 0 3 0 3
Upper Limbs Fatal 1 1 0 0 2
Not fatal 0 0 6 1 7
Unclear 0 0 1 0 1
Lower Limbs Fatal 1 0 0 0 1
Not fatal 2 0 3 2 7
Unclear 0 0 3 0 3
TOTAL 12 17 106 12 147
SOURCE: H. Frolich, Die Militannedizin Homer)s (Stuttgart, 1879).
is only applicable to the poetic discourse by analogy; consequently, I refuse

to use percentages or other statistical parameters arrived at by computa-
tion. They are based on false assumptions, and the refinement of the
mathematical operations in such figures endows them with only deceptive
legitimacy. To put it briefly, the conclusions to be drawn from statistical
analysis of Homeric wounds must be limited to the detection of tendencies
in the poet's thinking, which then allow us to make guesses about his
familiarity with military medicine in his own time and to discern some
specific aspects of it.
In terms of the number of victims, the most formidable weapon is the
spear (by "spear" I mean all casting weapons with a haft: spear, pike, and
javelin). However, in terms of lethal wounds, the sword is the most dan-
gerous of archaic weapons. In the Iliad) no one wounded by a sword
survives. The deadliness of the result is less certain when the wound is
from a spear or a stone. Finally, the bow leaves one with a fair chance of
surviving: roughly one hero of every two hit by an arrow does not suc-
cumb to his wound. In general, the greater the distance between adversar-
ies when a given weapon is used, the less effective that weapon becomes.
Homeric warriors know where to strike to finish off their enemies
quickly. In the Iliad) the result of a wound depends more on the region
and the organs hit than the weapon used. The Homeric descriptions of
woundings attest to an excellent knowledge of vulnerable points in the
human body, of the anatomical disposition of the principal organs, and of
the most likely consequences of lesions to each. Aeneas, Aphrodite's mor-
tal son, is struck by a huge rock on the hip, at the place called the "cup-
socket,'' where the thigh turns into the hip-bone. The blow tears the skin,
"breaks the two tendons" (these could be the ilio-femoral and the pubo-
femoral ligaments, or the tops of the crural and vastus externus muscles),
and crushes the cotyloid socket. Although the rock was thrown with
superhuman force (its weight was such that "no two men could carry /
such as men are now''), Aeneas does not die from this wound-he doesn't
even lie down on the ground immediately. He falls to his knees and faints
from the pain: "The fighter / dropping to one knee stayed leaning on
the ground with his heavy / hand, and a covering of black night came
over both eyes." In this state of shock, he would have been an easy mark
for enemies wanting to slay him and plunder his weapons, but his mother,
Apollo, and Artemis protect him. Transported miraculously to a distant
sanctuary, Aeneas is cured so quickly and completely that he returns to
combat the same day, full of vim and vigor. 70 This narrative is a skillful
interweaving of realities experienced in daily life with the marvels of an
imagined world. While the presentation of the wound and its conse-
quences is realistic, the rapid cure and above all the integral restoration of
the damaged body are contrary to observed experience in similar cases-
but those events are expressly attributed to divine intervention.
With all his might, Patroclus hurls a jagged rock at Rector's charioteer,
Cebriones, and it hits him right on the forehead: the rock crushes both
brows and, unhindered by the bones, dislodges his eyeballs from their
sockets. Wounded mortally, the Trojan warrior "vaulted to earth like a
diver. " 71 His fall resembles the fall of the Paphlagonian charioteer men-
tioned above: struck on the temple by the sword of Antilochus, he falls
"gasping ... from the carefully wrought chariot / headlong, driven
deep in the dust.' ' 72 There are thirteen other instances when mortally
wounded warriors fall forward, and fourteen when they fall backward.
Each time th_ereis a good reason, either physiological or physical, to justify
what the poet says. Men die in the Iliad falling backward or forward, stiff
or sagging, gasping or crying out, but always in a way compatible with
what a modern physician would predict given the localization of the wound
in question. 73
Head wounds are rightly considered especially dangerous. Once a lesion
on that part of the body is sufficiently important, death is inevitable.
Wounds to the face from a spear or sword are particularly shocking. An
example will reveal the atrocity and the realism of their description:
Idomeneus stabbed Erymas in the mouth with the pitiless
bronze, so that the brazen spearhead smashed its way clean through
below the brain in an upward stroke, and the white bones splintered,
and the teeth were shaken out with the stroke and both eyes filled up
with blood, and gaping he blew a spray of blood through the nostrils
and through his mouth, and death in a dark mist dosed about him. 74
Actually, a warrior's face was only partially covered by his helmet, which
served to protect him fairly well from cut~ing and thrusting weapons. To
be sure, the helmet was vulnerable to an especially violent spearcast, as in
the sad case ofHippothoos, the Pelasgian chieftain. The Telamonian Ajax
leaped at him and
struck him at dose quarters through the brazen cheeks of his helmet
and the helm crested with horse-hair was riven about the spearhead
to the impact of the huge spear and the ,veight of the hand behind it
and the brain ran from the wound along the spear by the eye-hole,
bleeding. There his strength was washed away. 75
Nor was it enough to be helmeted to be safe from the dread consequences
of a blow to the head from a hard and heavy object like a club or a rock.
Here, for instance, the Trojan warrior, Erylas, is struck by a rock on the
dome of his skull: "All the head broke into two pieces / inside the
heavy helmet, and he in the dust face downward / dropped.'' On the
Achaean side, Epeigeus died in exactly the same way. 76 But during his
battle with Diomedes, Rector's helmet saved his life. Diomedes' javelin
struck the top of it, but bronze repelled bronze. Shaken up, Hector
withdrew at a run, then fell to his knees in a faint. Finally, he "got his
wind again," stood up, and remounted his chariot. 77 The epic repeats here
verbatim the lines used to describe Aeneas's loss of consciousness when
wounded on the hip. 78 From a medical point of view, however, the situa-
tion is a little different: Hector has no visible wound. 79 He has suffered a
concussion, a disturbance of the brain that causes a transient loss of
consciousness. 80
Nowadays, concussions are almost inevitable for boxers. In view of the
boxing rules in antiquity, it was even worse then than now. Here is the
Homeric account of a sporting event during the funeral games of Patro-
clus, which pitted the Argive leader, Euryalos, against the warrior Epeios,
skilled in boxing:
Their heavy arms were crossing each other,
and there was a fierce grinding of teeth, the sweat began to run
everywhere from their bodies. Great Epeios came in, and hit him
as he peered out from his guard, on the cheek, and he could no longer
keep his feet, but where he stood the glorious limbs gave.
As in the water roughened by the north wind a fish jumps
in the weeds of the beach-break, then the dark water doses above him,
so Euryalos left the ground from the blow, but great-hearted Epeios
took him in his arms and set him upright, and his true companions
stood about him, and led him out of the circle, feet dragging
as he spat up the thick blood and rolled his head over on one side.
He was dizzy when they brought him back and set him among them. 81
Wounds to the,, neck are relatively frequent, considering the small size
of this region of the body. This is because it is at once very delicate and
poorly protected by the warrior's armor. 82 Plotting Rector's removal,
Achilles was eyeing his "splendid body, to see where it might best /
give way.'' All the rest of his body was covered with arm or,
yet showed where the collar-bones hold the neck from the shoulders,
the throat, where death of the soul comes most swiftly; in this place
brilliant Achilles drove the spear as he came on in fury,
and clean through the soft part of the neck the spearpoint was driven.
Yet the ash spear heavy with bronze did not sever the windpipe,
so that Hector could still make exchange of words spoken. 83
In several other instances, a warrior plunges his spear into his enemy's
throat, cutting the carotid arteries 84 or the whole neck from one side to
the other. Sometimes decapitation is the coup de grace administered to a
powerless, fallen enemy. Deucalion, his arm already transfixed by Achilles'
spear, awaits death resignedly: "Achilles struck with the sword's edge /
at his neck, and swept the helmed head far away, and the marrow / gushed
from the neckbone. " 85
The number of wounds to the trunk or abdomen is particularly high in
the epic. They account for more than half of all wounds whose localization
is specified. Doubtless this high frequency is consistent with what actually
happened in the Bronze and Iron Ages, when the decisive role in battle
was played by soldiers in heavy armor. It is explicable in view of the central
position and relative size of the trunk as a target for casting weapons and
arrows. The chest itself was heavily protected, as much by body arm or as
by the shield, which was carried in the left hand and covered the cardiac
region. 86 As a result, Homeric champions sought to hit their enemies at
waist level and on the right-hand side, "under the phrenes) in the liver. " 87
According to Homer, a wound to this organ inevitably resulted in death.
In fact, given the prevalent conditions, there could only have been very
rare exceptions to this rule. The experience of modern army surgeons
attests to the extreme seriousness of sword wounds that cut across the
diaphragm and thus open both the thoracic and abdominal cavities. 88
The belly was also a region where wounds were unforgiving. 89 Having
wounded Diomedes in the foot with an arrow, Paris shouts in delight that
he did not miss his enemy but. is unhappy that he failed to hit a fatal spot:
"You are hit, and my arrow flew not in vain. How I wish / I had struck
you in the depth of the belly and torn the life from you. " 9° For instance,
Agamemnon kills Deikoon by driving his spear through the man's shield
and belt into his belly. 91 Or Meriones took Adamas's life by a similar wound:
Meriones dogging him threw with the spear
and struck between navel and genitals where beyond all places
death in battle comes painfully to pitiful mortals.
There the spear stuck fast driven and he, writhing about it,
gasped as an ox does when among the mountains the herdsmen

have bound him strongly in twisted ropes and drag him unwilling.
So he, stricken, gasped for a little while, but not long,
until fighting Meriones came close and wrenched the spear out
from his body, and a mist of darkness closed over both eyes. 92
Like decapitation, the death-blow to the belly was a coupdegrace dealt to
a defeated but not yet dead enemy. The account of the battle between
Thoas and Peiros shows that the epic poet knew that abdominal wounds
were more serious than wounds to the lungs:
Thoas the Aitolian hit Peiros as he ran backward
with the spear in the chest above the nipple, and the bronze point fixed
in the lung, and Thoas standing close dragged out the heavy
spear from his chest, and drawing his sharp sword struck him
in the middle of belly, and so took the life from him . . . 93
After Pa trod us, a victim of divine terror, had already been struck by sur-
prise in the back between the shoulder blades, Hector finished him off
with a blow to the same region. 94 Wounds to the lower belly at once
produce a state of shock. They are not always immediately fatal, but it was
not known how to combat internal hemorrhaging or peritoneal infection
caused by the spilling of intestinal contents or the consequences of the
intra-abdominal effusion of urine. Lesions of the bladder were doubtless
frequent and deadly. The case of Pherekles serves as an illustration of a
lethal wound to that organ by a strange but anatomically possible route:
This man Meriones pursued and overtaking him
struck in the right buttock, and the spearhead drove straight
on and passing under the bone went into the bladder.
he dropped, screaming, to his knees, and death was a mist about him. 95
Going from back to front and from below to above, the spear traversed
the gluteus maximus, the ischio-pubic foramen, the bladder, and the pubic
arch. 96
I now come to wounds of limbs. 97 There are relatively few of them,
and, as should be the case, most often they are not mortal. Their relative
infrequency in the epic is certainly not in conformity to reality, to the
abundance of banal wounds to arms and legs. The p·oet prefers to relate
serious and spectacular events. One case merits attention. When Teucer is
drawing the string of his bow, a sharp rock thrown by Hector hits him ''at
the side of the shoulder, where the collar-bone separates the chest from
the neck, at the decisive spot." The shock broke the string and "his arm
went dead at the wrist. " 98 The scene suggests a basis in observed reality
(lesion of the brachia! plexus?), since details like these do not arise from
imagination alone. The same is true for remarks about the convulsive
movement of the jaws caused by a fatal wound to the nape of the neck, 99
or the "shooting pains" from a relatively light leg wound that arise only
after "the sore place was dry, and the flow of blood stopped," 100 or, again,
the attack of "harg pain [coming] over his flesh" not at the moment of an
arrow wound to the foot but when the arrow is extracted. 101
A few overly enthusiastic medical historians could not keep themselves
from calling Homer the oldest Greek doctor known. Before becoming a
poet, he was an army surgeon, or at least a kind of priest with access to
medical lore accumulated over centuries. 102 According to Otto Korner, the
Diomedeia (Iliad s) is different from the rest of the poem in its surprising
knowledge of "anatomical topography. " 103 It is tempting to conclude that
its author had medical training, that he was a real precursor of Hippo-
crates! Sadly, such hypotheses, however attractive they may be to a medical
audience, are unprovable. 104 The Homeric epics contain no more medical
knowledge than a "gentleman" of the eighth century B.c. with some
experience of warfare can be expected to have had.
Homer has brief but undeniably realistic descriptions of cuts or surface
wounds. 105 Their treatment is fairly crude (rational in the Iliad and magical
in the Odyssey),and the poet only speaks of it for fresh wounds. 106 The
Iliad calls the loss of tegumentary continuity helkos)107 a term taken up by
the Hippocratic physicians in a technical sense that for a long time com-
prises both cuts and ulcers. 108 The word trauma never occurs in Homer,
despite numerous occasions in which it could have been used. 109
It is a surprising and completely abnormal thing that the Homeric he-
roes are never sick as a result of their wounds. They either die of them or
return very quickly to normal activity. Hector fights like a lion after suffer-
ing a concussion and a thoracic contusion that makes him spit blood.
Teucer, whose shoulder injury and arm paralysis were discussed above, is
in such a state after it that he sobs and has to be carried back to the ships
by his companions. One day later, he is already fighting bravely against
several adversaries. Agamemnon, Odysseus, and Diomedes all participate
successfully in the funeral games for Patroclus, but they were all wounded
in battle the day before. The rule in Homer is that the heroes either die or
reemerge in good health; that wounded persons can be in a state of trau-
matic shock, but never have any fever; that cuts bleed but are never
inflamed; that tissues can be torn and smashed but never suppurate. To
be sure, these distortions of reality reinforce the poetic design and flow
from the very nature of epic narrative, which will not stand for inactive
personages on the scene. The heroic life is inconceivable in the absence of
physiological completeness. D·eath on the field of battle in the prime of
life is glorious; slow decay is its shameful opposite. Even in the back-
ground, chronic diseases would be incongruous in this imagined world.
True Diseasesin Homer

Other than the pathological states discussed above, which are disabili-
ties, 110 and the sudden deaths attributed to the invisible arrows of Apollo
and Artemis or the thunder of Zeus, and aside from wounds whose etiol-
ogy is obvious, the Homeric epics do speak of mental aberrations and cases
of intoxication, they mention a few banal pathogenic factors, and last, in
an unfortunately off-hand way, they talk about internal diseases.
Several verses of the Homeric poems tell us that the gods can disturb a
man's spirit, cloud his reason, and strike him with madness. Nevertheless,
having decided not to include mental illnesses in my inquiry, I pass over
the depressiveness of Bellerophon and the manic behavior of a few others.
In any case the Homeric texts add little to the medico-historical knowledge
of psychic health. Not until Greek tragedy do the poets provide us with
masterful, almost clinical descriptions of neuroses and psychoses. 111
Rage (lussa) poses a separate problem. Although in Homer this term
denotes a kind of fury and not the disease (rabies) that will later bear the
same name, it is likely that there are very old links between the fury of
fighting men and rabid dogs. 112 Two other words that are part of modern
medical terminology are already attested in the Iliad: aphasia (amphasieor
aphasia) and asthma (asthma). For a physician nowadays these names are
attached to specific diseases. Such is not their meaning in archaic literary
texts or even in the classical and Hellenistic Greek medical writers. Aphasia
in the ancient sense of the word is nothing more than difficulty in speaking
when one is the victim of violent emotion. Antilochus has just learned
from Menelaus of the death of Patroclus and is stupefied by the news.
Stricken with horror, for a long time he is seized with "aphasia," that is,
he can't say a word. 113 The modern nosological sense of the word "apha-
sia" (speech disability caused by lesions in a specific region of the brain)
dates only from 1864. Likewise, if Hector is struck by a rock and suffers
from asthma and perspiration, this simply means that he has trouble
breathing and is sweating. 114 His complaint is dyspnea as the result of a
contusion of the chest and not an attack of suffocation strictly speaking.
The word asthmazo is used now and then in ancient literary texts to indi-
cate that a person is panting, either after running or some other form of
exertion or during death-agony. In the Hippocratic corpus asthma denotes
accelerated and difficult breathing, or shortness of breath. Despite techni-
cal usage in nosological contexts, it remains a term for a clinical symptom.
In specific cases it can be used for patients with bronchial or cardiac asthma,
but that is by no means the rule. 115
Hom.er knows of the poisonous effect of snakebite, the use of poisoned
arrows, and the existence of venomous plants. The Homeric word phar-
makon is a middle term between our concepts of poison and of medicine.
As for banal pathogenic factors, I recall that Hector recommends disori-
enting the Achaeans with smoke 116 and that Odysseus speaks of the dele-
terious effects of cold, specifically of morning frost, on persons who are
scantily clad. 117 People have always known that great exertions can break a
man. This is how the poet describes Odysseus's state when he lands on
the Phaeacians' island, after days on a raft at sea and an exhausting swim
to reach dry land:
Now he let limp both knees
and his ponderous hands; his very heart was sick with salt water,
and all his flesh was swollen, and the sea water crusted stiffly
in his mouth and nostrils, and with a terrible weariness fallen
upon him he lay unable to breathe or speak in his weakness. 118
Was such prostration, brought on by plainly natural factors and easily
relieved by rest alone, really considered a nousos)a disease strictly speaking,
by the Greeks of the archaic period? It is doubtful that it was. In the
Homeric world, true disease has a divine etiology. The Cyclopes, for in-
stance, who are Odysseus's dupes, so answer Polyphemus's call for help:
"If alone as you are none uses violence on you / why, there is no
avoiding the sickness sent by great / Zeus. " 119 Long ago, Celsus drew
attention to the fact that during the Trojan War the doctors, "in the
Homeric account, were of no help against pestilence (pestilentia)or against
various diseases (in variisgeneribusmorborum) since they are only presented
as treating wounds with remedies and by the knife.'' 120 And the poet does
call the destructive plague that is started by the arrows of Apollo an "evil
disease. " 121 Along with a notion that sudden death and chronic disease are
of divine origin, since earliest antiquity and especially in regard to pesti-
lence there was also a specific belief that invoked impurity or the violation
of tabu as a causal explanation. 122
Returning to the exhaustion of the shipwrecked Odysseus on the shore
of Scheria, some modern medical historians have wished to consider that
passage the oldest description of seasickness. 123 I do not find such an
interpretation satisfactory. Instead another passage in the Odysseyreminds
me of that syndrome, the one in which Odysseus tells how he and his men
sailed off the coast of Crete, slipping along smoothly "like sailing down-
stream" and "without sickness" (anousoi).124 Greek sailors must have
known seasickness, since it is practically certain that during at least the last
few ten thousand years there have been no essential changes in the anatom-
ical structures and functions of the inner ear that are responsible for this
sickness provoked by the movements of a ship in water. Nevertheless,
other, more serious diseases must have haunted the sailors of old, and the
poet could well have mentioned them and not just a minor, transitory,
and, frankly, somewhat silly ailment. Indeed, the word for seasickness is
attested in Aristophanes . 125 Although it does not occur in Homer, the
nosological term nautfa (or, in Ionic dialect, nausie) is assuredly very old.
Its primary meaning, seasickness, or more precisely ship-sickness, allows
for no doubt despite the fact that in its oldest occurrence, in some satirical
lines of the lyric poet Semonides (late seventh century B.c.), it denotes
nausea in general. 126
Since ship-travel in Homeric times was only coastal, the sailors had no
DISEASES IN THE ANCIENT GREEI< WORLD
fear of scurvy, but the frequency and proximity of ports-of-call did not
protect them from infectious diseases, and in fact they must have been
especially exposed to diseases endemic to coastal zones, principally infes-
tation by pathogenic protozoa. Water and insects, both germ carriers,
must have been the chief factors in archai~ naval pathology. Although the
important· role they played was unknown, their consequences could not
be ignored. It is not hard to understand why Odysseus considers it remark-
able that at one stage of his voyage no one on board was sick. There is
every reason to believe that sailors in those times suffered often from
typhoid, malaria, and dysentery. Those who reached Africa must have had
schistosomiasis. Unfortunately, our literary sources are silent on the
subject.
Likewise, in a military camp, and especially during a prolonged siege,
infectious diseases and those caused by various deficiencies could not have
been absent. And yet the Homeric heroes never have intestinal or renal
colic-they don't even catch cold. The historian of diseases can only regret
that moral and aesthetic considerations heeded by the epic deprive him of
information about the one category of pathology that is fundamental to
his investigation.
The scene of a sick man in bed, according to Daremberg, is completely
alien to the epic; no hero takes time out for a bout of pneumonia or a case
of diarrhea. 127 As far as I know, only one passage in archaic Greek literature
speaks of a sick man in bed. Homer uses the metaphor of the curing of
sickness to express the joy Odysseus feels once he senses the shore after
thrashing about in heavy swells:
And as welcome as the show of life again in a father
is to his children, when he has lain sick, suffering strong pains,
and wasting long away, and the hateful death spirit has brushed him,
but then, and it is welcome, the gods set him free of his sickness. 128
The poet gives the epithet stuger6s'hateful, abominable, odious' to the

malevolent spirit who causes suffering (al!Jea)and who slowly dissolves
(tik6menos) one's flesh. In other Homeric texts, this epithet applies to
disease itself, as in
many times the good old man Polyidos had told him
that he must die in his own house of a painful sickness (nousoihup) a1lfaliei)
or go with the ships of the Achaians and be killed by the Trojans.
He therefore chose to avoid the troublesome price the Achaians
would ask, and the hateful sickness (nous6nte stugeren)
so his thumos might not be afllicted. 129
Homer never speaks of diseases but only of one disease, or rather, the
disease. This disease par excellenceis a kind of chronic wasting away or
consumption. The historian of diseases would like to be able to say it is
phthisis (pulmonary tuberculosis) or malarial cachexia, but the poet's lan-
guage is too impn;,cise. Without any doubt, these two diseases were the
major components of the nosological reality behind the Homeric texts. I
can be positive in this regard thanks to cross-checking from nonliterary
sources, since the testimony in Greek texts prior to the Hippocratic corpus
is too generic and ambiguous to support, by itself, the diagnosis of tuber-
culosis and malaria. In a lovely passage cited above, Odysseus's mother
draws the distinction between gentle death from divine arrows or grief and
death from the disease ''that drives out the thum6s by the hideous con-
sumption of flesh. " 130 To begin with, it is "'North noticing that the epithet
(stuger6s)given elsewhere to the malevolent spirit and the disease itself is
here applied to consumption. The word I translate "consumption" is
tekedlJnand not phthfsis) a term that does not appear in Greek literature
before Herodotus. But there is no doubt ,,about the meaning of this sub-
.
stantive, which is a derivative of the verb teko 'melt, dissolve' (applied, for
example, to melting snow). The authors in the Hippocratic corpus use
this verb often in connection with pulmonary consumption. 131
Some historians believe that the latter disease is mentioned by two tragic
poets of the classical period, but the texts in question are very vague. In
the play of Sophocles that bears her name, Antigone is told by the chorus
that she "departs toward the hiding place of the dead without being
wasted by consuming diseases. " 132 And in Euripides' play about Alcestis,
the heroine is dying to save the life of her husband and the servant woman
says that "the disease consumes her. " 133 From a strictly medical viewpoint,
these passages do not support any specific diagnosis.
The situation is no better for the attestation of malaria in literary texts.
Speaking of the star called Orion's Dog, Priam says that "it is a
sign of evil / and brings on the great fever for unfortunate mortals.'' 134
By 1700 Adam Brendel was interpreting this passage as an allusion to the
upsurge in malarial fevers during summer and fall. For the Greeks the rising
of Sirius with the sun heralded hot weather, the dog days. This was the
time that malaria raged. 135 But Charles Daremberg opposed Brendel's hy-
pothesis and ..refused to accept that puret6s in this passage means fever in a
medical sense. According to him, it is "quite simply a period of intense
heat that just exhausts poor mortals. " 136 This may seem a surprising idea,
but it only takes up where an ancient quarrel leaves off. A scholium teaches
us that the learned men of antiquity were unsure of the meaning of this
Homeric line: "Note that the word puret6s occurs only here and that it is
used in its ordinary sense and not, as some assert, to denote hot
weather. " 137 To me it seems that Daremberg's interpretation renders the
Homeric text banal and pointless. I prefer to leave puret6s in its medical
sense, which is well-attested all through Greek literature. However, Bren-
del and in his train Korner and a good number of other historians seem to
me to have gone too far in reducing at all cost these Homeric fevers to the
intermittent and pseudo-continuous fevers of modern pathology. The ep-
DISEASES IN THE ANCIENT GREEK WORLD
idemiology of infections due to salmonellae is cautionary in this regard:

the fact that acute endemic fevers rise in summer and autumn does not
prove they are malarial. When all is said and done, the real interest of the
remark in Homer is not that it proves the existence of malaria in Homeric
times, but as Fernand Robert did well to emphasize, that the relation
between seasons and epidemic diseases had been observed, prefiguring,
albeit from afar, Hippocratic medicine. 138
From Hesiod to the R-esocratics

After reading and rereading Homer, I pass on to Hesiod, the shepherd
from Boeotia. According to Herodotus, these two poets gave the Greeks
the genealogies of the gods and their names and functions. Hesiod's The-
ogonyretells the birth of the world, the cosmogonic struggle of the new
forces, and the origin of civilization. It is a mythological account transpos-
ing reality into symbolic tales. Such an approach is hardly conducive to
the recording and transmission of concrete pathological information. Even
when such information is in the myth, we cannot be sure of decoding
correctly the language and images in which it is embedded.
Mythology likewise predominates in the second of the two principal
poems of Hesiod, although, as its title suggests, Worksand Days is a didactic
poem closer to earth than sky, closer to men than gods. Conspicuous in it
are the myth of the successive generations of men and the myth of Pan-
dora. I have already spoken of them in the Introduction, since these two
related myths provide a solution, a still prerational solution, to the prob-
lem of the origin of diseases. Since ancient times and still today, there have
always been learned men who think that these tales of Hesiod relate, in
their own way, a historical truth, namely the progressive decline in health
during protohistorical times that resulted in the sudden appearance of new
diseases. Even if this is true, the myth is no proof. Doubtless Hesiod did
not invent the notion of a Golden Age, a legendary time when the gods
spared men the misery of disease. He adopted it from a very old popular
tradition. But it makes little difference, since the theme of paradise lost
did not have its origin in a confused collective memory of medically idyllic
times. It is a fantasy whose creation and projection onto the distant past
of humanity have to do with depth psychology and not historical recall.
The myth of a Golden Age has various functions, among them the satis-
faction of a psychological need stemming from the irretrievable loss of
childhood and the desire to justify certain social changes. A historian might
conclude that, in reviving this old theme and elaborating it by making
himself the prophet of the Iron Age that will take over from the Age of
Heroes, Hesiod bears witness to a social crisis in his time and to the
necessity that peasants like him leave behind the aristocratic ideology of
Homer. All this is surely true, but it does not advance the study of disease
in ancient Greece.
Even so, once he has transferred his attention from the lofty deeds of
war to the rigors of daily labor, Hesiod teaches us a few details about the
demands and the risks of rural life that are determinants of health. For
instance, he exhorts the Boeotian farmer to prepare himself a shelter and
to provide himself with nourishment for the cold season,
lest bitter winter catch you helpless and poor and you chafe a
swollen foot with a shrunk hand. The idle man who waits on empty
hope, lacking a livelihood, lays to heart mischief-making. 139
This is the first mention in history of starvation edema. 140 Today we know
that lack of protein provokes diffuse, cold edemas localized in the lower
body parts (as a result of gravity). Hesiod provides the perfect clinical
picture of this deficiency as he describes the characteristic attempt of the
sufferer to reduce the swelling by applying pressure around the ankles and
as he emphasizes in the fewest words the contrast between swollen foot
and withered hand. Crises of subsistence are also responsible for a famous
line of his in which famine (limos) and pestilence (wim6s) sit side by side.
Hesiod also knows that winter's cold aggravates rheumatism ("breaks the
back"). 141 But despite Plutarch's view that "it appears that Hesiod was a
physician," 142 research on his vocabulary is disappointing: he hasn't a
single term of pathology that is not already known from the Homeric
texts. 143
Nevertheless, two points are peculiar and worthy of comment. First,
there is the fact that a work devoted to country life makes no mention
whatever of fever. Jones finds the silence especially deafening since Boeotia
was marshy country with a climate favorable to mosquitoes. 144 It is known
that later on malaria was rampant there. Was it not yet endemic around
the end of the seventh century B.C.? The absence of evidence is only a
weak indicator, but it could become significant in the presence of other
arguments. _
Second, Hesiod speaks of diseases)that is, he uses the word nousosin the
singular and the plural, unlike Homer, who uses it only in the singular:
"Countless plagues wander among men; for earth is full of evils and the
sea is full. Of themselves diseases come upon men continually by day and
by night, bringing mischief to mortals silently; for wise Zeus took away
speech from them. " 145 So for this author diseases are numerous and can
attack men of themselves, as they please (aut6matoi)-by a kind of intrinsic
causality and not by an individual divine decision in each case. For Homer,
disease is outside nature: altogether alien to man and dependent on divine
whim, it escapes the order of nature. By contrast, Hippocratic doctors
consider diseases disturbances in man's natural equilibrium. Though un-
harmonious, they are still part of nature and consequently obey certain
rules. Hesiod is a witness to the reclaiming of disease by nature, of nosol-
ogy by physiology . 146 Although he stopped halfway, the philosophers of
the sixth century B.C. did resume the destruction of the metaphysical and
ontological concept of disease and its replacement by a natural and func-
tional concept. ,
Around 500 B.C. Alcmaeon of Croton stated clearly what a natural
disease is. His formulation borrows a political metaphor. It is important
not only as an expression of Pythagorean ideas about evil as a loss of
equilibrium, but also, especially as regards my subject, because it reflects
the concrete experience of contemporary physicians concerning the main
centers of pathological disturbance and their apparent causes:
Health is maintained by the equal rights (isonom[a) of the qualities of the wet, the
dry, the cold, the hot, the bitter, the sweet, etc., but exclusive power (monarkhfa)
among them produces disease. Exclusive power of one [of the opposites] corrupts.
Cases of disease can be explained as to their causes by an excess of heat or cold and
as to their occasion by an excess or deficiency of nourishment; as to their locale,
blood, marrow, and brain are affected. However, diseases also arise from external
occasions, such as certain waters, a place, expense of effort, torture, or the like.
Health is the mixture of the qualities in proper proportion. 147
According to this statement, diseases are not the result of direct actions by
the gods, nor are they whimsical demons whose behavior cannot be fore-
seen. Even so, one disease in particular kept the epithet hier6s'holy' long
after Alcmaeon. 148 An epileptic fit strikes the whole of one's being sud-
denly, without warning, and in spectacular fashion it twists and shakes the
body and deranges the mind, to such an extent that it suggested an extra-
human presence and could not at all be thought of as comparable to other
diseases. Thus epilepsy was considered a sacred disease, or rather, the sacred
disease: he hierenousos.149 It is so designated from the first line of a famous
treatise in the Hippocratic corpus, which, despite its traditional title, the
SacredDisease)is actually concerned to rebut the reason for this name and
thus to storm the last citadel of the magico-religious concept of disease:
As for the so-called sacred disease, this is the way it is: I hold it to be no more
divine or sacred than the other diseases; it has a natural origin from which it derives
just like other diseases. Men have regarded it as divine because of their confusion
and their wonder at its resemblance to no other disease . . . Those who first
attributed a sacred character to this disease were men whose kind one can still find
today: magicians, exorcists, charlatans, quacks. 150
The clinical descriptions found in this treatise and in other Hippocratic

texts fully justify the usual retrospective diagnosis of the sacred disease:
generalized epilepsy, more precisely grand mal epilepsy, is its central ele-
ment. Still, the features in the clinical picture are vague enough to permit
the inclusion of several other pathological states, for instance, certain other
forms of epilepsy, sonvulsions caused by encephalitis, severe spasmophilia,

eclampsia, and, to be sure, hysteria. Despite this possibility of a contami-
nation by other clinical entities, the conceptualization of the sacred disease
would not have occurred had there not been, as its basis, a profound,
long-term, and direct acquaintance with grand mal epilepsy.
So there is not a shadow of a doubt concerning the antiquity of epi-
lepsy .151 The Hippocratic physician's observation of the role of heredity in
its genesis 152 shows that the disease in question must mainly be what is
today called "essential" epilepsy, not just its symptomatic forms. It is true
that the treatise cited above, which gives the first secure description of
epilepsy, only dates from the end of the fifth century B.c., but the polemic
it opens sets back the origin of the disease's name to such a distant past
that the identity of its inventors was unknown. The true origin of the
name "sacred disease" was already long forgotten. 153 ·
Cambyses II, king of Persia from 530 to 522 B.c., was regarded as an
epileptic. In a fit of anger, he kicked his pregnant wife until she miscarried
and died as a result. To explain this misdeed and a host of other abnormal
actions of Cambyses, Herodotus says that ''there is, in fact, a story that he
had suffered from birth from a serious complaint which some call 'the
sacred sickness.' There would then be nothing strange in the fact that
serious physical malady should have affected his brain. " 154 Greek mythol-
ogy accuses Heracles of an attack of madness during which he killed his
sons. And one of the ancient names for epilepsy was "Heracles' disease."
It is not uninteresting that modern medicine fully confirms the occurrence
of episodes of homicidal delirium in some epileptics. Diogenes Laertius
ascribes to Heraclitus of Ephesus the opinion that falsely imagining some-
thing is the same as suffering from the sacred disease. 155 The authenticity
of the passage is not assured. 156
On the other hand, tradition links to the life of Heraclitus, in a reliable
way despite considerable disagreement about the details, the mention of
another disease, dropsy, of which he himself was a victim. Here is the
principal variant on the philosopher's death:
He became so misanthropic that he withdrew and went off to live on herbs and
plants in the mountains. But when this diet made him dropsical, he returned to
the city and consulted the doctors, asking them about his condition in the form
of a riddle: could they change wet weather into a drought? Since they did not
understand, he shut himself up in··a stable, hoping to cure himself and dry up the
water by the heat of manure, with which he covered himself. To no avail-he
ended up dead of it at age sixty. 157
It would be pointless to hold forth on the possible causes of Heraclitus's

death. In terms of modern medicine, dropsy is not a disease but a syn-
drome. A strict vegetarian diet can result in hypoproteinemic edema, but
the term hudor (or hudrops)most likely indicates significant effusions in the
body cavities and swelling of the abdomen. One should consider, in the
case of a 60-year-old leading the life of an anchorite, a very serious ailment

such as peritoneal tuberculosis, heart failure, or cirrhosis of the liver.
Whatever the correct diagnosis, the medical historian will lend his credence
to the above account rather than that of a certain Ariston, according to
whom Heraclitus ''was cured of dropsy ~d died of another disease.'' 158
Lyric Poets of the Seventh and Sixth Centuries

From the standpoint of chronology, it is now incumbent on me to
scrutinize the works of the first Greek lyric poets. Sadly, they have survived
only in bits and pieces whose existence we owe to citations in later authors
and to papyri from Hellenistic Egypt. Despite the fragmentary and dis-
jointed nature of this historical documentation, we owe it great respect
since the message it transmits is true-to-life. Unlike Homer and Hesiod,
the lyric poets of the archaic period are not fond of telling tales of far-off
gods and heroes, nor of teaching proper social behavior; what they want
and what they do is to sing, to purge themselves of their own miseries,
their most intimate problems. In this genre of poetry, the main theme is
immediate, personal experience of the world, so it would be surprising not
to find in it reference to disease and death.
The oldest lyric poet whose work is at least partially accessible to us is
Archilochus of Paros (first half of the seventh century B.c.). He is a strong
personality, an anti-hero. 159 Enamored of life and passionately attached to
the sensual exaltation it can offer, he is a soldier who demythologizes
heroic death. Dead bodies, he says, serve only as fertilizer. He confesses,
without pretending to be ashamed of it, that he once left behind his shield
on the field of battle. And he waves off the moral significance of the act
by exclaiming: I saved my life! He admits it freely, just as he declares
publicly through his poetry that he is a bastard, born of an aristocratic
father and a slave mother. 160 A realist, Archilochus bursts the seamless
unity of virtue and physical beauty that was the rule in Greek poetry until
his arrival. According to him, it's better to have as your officer "a squat
fellow with knock knees" and a good heart than a slender dandy infatuated
with his own curls. But the poet much esteems his mistresses' beauty, and,
the ultimate inconsistency, gets himself killed on duty.
A line of Archilochus contains the following expression: "tumor be-
tween the thighs. " 161 The word used is phurna'tumor, swelling, abscess.'
Daremberg noted that the only interest of this passage lies in its use of a
word from the vocabulary of pathology. 162 But more can be said if we
accept the reconstruction of Archilochus's Epodesproposed by Fran~ois
Lasserre, who joins several fragments by analogy with Horace's Epodes.163
The verse that mentions the tumor between the thighs was probably part
of a satirical description by the poet of his former fiancee, featuring ridi-
cule, not to say caricature, of an aging woman who still thought herself
beautiful. This verse seen1s to follow a fragment that evokes her withering
skin and other ravages of old age. In this context, the phuma in question
could be a prolapsed uterus or vaginal wall, if not a hernia or a vulva
edema.
A scholiast in Theocritus noted that Archilochus also used the word
phut6n in the sense of phuma. 164 By chance the line was found on a papyrus
from Oxyrynchus . 165 The context is mutilated, but something like the
following is still readable: " ... physician ... cutting ... since I know
an excellent remedy, a very different one, for this kind of tumor ... I
.
propose ... evils . . . b every care ful ... o fl 1nen . . . "166 I t 1s ·61e
. 1mposs1
.
to be sure of the meaning of this fragment. Bonnard thinks that "the
mention of a physician, the words for swelling and cure could mean that
poverty had cost the poet his health." In other words, the poet is describ-
ing starvation edema. His poem would then have inspired an epode in the
Catalepton Ve1l]ilianum that speaks of a person whose feet are swollen from
hunger. 167 Even though Archilochus complains impressively in other
poems of the paucity of food on Paros, 168 I doubt that Bonnard's expla-
nation is the right one. The term phut6n has botanical connotations and
suggests an excrescence rather than a diffuse edema. It would seem to
indicate a newly formed tum or, unless it is being used simply in an erotic
sense without any medical connotation.
According to Daremberg, Archilochus spoke of "ablation of the geni-
tals" and "perhaps knew of pediculosis. " 169 That is unprovable, since the
texts in question are short fragments without context. In the first case, the
phrase "he cut the tendons of his limb / member" 170 could refer not to
castration but, as Lasserre suggests, to a fable in which a lion eats a con-
ceited deer. In the second, there is nothing to indicate that the expression
"devoured by vermin" 171 is an allusion to a disease. Nor is it likely that
the curse invoking Sirius and the heat of the dog days has any relation
whatever to malaria. 172
Finally, and not surprisingly, Archilochus sings of love, its passion as
well as its physical aspect. To describe the passion of love, he uses language
that recalls the clinical description of a disease: love's desire excites the
heart, sheds a dense mist over the eyes, and robs one of hearing; it "breaks
the limbs,'' takes one's breath away, and causes terrible pains that pierce
the bones. 173 Alcaeus also sings of the disease of love, and Sappho does so
superbly: the heart melts in one's breast, one cannot speak a word, the
tongue breaks, a subtle fire runs under one's skin, a mist comes over one's
eyes, the ears buzz, sweat streams down one's body, a shudder seizes the
man seated across from the loved one, one gets greener than grass and feels
like dying. 174 To be sure, this is a poet speaking, at least in part, through
metaphors. But that does not alter the fact that Archilochus, Sappho, and
Alcaeus proceeded to examine almost clinically the '' madness of love,''
and that their poems are the oldest detailed descriptions of a psychoso-
matic state perceived to be abnormal. Medical literature and countless

anecdotes from antiquity show that the passion of love was in fact lived,
diagnosed, and treated like a disease. 175
The Greek lyric poets appreciated almost as much another thing that
beautifies life and can also threaten healt~: wine. In the Homeric epics,
the heroes never miss a chance to mix bowls of wine "as sweet as honey."
A line in the Iliad declares that it greatly swells the might of a tired warrior.
The lyric poets go further in the same line, praising drunkenness as an
ecstatic experience. The success of Anacreon's drinking songs illustrates
the role of alcoholism in Greek society. Poets and doctors described, each
from their own viewpoint, alcoholic intoxication. 176 However, chronic
alcoholism as a pathogenic factor completely escaped medical interest.
Is wine really more dangerous than water? The answer cannot be given
outside a specified historical framework. It depends on the supervision of
water quality, mean longevity, social activities, and so on. Wasn't it pref-
erable for an ancient Greek to risk cirrhosis of the liver at an age beyond
average life expectancy than dysentery from water in the flower of one's
youth? Anacreon, a great drinker, "exceeded the limits of human life"
according to Valerius Maximus; he died after the age of 85. An example
like him did not inspire abstinence. Still, the Greeks preached temperance
early on. Theognis of Megara insists in one of his elegies that "it is bad to
abuse wine; used in moderation, wine helps, not hurts. " 177
As a rule, wine was mixed with water and not drunk straight, in contrast
to the customs Greeks attributed to Scythians and Thracians, who were
considered nations of drunkards. According to Herodotus (VI, 84), the
Spartans ascribed the madness (perhaps a case of delirium tremens) that
destroyed their king, Cleomenes, to the fact that he had learned from the
Scythians to get drunk often and on unmixed wine. In this connection I
cannot resist the temptation to cite a funerary inscription that is, to be
sure, late, but whose topicality is not restricted to a single moment in
Greek history: "Asclepiades, son of Anaxippos, an Ephesian. Twenty-two
years of age, I drank in one gulp a large amount of unmixed wine and
died, spitting blood. " 178 If this victim had not been so young, the most
tempting diagnosis would be rupture of esophageal varices caused by cir-
rhosis of the liver. Actually, the fatal hemorrhage could easily have been
pulmonary or gastric as well.
The poet Alcaeus draws a strange picture of Pittacus (ea. 650-570 B.c.),
the tyrant of Mytilene who was canonized as one of the Seven Sages. If
we believe this account, Pittacus was less appealing to look at than to hear:
Alcaeus calls him sartipousor sarapos('splay-footed'), because he had flat feet and
shuflled; kheirop6des('chapped-footed') because he had cracks and chapped skin on
his feet;gaurex ('braggart') because he was arrogant;phuskonandgastron ('potbellied')
because he was obese, and again zophodorpfdas('dining in the dark') because he was
stingy with lamp oil, and agasurtos('slob') because he was slovenly and dirty.179
The ancient world~has left us other portraits of this kind in which, taking
off from the real or imaginary appearance of an Aesop or a Socrates, their
aesthetic defects border on the pathological.
Hipponax of Ephesus, a great lyric poet of the mid-sixth century B.C.,
was so ugly as to be the butt of jokes in his home town. In exile at
Clazomenae, his life was miserable. His poetry calls up the life of the
pauper: Hipponax knows what it is to be hungry and to shiver with
cold. 180 He complains that the god Plutus must be blind, since he gives
him nothing. 181 In Hipponax's surviving fragments, there is mention of
famine, chilblains (khimitla)) blisters (phofdes))incest, and castration. 182
The poet says of a rival that he is ill from gluttony. In this highly mutilated
text, the wordgasme meaning "abdominal cramp, colic" appears. At least
that is the explanation given by a scholiast commenti!}g on the passage in
a second-century A.D. papyrus. 183
It is a shame that only a few scraps survive of a poem by Hipponax on a
subject that may have medical implications. Long ago Daremberg re-
marked that in the fragments of this poet is to be found a passage depicting
''an unknown man who passed blood in his urine and bile from his
anus." 184 He even proposed as his diagnosis a vesico-rectal fistula. But
Daremberg only knew a single verse of the poem that is cited as an example
in a work of the Alexandrian grammarian Herodian. The discovery at
Oxyrynchus of a papyrus that contains the beginning of the verse allows
for an improved understanding of its context. 185 Here is the verse Hero-
dian cites to illustrate the usage of the verb omikhein 'make water, urinate':
" ... he urinated some blood and discharged some bile." On the papyrus
the passage continues as follows: "but as for me ... and my teeth are
all chattering in my mouth ... I run crazed ... fearing ... this per-
son ... " 186 Little is clear, and unfortunately the larger whole suggests less
medical significance than the single verse had led one to hope for. Accord-
ing to Olivier Masson, the subject is two victims of a great fright: first, the
narrator, whose fear makes his teeth chatter, then the other person, "on
whom the fear has a violent effect. " 187 If this interpretation is correct, the
expressions that seem to designate hematuria and bilious stools must be
considered metaphorical. Fright, no matter how terrible, cannot cause
blood to appear in the urine of a healthy person. The most common cause
of this symptom in classical an,d Hellenistic antiquity was stones in the
urinary tract. So the line from Hipponax does retain some interest for the
history of diseases in Greece.
In the second half of the sixth century B.C. Theognis, an aristocrat from
Megara, stresses in his elegiac poems the hereditary aspect of physical
virtues and defects 188 and maintains that poverty is worse than fever
(ep{alos).189 "Let's nip in the bud," he exclaims, "the ills of those near and
dear; let's try to find a salve for the ulcer as soon as it forms. " 190
I conclude this chapter by citing a few verses from an ode of the lyric
DISEASES IN THE ANCIENT GREEK: WORLD
poet Pindar in which he speaks of the primordial diversity of diseases.

True, this text belongs to the fifth century B.C. (it was composed probably
ea. 473), but it clearly preserves a reflection of the tripartite division of
diseases and their treatments that, as Benveniste and others think, was an
archaic characteristic of Inda-European th,ought. 191 While telling the tale
of Asclepius and his apprenticeship to the centaur Chiron, the poet de-
scribes as follows the latter's medical technique:
They came to him with ulcers the flesh had grown,
or their limbs mangled with the gray bronze, or bruised with the stone flung
from afar,
or the body stormed with summer fever, or chill; and he released each man and
led him
from his individual grief. Some he treated with guile of incantations,
some with healing potions to drink; or he tended the limbs with salves
from near and far; and some by the knife he set on their feet again. 192
According to this text, there are three sorts of diseases: those born in the
body without visible cause, those due to wounds, and those caused by
climate. Likewise, their treatment is of three kinds: charms, medicines,
and surgical interventions. For the history of diseases, I take note of the
importance given to "spontaneous ulcers," that is, skin diseases.
With the cultural flowering that takes place in the fifth century B.C., the
literary documentation in Greek texts becomes extremely rich and varied.
Historians like Herodotus and Xenophon provide priceless information
about diseases and about medicine. 193 The same is true for the poets and
philosophers of the Age of Pericles. The theater becomes a more and more
valuable source of evidence for understanding all aspects of daily life. How-
ever, for my subject all this documentation can take on only secondary
significance, since it is displaced by the appearance of technical writings by
physicians. Therefore I will only use these historical, philosophical, and
poetic texts insofar as they complement or illuminate what we learn from
the Hippocratic corpus. But before taking up that subject, a different
perspective based on unwritten sources demands attention. Until now
their contribution to the study of pathology in ancient Greece has been
neglected, at least in works of historical synthesis.
Chapter Two
PALEOPATHOLOGY
Evidencefrom Ancient Boneson Diseasesin Greece
The anthropological and medical examination of ancient human remains

is all the more valuable as a source of evidence for its being "objective"
and so dispensing with both the benefits and the drawbacks of language.
Nevertheless, the exploitation of this nonverbal source material has only
recently been embarked upon. Before that could happen, archaeologists
had to bring to light a sufficient quantity of ancient human remains and
date them correctly. Another necessary precondition was that the rapidly
expanding field of pathological anatomy forge the conceptual tools to
proceed from the identification of structural changes to their clinical and
epidemiological significance. Finally, the two disciplines, archaeology and
pathology, had to meet and adjust to their respective requirements and
techniques.
The Historical Developmentof Paleopathology
Paleo pathology, defined in 1913 by Sir Marc Armand Ruffer as ''the

science of diseases whose existence can be demonstrated on the basis of
human and animal remains from ancient times," 1 made its discreet and
timid de but toward the end of the eighteenth century, that is, after the
founding of organic anatomopathology by Morgagni. In 1774, J. F. Esper
recognized a fracture of the pelvis in a Pleistocene mammal and described
a bone tum or on the femur of a prehistoric cave bear. Modern examination
of this femur supports the diagnosis of a fracture with excessive callus
formation; but it is important, historically speaking, that Esper thought
he had found a primitive malignant tumor, an osteosarcoma. 2
47
The first investigators in this field, such as Cuvier and Walther, limited
their interest to gross traumatic pathology and to so-called rheumatic le-
sions in animal fossils. Not until the second half of the nineteenth century
did anthropologists and doctors attempt human paleopathology. Indeed,
it would have been fruitless to do so befor~ the complete revision, under-
taken by Rokitansky and by Virchow, of the doctrine and methods of
pathological anatomy. RudolfVirchow (1821-1902), the founder of cellular
pathology, himself made several major contributions to paleopathology.
However, as is often the case with great pioneers, at the same time as he
opened new horizons, he also engaged in misguided research. In order to
explain the frequency and special characteristics of osteoplastic lesions on
the vertebrae of prehistoric men and bears, Virchow created the nosologi-
cal concept of Hohl.engicht(cave gout), an unfortunate term because of the
incorrect etiological and pathogenetic explanations that its etymology sug-
gested. Worse still, Virchow interpreted as signs of rickets features that
were actually an anthropologic peculiarity of Neanderthal man. But these
blunders of an illustrious pathologist should not blind us to the value of
his other work for the constitution of a new branch of medical history.
Thanks especially to Rudolf Virchow and to the anthropologist and
surgeon Paul Broca (1824-80 ), the existence of pathological vestiges on
prehistoric human remains was accepted, and the interest of their system-
atic study was recognized. Parrot, Le Baron, Bartels, and others followed
in their footsteps, examining prehistoric and Gallo-Roman remains in
western and central Europe. Despite the incorrect interpretation of certain
lesions as rachitic or syphilitic, the work of these scientists produced new
and incontestable medico-historical knowledge. By the end of the nine-
teenth century, the presence of various ailments in ancient bones from
Europe had been proven: traumas, purulent and tuberculous inflamma-
tory processes, neoplasms, rheumatic ailments, tooth decay, rickets, and
so on.
The first three decades of the twentieth century amount to a kind of
Golden Age of paleopathology. Profiting from the overall progress of
knowledge in bone pathology and bacteriology, and with effective use of
the technique of histological sections and radiography, anthropologist-
doctors of this period identified a relatively high number of diseases that
left their traces in the remains of human and animal bodies. For reasons
having as much to do with methodological principles as with convenience,
the objects of choice were mummies and bones found in Egypt and on the
American continent (pre-Columbian civilizations) as well as prehistoric
bones exhumed in western Europe. The scientific achievements of this
period are recorded in the monographs of Marc Armand Ruffer, Roy L.
Moodie, and Leon Pales. 3
Paleopathology has been said to have made such progress that it should
no longer be satisfied with a simple medical examination of ancient lesions,
P ALEOP ATHOLOGY 49
but that "in contributing to the study of the evolution of pathological
processes and pathogeneses, it belongs to general pathology. " 4 In short,
pathology should serve history at the same time as history serves pathol-
ogy. But despite such a noble goal, paleopathological research seems to
have lost its inspiration. From the 1930s to the middle of this century, it
confined itself to confirming diagnoses made before or to multiplying
them. Pharaonic Egypt, pre-Columbian America, and prehistoric Europe
continued to fascinate some paleopathologists. Here and there one can
find a few cases cited for Roman Gaul or the European Middle Ages. Aside
from the first studies of John Lawrence Angel, which are still totally un-
known to historians of medicine, absolutely nothing is known about the
paleopathology of Greece and Italy in the classical era. 5 This explains why
E. H. Ackerknecht, when taking stock of modern paleopathological re-
search, even in 1953defined as its domain "the pathology of prehistoric
animals and man in prehistoric and unlettered societies.' ' 6 The restriction
proposed in this definition was soon shown to be inappropriate by subse-
quent developments in paleopathology. Even when written documenta-
tion of pathological events is at hand, paleopathological investigation is
not superfluous. Standing at the crossroads of medicine, anthropology,
and history, paleopathology is currently undergoing extraordinary growth
in the quantity and quality of specialized publications, in the scope of its
field of interest, in new modes of analysis, and in the organization of
special institutions.7 In my view, three tendencies characterize recent prog-
ress in paleopathology: the use of new scientific methods; the movement
from the study of isolated cases to the paleopathological investigation of
whole populations; and the widening of the research area to include clas-
sical antiquity, the Middle Ages, and even modern societies.
Today, microradiography is a useful complement to paleopathological
research. Calvin Wells has shown the importance of certain calcification
lines whose X-ray study makes it possible to evaluate the exposure of
children in an ancient population to famine or disease. It is known that
the long bones grow along the epiphyseal line and that the process can be
interrupted by undernourishment or by certain infectious or parasitic ail-
ments. When growth begins again, a transverse line of dense calcification,
called "Harris' line," forms between the diaphysis and the epiphyses of
the two sides. Harris' lines mark -~hebones permanently, and their number
indicates the frequency of distress during an individual's childhood. So by
examining all the tibias or femurs in a necropolis for such lines, we can
obtain valuable information on morbidity and subsistence crises. 8 By sim-
ilar observations of healing lines on the female pubic symphysis, it is pos-
sible to guess the number of times a woman has given birth, and, by
multiplying such results, to estimate the fertility of an ancient population.
Likewise, analysis of the microscopic structure of teeth provides informa-
tion about eating habits and nutritional deficiencies.
Among the new techniques, the most promising and extraordinary are
those of paleo-immunology. This branch of paleopathology is still in its
early stages. Thanks to sophisticated microserological processes, it may
become possible to determine the blood type (ABO system) of persons
long since deceased even if only a smal). quantity of dessicated bone re-
mains. The results obtained so far are encouraging, though it cannot be
ruled out that post-mortem impregnation of bone tissue by bacteria or
molds can falsify this type of serological reaction. Without much concern
about this as a possible source of errors, systematic research has been done
on the frequency of ABO groups among the ancient Etruscans, the prehis-
toric inhabitants of southern Italy, Egyptians of the Pharaonic period, the
ancient Khmers, and pre-Columbian American Indians. Given the genetic
stability and the relatively simple mode of transmission of these blood
traits, knowledge of their distribution in the past and today allows the
historian to reach significant conclusions about the origin and migration
of peoples. However, in the present state of our knowledge, this study
sheds no light on the history of diseases, since there is not a significant
enough correlation between blood groups and pathological phenomena.
From the viewpoint of paleopathology, it would be much more impor-
tant to be able to determine, from an examination of bone remains or at
least of mummified tissues, histocompatibility types of the HLA (human
leukocyte antigen) system. Jean Dausset, who proposed this system in
1958,and his collaborators and followers have shown in their recent studies,
first, that the genetic traits of the HLA complex are true "biological
markers'' that allow one to follow human migrations, and, second, that
there are associations between HLA antigens and certain diseases. For
instance, a carrier of tissue group HLA-B27 is 120 times more likely to
contract ankylosing spondylitis than an average person; the risk is increased
500-fold if the carrier of the tissue group is male. These medical discoveries
could open new horizons in paleopathology if a way is found to perform
antigenic HLA typing on ancient biological materials. Some as yet uncer-
tain results have been obtained on the tissues of pre-Columbian mummies. 9
All hope is not lost that specific immunological tests for various infec-
tious diseases can be applied in paleopathological research. There is noth-
ing surprising about the failure of numerous attempts at specific
immunological diagnosis from ancient bone or tissue: if the molecules of
antibodies are still present in them, it could only be in infinitesimal quan-
tities. To succeed, perfection or wholesale revision of traditional serological
methods will be necessary. In the near future, scanner electron microscopy
will perhaps make it possible to localize blood corpuscles, large organic
molecules, and pieces of nuclear chromosomal matter. Once identified in
this way, such highly specific organic structures as these can probably be
subjected to chromatographic and serological analysis.
Sometimes the ancient remains of human corpses contain the eggs of
PALEOPATHOLOGY 51
fossilized parasites 3-9-dtraces of bacteria, the true nature of which can be
difficult to determine. However, as against the consistent frustration of
current paleo-immunological research on the antibodies that pathogenic
agents give rise to in a host organism, it has been possible to prove, by a
standard immunological technique, that pathogenic bacterial antigens are
present in ancient organic materials: in 1976, a team of American research-
ers was able to demonstrate the existence of salmonella antigens (most
likely type D, i.e., Salmonella -typhi) the agent of typhoid fever) in the
intestinal contents of a Peruvian mummy. 10 Microscopic examination of
coproliths and deposits found ~uring archaeological excavations in latrines
or similar places makes possible the discovery of eggs and cysts of various
intestinal parasites, as has been shown in several cases going back to the
Middle Ages or the Roman Empire. 11 In addition, we have at our disposal
today methods of chemical microanalysis that can serve to determine the
concentration of normal elementary components of bone tissue-the cal-
cium, strontium, and phosphorus values are especially useful indicators of
health-or to detect the presence of inorganic toxic substances, for in-
stance, lead in the bones of ancient potters. 12 None of these new tech-
niques has yet been exploited in paleopathological studies of materials from
Greece. The reason I have stressed the possibilities that they offer the
historian in search of original approaches is to sketch a program and open
perspectives, not to criticize better, in the rest of this chapter, the results
already obtained. My purpose is to call attention to a lacuna in historical
research on archaic and classical antiquity.
Careful examination of human remains can provide information not
only on serious diseases and grave lesions but also on small, almost normal
factors, such as wear and tear from age or work, longevity, the state of
bodily development, or undernourishment. Just as in political or eco-
nomic history, the interest of contemporary researchers is moving away
from exceptional events and closer to common facts, to quantifiable evi-
dence about daily life. At first, the curiosity of paleopathologists was piqued
by the somewhat baroque aspect of certain "cases" and by the antiquity
of the specimens. Nowadays the preference is for a necropolis instead of
an unusual skull, interest is not limited to singular "discoveries," and
there is great willingness to emerge from the domain of prehistory. Indeed,
it is not important to have paleopathological proof of the existence of a
disease within a historical population for which written documentation
of that fact is to hand. What matters-and this in itself transforms
paleopathological investigation-is evidence of the anatomopathological
peculiarities, the diachronic and geographic prevalence and distribution of
the disease in question. By offering information on gender, lifespan, height,
body build, and even fertility, the systematic examination of bones and
mummies enriches historical demography and makes possible the creation
of a new branch of it, called "paleodemography." The American anthro-
pologist John Lawrence Angel has spelled out in brilliant fashion the ideo-
logical and methodological bases of this new approach, and he has
demonstrated concretely its practical applications. Nor is it without inter-
est for our subject that Angel had done most of his work on Greek
necropolises. 13
All ancient populations have several characteristics in common: reduced
longevity, frequency of inflammatory processes and of trauma, rarity of
tooth decay, and presence of tuberculosis, among others, but in numerical
variations that can be significant. We emphasize the crucial importance of
the statistical analysis of paleopathological evidence. The main advantage
of paleopathology is to by-pass the mediation of language and conse-
quently to become the complement or even the replacement of written
sources. However, the nature of the material remains, chiefly the hard
portions of the human body, poses severe limits on the number of diseases
and the biological facts that can be arrived at in this way. Considering that
each ancient human bone is a historical witness and that in its examination
the borders between normal and pathological fade, Vilhelm M~ller-
Christensen has proposed that the branch of science that concerns itself
with them be called "osteoarchaeology. " 14 Its area of study is in one sense
broader than that of paleopathology, since it also studies the nonpathol-
ogical aspects of bones. In another sense, osteoarchaeology is narrower,
since its purview includes no direct evidence of disease other than bones
and teeth of human or animal origin. In any case, the definition of osteoar-
chaeology is well-suited to the conditions of research in Greece, given the
absence of mummified tissue and the significance of demographic data.
Moreover, the term appropriately brings to mind the links of this disci-
pline with archaeology.
Earliest Osteoarchaeological
Researchin Greece
Paleopathology has taught us more about the character and frequency
of diseases in Egypt in the time of the Pharaohs, and even during the
Hellenistic and Roman periods, than the most determined and ingenious
analysis of the literary tradition. But the situation in Greece is altogether
different. The results obtained up to now are very incomplete, especially
for the classical period. We will not waste time blaming the archaeologists
who, while excavating archaic and classical sites, too often treated human
remains in a way that has made it difficult if not impossible to undertake
paleopathological examination of the unwritten medical documentation.
The famous Heinrich Schliemann (1822-90) tells how, during his first ex-
cavations at Hissarlik, Mycenae, and Tiryns, he would come across skele-
tons that he could neither examine with the necessary care nor preserve
for subsequent study. Here is an excerpt from his account of the excava-
tions of 1876:
PALEOPATHOLOGY 53
One of the most interesting objects I found at Tiryns is the skeleton of a man, at a
depth of five meters. The bones are petrified, but I attribute this phenomenon to
the nature of the soil in which the skeleton happened to be encrusted. Some of
the bones are considerably swollen because of the humidity; this is probably why
the lower jawbone is of such enormous thickness. Unfortunately, I could only
preserve a portion of the skull. 15
It is really a shame that this specimen was lost, since we doubt that Schlie-
mann' s opinion on the pseudo-pathological, post-mortem nature of the
mandibular swelling is well-founded.
In the midst of his archaeological discoveries, Schliemann made the
acquaintance of RudolfVirchow, at the time not only the greatest author-
ity on pathology but renowned as well for his competence in archaeology,
anthropology, and ethnography. A friendship grew up between the two. 16
The Berlin professor's backing was a great help to Schliemann who, as a
dilettante, was having difficulty gaining an audience in certain official sci-
entific circles. In 1879, Virchow traveled to Greece and Turkey, examined
skulls found in ancient tombs and kept at Athens, participated in the
excavations at Hissarlik by Schliemann 's side, and made a series of medical
observations in the region. 17 Schliemann entrusted him with the bulk of
his osteological finds and deposited his collections at the Museum of Eth-
nology in Berlin.
The formidable challenge of making the first detailed analyses of the
bones exhumed at Troy thus fell to Rudolf Virchow. He met it with the
publication, in 1882, of a monograph on the Trojan skulls, in which he
described the bone finds from Hissarlik as well as several skeletons discov-
ered by F. Calvert at Ren Koi and Hanai Tepe. 18 At the outset, Virchow
expresses his regrets concerning the small number of specimens analyzed
and their poor state of preservation. The archaeologists, he says, often
damaged, simply destroyed, or just buried bone finds, especially bones
other than the skull. At the time it was thought that only the skull was
relevant to anthropological research, the main goal of which was the deter-
mination of racial types. Although he was theoretically critical of this
attitude, Virchow generally adopted it himself in the practical portions of
his study. To be sure, he observed some anomalies of no consequence for
pathology (metopic suture, platycnemia, and the like), and he described
some pathological lesions as well, such as fusion of the cervical vertebrae
and the fracture of long bones. He determined skeletal sex and age group
and observed the way teeth were worn. He stressed the existence of sturdy
trochanters and hypertrophied bumps at points of muscular insertion,
signs that some males were exceptionally strong. But these are marginal
comments. In truth, the great pathologist neglected paleopathology. In
reality he was only interested in measuring skulls for supposed racial attrib-
utes. So, for example, he explicitly excludes the examination of two skulls
from Ren Koi because they were obviously pathological and therefore
might invalidate the racial significance of his measurements. Nothing can

be said against such a procedure, but it·is astonishing that Virchow does
not offer a word on the nature of the pathological distortions in question,
and that he completely excludes the two skulls from his investigation.
When bones from the ancient world wer~ before him, this expert in mod-
ern pathology became engrossed in purely anthropological problems.
In 1884, Virchow published a second study of Greek skulls. 19 In it he
analyzed bones from Behram Koi (ancient Assos in the Troad) and from
Cyprus. This publication includes an excellent presentation of two cases of
skull trauma. In 1890, another trip to Greece provided Virchow with the
opportunity to examine skulls dug up at new excavations in Athens. Fi-
nally, the discovery of a skeleton thought to be the mortal remains of
Sophocles resulted in the description of a very strange case of asymmetry
in the brainpan. 20 Several aspects of this work are admirable: the analyses
it contains are very precise, and its conclusions are original, perceptive,
and well thought out. I shall speak of it again in the systematic portion of
this chapter.
Clon Stephanos (1854-1914), a Greek doctor who had assisted Virchow
during the latter's trip to Athens, published an article in 1884 on Greece in
Dechambre's medical dictionary. 21 In point of fact, it is the best overview
of Greek medical geography in the nineteenth century. A paragraph in it
on anthropology informs us that at that time there was knowledge of only
about 90 ancient skulls exhumed in Greece and the lands it colonized (29
from Attica, 22 from Troy, 16 from Ionia, 4 from the islands, and 19 from
southern Italy and Sicily). They had been measured and studied by Vir-
chow, Nicolucci, Quatrefages, Broesike, and Zaborowsky, whose overall
conclusions were that the so-called "Pelasgians" were brachycephalic while
the "Hellenes" were dolichocephalic. Stephanos expresses himself circum-
spectly on this subject, stating that the ancient Greeks were, as distinct
from the Pelasgians, "in great part dolichocephalic, though it isn't possible
to state absolutely that the early Greeks were dolichocephalic." He stresses
that the number of skulls examined was relatively small and that for most
areas in Greece not one ancient skeleton was known. Furthermore, Ste-
phanos draws attention to the fact that most Greek statues of nonmythical
persons are brachycephalic. 22
According to the measurements ofVirchow and Nicolucci, the average
cranial capacity of Greek skulls was ''one of the lowest known to anthro-
pology" (1388.7 cc. was the average for 18 skulls; more exactly, 1418 for
males and 1276 for females). The only explanation for this surprising and
embarrassing statistic was to judge it provisional and insignificant because
of the small number of specimens under examination. 23
The anthropologists discussed up to this point took no notice of bones
other than the skull, and as for pathology, their interest went only as far
PALEOPATHOLOGY 55
as the mention of a few healed fractures. In sum, Greek osteoarchaeology
began promisingly, but after the initial bursts of enthusiasm, scientific
curiosity in the subject seemed spent. Virchow's work had no immediate
successors. Even though continuing archaeological investigators brought
new bone n1aterial to light, its study was neglected. We do note the work
of the anthropologist Ionnis G. Koumaris, who published, in the begin-
ning of the twentieth century, several sets of skull measurements and an
article on non pathological variations in Greek skull bones (especially me-
topic sutures). 24 As a whole, paleopathological research in Greece was
stagnant just when it was achieving its greatest successes in Egypt, Amer-
ica, and France. In the basic works on paleopathology, that is, the mono-
graphs of Ruffer (1921), Moodie (1923), and Pales (1930), there is no
information whatever about Greece. That fact corresponds to the com-
plete lack of osteoarchaeology in Greece during the first three decades of
the twentieth century.
In 1951, Sigerist summarized the progress that had been made in paleo-
pathology around the world during the first quarter of the twentieth cen-
tury and compiled a fine bibliography on the subject. 25 He also was
completely silent on the subject of Greece. But the silence was no longer
justified, since in the meantime C. M. Furst had published studies on
skeletons from the Argolid and Cyprus, 26 E. Breitinger had analyzed bone
finds from the German excavations of the Kerameikos in Athens, 27 and,
most important, J. L. Angel had begun the anthropological and paleo-
pathological investigation of the new American excavations and had al-
ready undertaken a complete review of all ancient Greek bones. 28 While
Furst and Breitinger perfected the purely anthropological approach and
still limited themselves to gross pathology (fractures, bony growths, etc.),
Angel gave Greek osteoarchaeology a new direction.
John Lawrence Angel, a British (born London, 1915) anthropologist
working in American scientific institutions-he was curator of the Division
of Physical Anthropology at the Srnithsonian Institution in Washington-
is at the moment the almost unique representative (one might almost call
him the personification) of the osteological paleopathology of prehistoric,
archaic, and classical Greece. Angel's first trip to Greece goes back to the
fall of 1937 when, as a young graduate student, he joined the team excavat-
ing the Athenian Agora. From the start he wished to unravel the social
biology of this area, or, more precisely, to find, at the very place where
Western civilization had been born, anthropological responses to the ra-
cialist ideas of Nazism. 29 His ambition was to make a systematic study of
the changes in the biological substratum of the populations that had suc-
ceeded one another in Greece and the neighboring regions since prehistoric
times. Accordingly, he tried to localize and examine all the ancient skeletal
remains from Greek lands, whether they were in private or public collec-
tions. At the time of this, his first review of osteoarchaeology in 1937-38,

Angel was able to analyze 455 skulls and 132skeletons dating from Neolithic
times up until the Byzantine period. In his publication of the results of
this inquiry, he remarks on the poor state of preservation of the bones due
to marked seasonal fluctuations in Greek soil conditions, specific burial
customs during some periods of Greek history, and the carelessness of
archaeologists with human remains. 30
As for the attitude of the archaeologists, the situation has changed since
the 1930s and especially since World War II. So Angel was able to profit
from the rich harvest of human bones excavated by D. M. Robinson at
Olynthus, H. A. Thompson in the Athenian Agora, J. L. Caskey at Lerna,
G. Daux and P. Courbin at Argos, C. W. Blegen at ancient Pylos and in
the Troad, P. Dikaios at Khirokitia (Cyprus), J. L. Benson at Bamboula
(also Cyprus), R. J. Rodden at Nea Nikomedia, M. J. Mellink at Karata~,
T. W. Jacobsen in the Franchthi Cave, J. Mellaart at <_::atalHiiyiik,
S. Dietz at Asine, and others. By 1973, Angel had examined the mortal
remains of approximately 2,200 persons from the Paleolithic up until the
modern era. For the first time, it was possible to study systematically the
sum-total of bones found in the relatively important necropolises, and to
draw sure conclusions about the biological traits of populations, not just
individuals. From this point of view, Angel's monograph The People of
Lema is a model of the genre. 31
Besides this American research, one should cite the studies by Robert P.
Charles on the skeletons discovered during the French excavations in Ar-
gos and on those of the ancient inhabitants of Crete, the examination
of bones from the Kitsos Cave in Attica done by J. Dastugue and
H. Duday, the articles by N. G. Gejvall and F. Henschen, C. S. Barsocas,
A. D. Tsouros, H. G. Carr, and D. M. Hadjimarkos on specific paleo-
pathological and paleodental questions, and the work of V. G. V allaoras
and M. S. Senyiirek on paleodemography. 32 Even so, Greek osteo-
archaeology is a branch of scholarly research in which much remains to be
done. Though the current total of skeletons examined surpasses 2,000,
oftentimes the specimens are incomplete and fragmentary. They are also
scattered in time and space. The contents of prehistoric burials are now
better known than those of the classical period, for which the number of
skeletons examined is well below 200. For most archaeological sites, the
local sample is very small. There is a relative abundance of bones from
peripheral areas, such as Anatolia, Macedonia, and Cyprus, and although
some places of major significance are well-represented (Athens, Mycenae,
Lerna, Argos, etc.), regrettably, the most interesting places for historians
of medicine are totally lacking in osteoarchaeological finds. We know
nothing of the paleopathology of Cos, Cnidus, Thasos, or Epidauros, to
mention a few sites.
PALEOPATHOLOGY 57
Traumas
Wounds to the skin, muscles, and internal organs are only by exception
amenable to paleopathological investigation. The opposite is true for frac-
tures, which affect the solid and long-lasting parts of the body and also
leave recognizable traces even when they heal perfectly. The percentage of
Aegean bones that were fractured during life is not excessive. It is on the
order of10 percent (of the number of individuals, not the number of bone
specimens). The real frequency of fractures must have been greater, since
a negative diagnosis cannot be confirmed in the case of incomplete skele-
tons, which are, unfortunately, especially numerous. Despite the need for
this correction, the frequency of fractures in ancient populations does not
seem close to that of current societies with developed technologies. The
difference is explicable partly in terms of increased longevity, partly in
terms of new activities, such as transportation by powerful and relatively
dangerous vehicles, sports, and so on.
As far as the prehistory and ancient history of Europe (including Greece)
are concerned, it appears that the risk of fracture is inversely related to the
progress of civilization. 33 The oldest completely preserved skeleton found
in Greece dates from the Mesolithic era (ea. 7000 B.c.). It was discovered
by Thomas W. Jacobsen in the Franchthi Cave, a prehistoric site along the
Gulf of Nauplia. It is not a coincidence that this skeleton is that of a man
who perished at around age 25 from a series of blows to the front of his
skull. So the most remarkable bone evidence from prehistoric Greece makes
its entrance in a context of violence, of struggle between men. 34 In most
osteoarchaeological cases of skull trauma, the locale and nature of the
lesions make it clear that they result from conflict, from intended acts and
not from accident. The Franchthi man was bashed on the head by a rock
or a club, and the scars and oblong dents on certain skulls of the Bronze
and Iron Ages lead one to suspect the use of blades. Of two individuals
with head wounds at Karata~ (the only two cranial traumas in 560 skeletons
from the Early Bronze Age), one succumbed to a fracture of the temporal
bone (skull 522 J(a. ), while the other survived a wound to the top of the
skull that reached a depth of 5 millimeters (165J(a.). 35 Among the princes
or aristocratic warriors buried in Grave Circle B at Mycenae (Middle Bronze
Age, 1650-1450 B.c.), two were_ wounded in the head. One skeleton (59
Myc.) is that of a sturdy man, very tall and well-built, with noticeable
depressions in the skull vault above the left eye and behind the left parietal
boss. These are apparently the results of rugged combat, yet they did not
cause this chieftain's death, which overtook him at the age of 50 years or
more. The other skeleton (51 Myc.) is that of a 30-year-old man of excep-
tional strength. On the right of his frontal bone he has an oblong, shallow
depression (23 mm. long) that is the mark of a healed wound, and higher
to the left the traces of a fatal fracture, unsuccessfully treated by a surgical

intervention. 36 Such details are indeed reminiscent of the exploits and
destinies of the Homeric heroes.
Among the skeletons from Lerna, most of which date from the Middle
Bronze Age, Angel has counted 10 cases of cranial trauma. Of the 10, 8 are
relatively mild or, at least, completely healed. Among those with well-
healed unions, 4 are on male skeletons (IS) 23) 66) and I32 Ler.) and 4 on
females (38) 59) I78) and I82 Ler.). To take a closer look at one example,
consider the skeleton 59 Ler.: it's a young woman, about 25 years of age,
short and slight; on her skull, just to the right of the vertex, there is a
sharp, traumatic depression without loss of substance. 37 In 2 of the 10 cases
from Lerna the wounds probably resulted in death (I8I and I89 Ler.), as the
skull splits and the state of the fractured edges indicate. The skeleton I8I
Ler. is the mortal remains of a man about 45 years old, short but with an
athletic body, who suffered from deformations of the vertebrae. His skull
presents an irregular fracture with loose bone fragments; the inner table is
split off and curled as if by an imploding force. 38 He might have been hit
by a stone or club, or taken an especially bad fall. Wounds similar to this
one are described in the Hippocratic corpus. 39
On a skull from the main necropolis of Argos (I68 A13.; protogeometric
tomb, tenth century B.c.), probably of a female in her middle twenties,
R. P. Charles noticed that the bone wall over the left frontal eminence was
crushed as the result of a violent trauma. It is interesting to note that
despite the seriousness of the lesion, which certainly did significant damage
to the central nervous system, the victim survived the accident: the inside
of the bone shows breaks that were completely healed. 40
In the Kerameikos cemetery in Athens, Breitinger found an old man's
skull with a scar of traumatic origin in the form of a crater on the left side
of the frontal bone (Ioo C I<eram.). The cervical vertebra of an adolescent
whose head was buried in the protogeometric era near the Hephaisteion
in the Athenian Agora has the marks of decapitation. 41 During the older
excavations at Assos in the Troad (now the Turkish village Behram Kai), a
male skeleton was discovered (I Ass.) dating from the time of the Lydian
conquest or the first Persian occupation (sixth century B.C.). The skull
shows signs of two wounds that were made by a sharp weapon. On the
frontal bone, there is a fracture line 30 millimeters long, well healed, which
stretches diagonally from the middle of the forehead to the middle of the
top edge of the left eye socket. An additional scar, parallel to the first but
very short, cuts across the lower edge of the same socket. Also in this spot,
the blade of the weapon detached and lifted a small splinter of bone that
managed to heal in an abnormal position. This man, probably a warrior,
certainly lost an eye and probably suffered brain damage from the shock,
but he did not succumb to the blows whose traces we have just described.
PALEOPATHOLOGY 59
He died at an advaqced age, as his heavily worn teeth, among other things,
attest. On another male skull (2 Ass.) uncovered in the same place as the
previous one but dating from the third century B.C., Virchow recognized
the signs of a broken nose. 42 Three persons of the protohistoric era in the
Argolid, two men at Lerna (I8 and 50 Ler.) and one woman at Asine (I8
FA.), have broken noses, probably the result of accidents or combat. One
the same occasion the woman from Asine also lost several of her front
teeth. There are exostoses above her right wrist, interlocking radius and
ulna; they suggest a defensive response to violence. Two women from
(:atal Hiiyiik (86 and Io8 CH.) and a n1an from Karata~ (I89 J(a.) had their
lower jaws broken near the chin. All three healed solidly. 43
As might be expected, skull injuries, especially the more serious ones,
occur more often in men than women and statistically tend to be situated
on the left- rather than the right-hand side. Discussing such wounds in the
people of Lerna, I mentioned the skeleton 59 Ler., that of a slight woman
hurt on the right side, purposely to stress the existence of atypical cases.
The localization on the left side is more frequent because, then as now,
usually the right hand wielded the weapons. The greater number of wounds
in males is easily understood if we assume that skull traumas are mainly
caused by violence between people. The same preponderance of the male
also occurs, though to a lesser degree, in the statistical distribution of
certain long-bone fractures that are usually the result of accidents. Taking
all fractures together, the relation between female frequency and male
frequency is on the order of 1:4. Males were more exposed to the danger
of fractures not only as warriors and fighters but also by the nature of their
activities outside the home.
Angel drew attention to the fact that fractures in children are rare. Of
100 children's skeletons in Lerna and Argos, he noticed only 1 with a
damaged bone, and that was probably caused by a difficult birth, not an
accident thereafter. 44 Does this mean that the children of certain Greek
populations were especially well protected and watched and that they led
a relatively uneventful life? The exceptional specimen just mentioned is an
infant, dead after scarcely a month of life, whose right collarbone shows a
swelling, probably a callus, at the junction of its middle and outer thirds
(I62 Ler.).
Generally, fractures of the collarbone (for instance, an adult case in (:atal
Hiiyiik and also one in Karata~) and of the humerus are accidental. Three
instances deserve mention here. On a Neolithic humerus from (:atal Hii-
yiik (IOI CH.) belonging to a 40-year-old woman, the shaft, which had
broken midway, at deltoid insertion, has knitted solidly and with slight
medial angulation, but the upper piece is rotated. The rotation must have
resulted in functional difficulties. A male skeleton from Asine (IIO As.) has
a right humerus fracture below midshaft with about 15° angulation, short-
ening, and thickening, along with arthritic lipping at the elbow and ero-
sion of the capitulum. 45 On the other hand, a comparable midshaft fracture
in a woman from Mycenae (58Myc.) healed remarkably well. 46
If a much larger number of specimens were at our disposal, it would be
possible to use the way upper and lower. arm fractures heal to date the
introduction and spread of the therapeutic practices codified in the Hip-
pocratic corpus. According to the author of the treatise Fractures) ''The
treatment of a fractured arm is not difficult, and is almost any practitioner's
job. " 47 After this peremptory statement, the Hippocratic author goes on
to describe in minute detail how carefully to proceed in fitting together
the fragments, how to put a limb in traction in some cases, and in every
case how to keep the broken bone in place with bandages and devices. 48
The case 58 Myc. of a perfectly healed arm fracture belongs to the Middle
Bronze Age. This woman, who was buried in a royal tomb, was an aristo-
crat. As a consequence of her social position, she could call upon highly
qualified physicians. When the skeletons of persons of the same era but of
lower social standing can be examined, as with the inhabitants of Asine
and Lerna, more often than not fractures of the arm have faulty unions,
healed in abnormal positions. However, in the case of Lerna it should be
noted that we find fractures of the forearm and not of the humerus.
Two types of fracture of the forearm dominate the Greek osteoarchaeo-
logical record: (r) breaks of continuity in the middle of the ulna resulting
from a defensive movement (holding up the arm to parry blows aimed at
the head or the trunk) and (2) lesions of the lower portions of the radius
and the ulna caused by falls (from breaking a fall with one's arm). In
prehistoric times most fractures like these healed poorly and resulted in
persistent dysfunctions. Forearm bone trauma disabled its victims and was
a nagging, unsolved problem for the bonesetters of the day.
To exemplify the first type of fracture, we cite a Neolithic case from
<;atal Hiiyiik: in a 30-year old male, the left ulna had been broken in the
middle of the shaft. The fragments are joined together by a kind of fibrous
pseudo-arthrosis with a cauliflower formation from the bridging callus (I84
CH.), 49 Such overdeveloped pseudo-arthroses form when the broken bone
is insufficiently immobilized during the repair process. The same type of
fracture of the ulna as the result of a defensive movement has been re-
marked on two other skeletons from <;atal Hiiyiik as well as on bones
from the Kitsos Cave (also Neolithic) and the necropolises ofKarata~ (Early
Bronze Age), Asine, and Lerna (Middle Bronze Age). Often the union is
imperfect and shows signs of periostitis. Again, a man from Lerna, aged
about 50, has both bones of his right forearm broken, radius and ulna, at
the junction of the lower quarter (73 Ler.). A blood clot filled the space
between the two bones, which became fibrous and mineralized, and a
bony bridge formed, fusing the radius and the ulna in a slightly pronated
position. 50 In Athens, in the Kerameikos, a skeleton was exhumed from
PALEOPATHOLOGY 6I
tomb 93 (Iron Age,,, twelfth-tenth century B.C.) of a man about 50 years

old. His right ulna has the marks of a well-healed fracture. 51 This probably
means that by his time effective methods for the extension and immobili-
zation of fractures were becoming known in Greece.
All the fractures of the ulna just mentioned are of the first type, the
result of a blow to the forearm raised in a parrying movement. The other
common type of forearm fracture occurs when a person falls forward onto
an outstretched hand, a reflex gesture that is intended to break the fall (the
so-called Pouteau-Colles fracture). It particularly affects the lower end of
the radius, with or without involvement of the ulna. As examples, I can
cite two cases, one from C.::atalHuyiik (SI CH.) Neolithic) and the other
from Lerna (207 Ler.) Middle Bronze Age). The first is an isolated fracture
of the radius of classic form in a young male. The second is more infor-
mative: a woman, who gave birth more than once and died at about age
30, had undergone, probably during adolescence, a fracture of the right
radius in the typical place. Along with it was a spiral fracture of the right
ulna. The bones healed solidly, but a dorsal deviation at the wrist and a
slight shortening of the forearm resulted. 52
Fractures of the metacarpal bones in the hand are attested in two indi-
viduals from Lerna, a man (89 Ler.) and a woman (7 Ler.). Usually such
lesions are the result of a direct, violent action, such as blows to the hand
with a blunt object. In the great majority of cases, fractures of the leg are
caused by accidents, usually falls. Angel catalogues four fractures of the
femur on Neolithic skeletons from C.::atalHiiyiik. The case of r65 CH. is
typical: the left femur of a woman who perished at around age 25. On the
upper third of the shaft there are traces of an old so-called greenstick
fracture, an incomplete break in which the outer layer of the bone is not
ruptured through its whole circumference. The bone healed well, without
being shortened or developing a large callus, but with slight medial angu-
lation and rotation of the upper fragment. 53 Two healed cases of greenstick
fracture of the femur were found at Karata~ (II7 and r22 J(a.). Lesions of
this kind are characteristic of children or adolescents-they are the result of
falling from a certain height, as from a ladder or when scaling a wall. Proper
fusion of such fractures is more the work of nature than of the physician's
art. The left femur of a female (r37 CH.) provides another example of a
greenstick fracture in which the young bone's elasticity saved it from a
complete break. In a protohistoric inhabitant of Asine in the Argolid, the
shaft of the femur was broken clear through, and the pieces were thrown
out of place, with the upper jammed askew into the lower. The result was
a spontaneous fibrous union, with a tough callus in very poor position,
shortening the leg and twisting it. 54
An accident may well be the cause of the deformation of a man's foot
in the Middle Bronze Age (r27 Ler.). And I mention in passing a fractured
seventh rib in a man from Lerna (88 Ler.), an ordinary mishap that needs
no special attention for a good repair. Fractures of the spinal column are
much more important, for they often give rise to troublesome complica-
tions on the part of the nervous system. If the spinal column of a woman
from Nea Nikomedia (Neolithic, ea. 6000 B.c.) was really broken clear
through before death, as Emily Vermel;lle suggests, then death surely fol-
lowed. The Macedonian tomb in which she was found provides a moving
glimpse of a family tragedy: the miraculously well-preserved skeleton of an
adult female, tall, sturdy, with excellent teeth, is curled up in a fetal
position and holding in her arms the skeletons of two children. Plainly,
they perished along with her. 55 Fractures of the vertebral bodies have been
described in another woman of the Neolithic era found at <_::atal Hiiyiik (s2
CH.), in a Middle Bronze Age denizen of the Argolid (76 Ler.), and in a
Mycenaean prince from the Late Bronze Age (25 Myc.). In all three cases,
the fractures are relatively benign and seem not to have brought on
paralysis.
The skeleton 25 M ye._,exhumed long ago by Schliemann from Grave 5 of
Circle A of the royal burials at Mycenae, appears, like the others, to be a
robust and muscular man who died at about the age of 40. Aside from a
few signs of spondylitis on the thoracic and lumbar vertebrae, there is a
definite wedging of the third lumbar vertebral body that is strongly remi-
niscent of a well-healed youthful fracture. 56 A similar case has been noticed
among the skeletons from Lerna (76 Ler.). It is a man who died at about
age 40, but unlike the Mycenaean prince, he was on the slender side. His
skeleton shows a disk injury between the ninth and tenth thoracic verte-
brae as well as a fusion (with a possible old fracture) between the eleventh
and twelfth thoracic vertebrae. 57
The process of repairing a fracture is often complicated by inflammatory
reactions due to infection. In about one-third of the prehistoric and pro-
tohistoric skeletons, the callus that results from the union of the fracture
shows traces of periostitis. There are also bone lesions in the absence of
fractures that attest to the existence of wounds in soft tissue that, as a
consequence of inflammation, have affected the adjacent bone tissue. Thus
a serious inflammation of the left ischium has been remarked in a 40-year-
old male from <_::atalHiiyiik (97 CH.), probably ·caused by an infected
wound of the buttocks. According to Angel, it might be the result of
being gored by a bull. 58 Comparable wounds sometimes occur in contem-
porary bullfights. Even so, there is no reason to exclude the possibility
that a weapon did the damage. It was indeed a weapon, either a javelin or
a sword or the like, that pierced the back and made an elliptical hole in
the right shoulder blade of a stout warrior from the Middle Bronze Age
(!}1 Ler.). 59 The inflammation around the perforation proves that the
wounded man, a 40-year-old, did not immediately succumb. In another
man of the same era and vicinity (175 Ler.), periostitis in the lower part of
the left fibula is the tell-tale sign of an infected leg wound. 60
PALEOPATHOLOGY
Cranial Trepanning
By means of archaeological dating of trepanned skulls exhumed in vari-
ous regions, it has been determined that the surgical procedure of trepa-
nation goes back to Neolithic times. 61 In Greece, we now know of at least
five sure instances of it, all dating from the Bronze Age or the archaic
period. On the skull of a young girl from the Neolithic era found at <;atal
Hiiyiik (256 CH.), the diagnosis of trepanation before death is possible but
doubtful. The skull has two small round holes 6 millimeters in diameter,
penetrating the right parietal bone near bregma and separated by only 15
millimeters, which does not suggest a surgical procedure with therapeutic
objectives. According to Angel, 62 the holes were made after death. If the
procedure was performed on a living subject, it had a magical-religious
purpose, namely, ritual murder. In this connection, we should mention
the strange marks on the skull of a Bronze Age man found at Arkhanes
(Crete). The skull in question (65Ar.) features two arc-shaped bony growths
situated symmetrically on the right and left parietal bones. A. Tsouros
suggests that they are lesions from "initiatory surgery," in other words, a
violent ritual operation on the scalp which irritated the bone beneath. 63
Indeed, the nature and regular placement of these bony growths on the
skull cannot be explained by any disease. They must be a kind of scar
consequent upon an inflammation brought on by some form of torture.
For example, the application of a burning metal crown to the victim's
head would have left such marks as these.
In 1979, not far from Arkhanes, at Anemospilia, the Greek archaeologists
Efi and Yannis Sakellarakis brought to light a small temple from the Mi-
noan era (ea. 1700 B.c.). Destroyed by an earthquake and the fire that
resulted from it, the temple has remained in the state in which the catas-
trophe left it. Three persons were inside at the fatal moment. The state of
their skeletons shows that two were crushed to death by the falling walls-
the defensive position of their arms in front of their faces and a recent
fracture of the thigh are notable signs of this-while the third, a strapping
18-year-old, has had his throat cut at an altar! The earthquake had inter-
rupted the performance of a human sacrifice.64 Robert P. Charles has
found on five Argive skulls (two from the Mycenaean necropolis of Deiras
and three from geon1etric tombs in Argos) multiple circular holes of.mod-
est size drilled into the bone wall. The holes are not unlike those in the
Neolithic skull 256 from <;atal Hiiyiik. They were made with a bit-brace
on perfectly healthy bones. There is no trace of scarring or infection. So
they must be the results of post-mortem interventions that, in the absence
of medical purposes, point to the survival of a very old mortuary cult. 65
We return to trepanation on living subjects. In the classical period, the
goal of this procedure was exclusively medical and rationally conceived.
From the time of the oldest Greek surgical texts, it is a well-codified
operation, described with great precision and thorough knowledge of its

indications and complications. The Hippocratic treatise Wounds in the Head
advocates trepanation for the treatment of skull trauma. The indications
are concise: bruising and above all fracture of the skull. Cuts or depressions
in the skull do not call for it. Chapter~ 30 and 31 of the treatise describe
how to clean and scrape the wound, the technique for perforating the
bone, whether with a cylindrical saw or a boring trepan, and bandaging
after the operation to prevent purulent inflammation of the bone and
decay of the brain membranes. 66
Osteoarchaeological finds in the ancient Greek world support the hy-
pothesis according to which the origin of trepanation is not to be found
in rational medical practice. It makes its initial appearance as a ritual act
first performed upon corpses, then on the heads of living persons. Experi-
ence taught that humans could survive the opening up of their brainpan,
and a crude technique for doing so was learned, which eventually made
possible the medical use of trepanation. In the presence of magical indica-
tions-as among certain primitive peoples of the nineteenth century who
trepanned those suffering from convulsions or headaches in order to pro-
vide a way out for evil spirits-or rational ones, such as fractures of the
skull vault, they began to operate on the afflicted. In the Mycenaean age,
professional surgeons had already mastered the technical side of this oper-
ation and formulated its indications: they were on the path that led ulti-
mately to the Hippocratic codification.
A female skull found in Karata~ and dating from the Early Bronze Age
(8I ICa.) has a relatively small (10 mm.) round opening on it and no trace
of any preexisting wound. Is it a magical trepanation? Another skull from
the same necropolis (522 ICa.) shows the failure of an attempt to trepan a
terrible traumatic rupture in the temporal region. 67 The skull 33 Ler. found
by J. L. Caskey at Lerna in a Middle Bronze Age tomb is still more
informative. It belongs to a young man about 22 years old who died from
a cranial trepanation, the indications for which cannot be discovered. On
his forehead there is a large (40 mm. by 60 mm.) opening of irregular
outline with a beveled edge (at least, the outer diameter is greater than the
inner one). The part of the bone that was cut out is broken into two pieces
that remain in place. They must have stayed there after the operation,
adhering to the skin. In the case from Lerna, the operating technique
seems very crude. 68
By contrast, an examination of the skull SI Myc. evokes admiration for
the surgeon's skill. True, this time the patient was an aristocrat, perhaps a
prince, who could afford to have a good practitioner. The skeleton in
question was exhumed by G. Mylonas from Grave Circle B at Mycenae
and is the mortal remains of a man of exceptional strength who was killed
at about age 30, probably in battle; the two wounds on his frontal bone
were discussed above. In addition, on the upper left of his frontal bone,
PALEOPATHOLOGY
at a distance of less than 1 centimeter from the coronal suture, is an oval

hole (measuring 27 mm. by 30 mm.), with a clean-cut, slightly conical edge
to it (the inner diameter is greater than the outer one). Two laminae of
the outer table of the vault, which fit the opening exactly, are preserved in
place, as though they had remained attached to the scalp. The correspond-
ing laminae of the inner table and the diploe have disappeared. Two ver-
tical fracture lines start from the hole down to the lower part of the frontal
bone: probably they represent the initial lesion that, along with complica-
tions inside the skull, motivated the surgery. The bone is cut surprisingly
cleanly, as though by a machine, and the edges of the opening are perfectly
smooth. From this we can draw two conclusions. First, the surgeon must
have had considerable experience performing this kind of operation as well
as hard, well-sharpened instruments. And second, the patient did not long
survive the wound, since otherwise the reaction of the bone tissue would
have rounded off the edges. 69
Like the two just mentioned, the fourth case also comes from the Ar-
golid during the Middle Bronze Age. The Scandinavian excavations at
Asine recently brought to light the well-preserved skeleton of a man who
died around age 35, some time after a surgical operation on his head (Io7
As.). On the left mid-parietal, about 30 millimeters from the sagittal su-
ture, there is an elliptical trepanation hole (17 mm. by 20 mm.) with a
healed bevel around it. This may represent, says Angel, the surgical re-
moval of fragments after a battle-axe wound, as in SI Myc.) but with survival
of the patient. 70
The last of the five cases is a young man in his early thirties whose
skeleton was removed from a burial in Argos. Judging from its archaeolog-
ical characteristics, the burial belongs to the early geometric period (900-
850 B.C.). According to the description by R. P. Charles, a cup-shaped
indentation caused by a deep-seated focus of infection is visible over the
left eye socket. A very distinct semicircular groove runs to the left of the
indentation. There was an attempt to treat the infection by surgical inter-
vention. Plainly, the intervention was unsuccessful, since the trepanation
was never completed. Probably the cutting of the skull was abandoned
owing to the patient's demise. 71
The ccskullof Sophocles))

and the Portrait of Menander
Schliemann 's dazzling archaeological discoveries were made on the basis
of an attentive and naive reading of classical texts. Inspired by ~he success
of this method, a Danish civil servant by the name of Munter decided, in
1893, to look for the tomb of Sophocles, using as a guide his own interpre-
tation of the written traditions about the great poet's last resting place.
According to the biographical account preserved in the Vita) Sophocles
was buried in a family tomb situated along the road to Dekeleia, exactly u
stades from the walls of the city. Traditionally, this was understood to be
the walls of Athens, since the tomb would then be in the deme of Co-
lonus, his birthplace. When excavations at Colonus brought forth nothing
that could be construed as the poet's burial place, Munter thought that
the walls in question might well be the, walls of Deceleia, whose site had
just been identified as the mound now called Palaiokastro. Using his com-
pass on the military map of this region, he fixed the point II stades distant
from Palaiokastro along the old road from Deceleia to Acharnae (modern
Menidi). On that spot Munter dug up three sarcophagi, one of which,
made of marble, contained the skeleton of a male whom the joyous dis-
coverer immediately proclaimed to be Sophocles. Beside the skeleton,
there was an iron strigil, a wooden stick, and some alabaster and clay pots
datable to the fifth century B.c. 72
It turned out that the three sarcophagi at Menidi had been broken into
at a relatively early date, so that the inscriptions normally found on them
were gone. As a result, identification of the skeleton by archaeological
means was impossible. Munter's hypothesis was violently attacked by Ger-
man philologists with strong and fairly convincing arguments, but they do
not constitute an absolute refutation. 73 The skull thought to be Sophocles'
was sent to Berlin, then to the Universal Exposition in Chicago; it was
finally deposited in a Copenhagen museum. In May 1893, RudolfVirchow
made a detailed report on this anthropological specimen to the Prussian
Academy of Science. 74 According to Virchow's analysis, none of the vari-
ous characteristics of the skull from Menidi went against the identification
Munter had proposed. Sophocles died about 406-405 B.c., a very old
man, probably in his nineties, and according to a consensus of historical
witnesses, he was hale and strikingly handsome. Despite the length of his
life, the Vita has him dying by accident, choking on a grape. 75 According
to Virchow, the skull Munter found is that of a tall, vigorous old man.
However, the state of preservation of its teeth, the abrasions on them, and
the overall degree of osteoporosis do not correspond to what one usually
sees in people over the age of 90. Unfortunately, Virchow says nothing
about the rest of the skeleton. Probably it had been destroyed.
Anthropological examination of the skull could have disproved the
identification of the Menidi bones with the last remains of the tragic poet,
but it is not sufficient to confirm it. What interests us here is not so much
the historical identity of the skull as the report of certain anomalies on a
specimen that is securely dated to the fifth century B.C. This particular
skull is markedly asymmetrical because the left rear section of it has been
flattened (technical term: plagiocephaly). The deformation is due not to
the skull's senile involution but instead to a pathological process in child-
hood that began with the complete union of the suture that joins the left
parietal bone to the temporal bone. According to Virchow, the union
must have been caused by an accidental trauma, such as a blow to the head
PALEOPATHOLOGY
or a fall, and not by an artificial deformation of the skull. Since the synos-
tosis hindered growth in one direction, the skull developed, compensato-
rily, in other directions. The asymmetry of the bone structures brought
about the displacement of the brain and an asymmetry in its hemispheres:
the right half must have been relatively larger than the left.
"According to current conceptions," wrote Virchow in his 1893 report
(p. 694), "one can deduce from it a predisposition to criminal activity,
though older pathologists would instead have associated it with the eccen-
tricities of a poet or a utopian." Neither opinion is acceptable to modern
anthropology. There is renewed interest in such questions from the per-
spective of neurophysiological experimentation with bisected brains. The
experiments of Sperry and Bogen show that the left hemisphere predomi-
nates in verbal activities, arithmetic operations, and literary expression,
while the right is better at realizing nonverbal thought and the perception
of shapes. 76 But the effect of pronounced cerebral asymmetry on intellec-
tual and psychosocial ·aptitudes is still unclear. The results of Sperry's
experiments cannot serve to identify the man whose skull was found at
Menidi. If it could, the predominance of the right hemisphere would
suggest a Phidias rather than a Sophocles.
No literary source speaks of Sophocles' plagiocephaly. The comparison
of the Menidi skull with ancient busts of the poet is not at all favorable to
the identification, but the argument is not decisive, since hair can mask
asymmetry of the skull vault. Only copies of the idealized portrait of
Sophocles have come down to us. I mention in passing what Plutarch tells
us about the head of Pericles (ea. 495-429 B.c.):
Agariste once had a dream that she had given birth to a lion, and a few days later
she was delivered of Pericles. His physical features were almost perfect, the only
exception being his head, which was rather long and out of proportion. For this
reason almost all his portraits show him wearing a helmet, since the artists appar-
ently did not wish to taunt him with this deformity. However, the comic poets of
Athens nicknamed him "schinocephalus" or "squill-head. " 77
Ancient sculptures can be informative about some pathological character-
istics of historical personages. It is also true that the blindness of Homer,
the snub nose of Socrates, and the hunchback of Aesop are cliches whose
sculptural or pictorial expression stems from legend, not clinical reality.
Perhaps that is not the case for the facial asymmetry that can be seen on
a very lovely Hellenistic head in marble, now in the University Museum
in Philadelphia. After a careful morphological examination of this sculp-
ture, the American neurosurgeon Temple Fay asserts that it faithfully rep-
resents a kind of facial asymmetry whose pathological character is beyond
doubt and whose subtlest details correspond to a precise nosological cate-
gory. According to this doctor, the late-thirties man represented in the
Philadelphia sculpture suffered, sometime before the tenth year of his life,
a lesion to the fronto-parietal region of the left hemisphere of the brain
that resulted first in spastic hemiplegia ("cerebral palsy") and in underdev-

elopment of the right side of the face and neck. 78 Although this childhood
accident may have had no negative effect on his intellectual gifts, its con-
sequences were not confined to an asymmetry in the physical development
of the body. If Fay's diagnosis is co~rect, this handsome man with a
melancholy look suffered from disturbance of his motor functions and was
even, perhaps, epileptic.
One must be very cautious in proposing a clinical interpretation in such
cases. From a recent study that reviewed substantial archaeological material
and made rigorous use of the quantification of morphological parameters,
we learn that asymmetry of the face and the brainpan in Greek statues and
bas-reliefs from the fifth to the third centuries B.C. occurs with a regularity
that clearly surpasses normal anatomical variation. 79 These are representa-
tions of gods, anonymous warriors, idealized and unidentifiable indivi-
duals. According to L. A. Schneider, there is a correspondence benveen
facial asymmetry and the position of the head, either in relation to the
activity being expressed by the body as a whole, or in relation to the
position of the neck. This scholar has not extended the research to por-
traits of historical persons, but from now on one should take account of
the fact that, from the fifth century B.C. on, exaggerated asymmetry of the
face is a technical tool, a craftsman's trick intended to enhance the aes-
thetic effect of a work of art.
The Philadelphia head is not the bust of a man standing still in a con-
ventional attitude, looking straight ahead. It is a fragment of a statue. The
head is bent on the neck and turned toward the right. So a certain asym-
metry in the face would be completely within the rules of art. The Phila-
delphia statue is the copy of a lost Greek prototype. Several other ancient
copies of the same original have survived (in Boston, Dumbarton Oaks,
Venice, Copenhagen, and elsewhere). All have clear facial asymmetry.
According to a consensus of art historians, the statue is the realistic portrait
of a historical personage. The prevailing view is that it represents the
Athenian poet Menander (about 342-291 B.c.), the leading figure in Greek
New Comedy. The identification is very probable but not absolutely cer-
tain. 80 It was recently corroborated by the discovery' of some third-century
A.D. mosaics found in the Khorapha quarter of Mytilene: they contain ten
scenes relating to the plays of Menander and a portrait of the poet labeled
with his name. The technique of this portrait is a little crude, almost like
a caricature, but even so one can see that the figure has crossed eyes and a
facial asymmetry that are not the result of an artist's botch. Here, then, is
an argument in favor of the "verism" of the Philadelphia head's asymme-
try, that it is the reflection of a real pathological phenomenon and not a
clever sculptor's trick. Still, one cannot help mentioning that the under-
developed part of the face is on the right in the sculptures but on the left
PALEOPATHOLOGY
in the mosaic. i:,erhaps that is a consequence of the way paintings and

mosaics were copied.
Written sources are very chary of details on the private life and bodily
characteristics of Menander. According to the Suda)he was cross-eyed, but
his spirit had piercing vision. If Alciphron can be believed in one of his
letters, Menander complains of poor health. According to an ambiguous
discussion in Phaedrus, he had an effete and langorous gait. He died at the
age of 52, probably trying to swim the harbor of the Piraeus. These few
biographical details are favorable to a diagnosis of infantile spastic paralysis,
but they are not a straightforward confirmation of it. Other events in
Menander's life, particularly his service as an ephebe and his success as a
lover, do not easily accord with the medical label of a handicapped person
and certainly exclude any serious or even moderate form of chronic motor
disability. 81
According to Jean Dastugue (1974), an adult atlas (first cervical vertebra)
found in a Neolithic cave near Laurion (2!337I(itsos) has an asymmetrical
posterior arch probably due to a difference in the diameter of the vertebral
arteries. If this etiology for the bone malformation in question is correct,
it is reasonable to suppose that the person in question suffered from cere-
bral hemiatrophy.
CongenitalMalformations) TumorsJand MetabolicDisorders

Anomalies of the spinal column, especially with regard to the number
of vertebrae and their fusion, are frequent among contemporary inhabi-
tants of Europe and Asia. The lumbosacral region is the part of the spine
n1ost subject to variations. This same genetic instability is very pronounced
in the ancient skeletons thus far exhumed in Greece. Of particularly high
frequency is the six-segment sacrum, which is chiefly due to the sacral
incorporation of the fifth lumbar vertebra. 82 There is a genetic tendency
for the pelvis to ascend and the lumbar curvature to lessen. Actually, these
phonomena are not properly a part of my inquiry, since they lack direct
pathological significance. However, certain vertebral fusions, especially in
the neck area, can give rise to neurological problems, and anomalies at the
lumbosacral joint can be a predisposition to lumbago and arthritic break-
down. As an example, I can cite the case of a Mycenaean aristocrat (58
M ye.) whose fifth lumbar vertebra, by a congenital anomaly, was incorpo-
rated into the sacrum and who had, at age 35, traces of degenerative arthri-
tis on the other lumbar vertebrae. 83
Another example deserves mention here even though it belongs to a
relatively late period. While excavating a Roman camp at Corinth in 1960,
H. S. Robinson discovered a tomb containing the well-preserved skeletons
of two persons buried at the same time, one next to the other. The
Swedish experts N. G. Gejvall and F. Henschen state that they are the
remains of an adult female and an adult male, both tall (181and 158.5 cm.)
and both afflicted with congenital spinal malformations. In the man, the
first cervical vertebra is fused with the base of the skull, and the second is
squat and has a very short denticulate apophysis. In the woman, the first
cervical vertebra is not completely formed (it consists of the two lateral
halves kept together by a pseudo-arthrosis), and the second is misshapen,
shortened, and fused in a block with the following two cervical vertebrae;
there is also a block fusion of the fifth and sixth cervical vertebrae and the
first, second, and third thoracic as well as a partial fusion on just one side
of the fifth lumbar vertebra with the sacrum. 84 Occipitalization of the first
cervical vertebra is a rare anomaly. In the male skeleton from Corinth it
surely is of genetic origin, as is the whole series of spinal malformations on
the female skeleton from the same tomb. Very likely the two were linked
by blood, not marriage. In both, the bone anomalies are so serious that
they must have been accompanied by significant neurological problems.
To be sure, parallelism in suffering does not explain their simultaneous
demise, unless we suppose it was a double suicide. For the record, a case
of spina bifida has recently been discovered on the skeleton of a young
woman from Tiryns (geometric period, ea. 900-700 B.C.). 85
In 1912, the French physicians Maurice Klippel and Andre Feil described
for the first time in a precise and detailed way a case of congenital absence
of the cervical vertebrae resulting in the ascent of the thoracic cage to the
base of the skull. This same anomaly, the Klippel-Feil syndrome, is repre-
sented with stark realism in a Hellenistic figurine from Smyrna. A coro-
plastic artist gave the clay the shape of a little fellow without a neck (aplasia
and fusion of the cervical vertebrae) whose face expresses suffering. The
diagnosis is secured by the symmetrical position of the head (which ex-
cludes a stiff neck), atrophy of the mastoid region, and the presence of a
pterygium colli. Often deafness and mental retardation are associated with
the Klippel-Feil syndrome. The cervical malformations compress the spinal
nerves and cause painful disorders. 86
On a terra cotta head from Corinth that dates from the middle of the
fourth century B.C., a Greek artisan has rendered with the precision of a
scientific observer all the morphological characteristics of a unilateral cleft
lip (cheiloschisis) as well as the secondary consequences of this congenital
malformation for the overall st1ucture of the face. 87
On two female skeletons from the Late Bronze Age excavated in the
Pylos region, Christos C. Bartsocas has good but not definitely convincing
reasons to recognize two specific bone ailments, Pager's disease on one
and cleidocranial dysplasia on the other. The first (osteitis deformans,
identified by Paget in 1877) causes remodeling of bones and deformities in
old people; it attacks several bones and progresses slowly. Heredity seems
to play a significant role, but the etiology of the disease is unknown. The
PALEOPATHOLOGY 71
second disease is a congenital disorder in skeletal development causing

defective formation of the skull, the collarbones, and the vertebral arches.
The same scholar has diagnosed this disease in an ancient iconographic
representation ofThersites, the deformed anti-hero of the Homeric epic. 88
A 35-year-old male whose skeleton was found at Argos in an old geo-
metric tomb (16 A13.)· ea. 900-850 B.c.) suffered from congenital disloca-
tion of the hip due to a malformation of the left hip socket, which is open
too wide. With the head of the femur out of place, excessive pulling on
the ligaments of the joint resulted, which then caused the production of
osteophytes on the vertebrae, the trochanter, and the hipbone. According
to the osteoarchaeological description of Robert P. Charles, the location
and quantity of these bony proliferations suggests an ankylosis of the left
hip as the final complication of an apparently minor congenital malforma-
tion. 89 Another case of congenital hip dislocation with less serious conse-
quences has been reported by John Lawrence Angel on a Greek skeleton
of the Mycenaean age. 90
Angel has also described cases of clubfoot on two skeletons from the
necropolis of Lerna. The first is an unusual calcaneo-navicular joint, prob-
ably a talipes valgus, in the left foot of a boy about 15years old (122 Ler.).
The femur on that side is thinner than its mate on the other side. Is it a
congenital clubfoot or one caused by an acquired pathological state? The
paleopathologist cannot decide one way or the other for this case or a
second one: the disability in the right foot of a 45-year-old man (127 Ler.)
is associated with a noticeable shortening of the right lower limb. The talus
is rotated laterally and downward, which completely unbalances the ankle
joint. The navicular, the cuneiforms, and the metatarsals are underdevel-
oped, while the calcaneus seems to have been fractured. It is possible that
this abnormal state results from a traumatic dislocation with fracture dur-
ing adolescence, but it could just as well originate in a congenital
malformation. 91
The Hippocratic physicians were perfectly well acquainted with congen-
ital hip dislocation as well as the principal forms of clubfoot. According to
them, congenital dislocation of the hip is not inherited but is rather the
result of a lesion arising during embryonic development. They accounted
for permanent deformities of the foot and ankle in the same way, evoking
mechanical factors like intrauterine compression or infantile trauma, not
knowing of their genetic determination and paralytic etiology. 92
It is practically certain that since his origin man has been subject to
disturbances in the biological controls on cell growth that are manifested
clinically as malignant tumors, leukemias, and the like. However, the
frequency of these ailments has not necessarily been the same in different
eras and throughout the range of geographic and sociocultural environ-
ments. Cancer, in the widest sense of the term, is a common disease for
modern man: in developed countries it is now the cause of death for two
72 DISEASES IN THE ANCIENT GREEIC WORLD
out of ten persons. The growth of such neoplasms often leaves traces on
skeletons, either when a tumor arises from the bony tissue itself (e.g. an
osteosarcoma), or when bones are attacked by the expansion and spread
of a malignant soft tissue proliferation. In modern man there are metasta-
ses into bone in at least one-quarter of~ cancer victims. It would suffice
to examine fewer than 100 skeletons from a twentieth-century cemetery to
be almost sure of finding a case of cancerous disease, but the results of
osteoarchaeological investigations in no way match these statistics. Though
it is true that the mark of cancer can be observed on human bones from
any and every age, the fact remains that its presence is exceptional in
specimens before the Renaissance. For the epochs stretching from prehis-
toric times up to the sixteenth century, its frequency is difficult to quantify
but certainly is a lot less than I case for every 1,000. 93
To our knowledge, no certain case of a malignant tumor has yet been
identified on ancient Greek bones. Angel mentions a probable case of
cancer metastases in one Late Bronze Age skeleton (unpublished). As for
neigh boring areas, an Etruscan example of malignant cranial metastasis has
been reported. 94 Aside from a few instances of benign tum ors, notably two
osteomas of the femur in Karata~, the osteoarchaeological evidence for
neoplastic diseases in Greek prehistory and history is extremely poor. One
should not draw hasty conclusions from this situation. Some works of art
appear to show that the Greeks knew of and portrayed both breast cancer
and sarcoma of the eye-socket, but such diagnoses from medical archae-
ology are not unimpeachable. 95 On the other hand, literary sources offer
abundant, reliable information on the existence of various malignant tu-
mors in the classical period. However, their frequency was so low that the
absence of traces of them in the osteoarchaeological record is not surpris-
ing. Considering the total number of skeletons that have been examined,
the current negative result does not allow us to state positively, for exam-
ple, that cancer was rarer in Greece than in certain geographic and cultural
regions for which there are positive paleopathological finds (prehistoric
Iran and northern Europe, Pharaonic Egypt, and pre-Columbian Amer-
ica). The infrequency of cancerous disease in ancient populations as com-
pared with our contemporary societies is partly explicable in terms of
differences in average life expectancy, chemical pollution, and the quantity
and nature of some particular radiations. All these circumstances are un-
deniably significant, but it does not appear that they adequately account
for the magnitude of the change in frequency from then to now. Whatever
the solution, genetic factors seem less to blame than environmental ones.
Among the "inborn errors of metabolism" (A. Garrod, 1909) that leave
their marks on bones, the most important in the eastern Mediterranean
are hereditary anemias and gout. The former raise such complex problems
that I must devote a separate chapter to them. As for gout, osteoarchaeo-
PALEOPATHOLOGY 73
logical reports and literary texts both attest to its antiquity in the Greek
world. A 35-year-old man whose skeleton was found at Lerna in a Middle
Bronze Age tomb (70 Ler.) must have suffered from it in his big toe, the
classic locus of urate deposits. The surfaces at that joint of his right foot
are markedly arthritic, and an interphalangeal enlargement of the space
between it and the next toe corresponds to the place taken, during the
man's lifetime, by a tophus, an accumulation of urate crystals. Although
the presence of arthritic exostoses on the rear part of this foot might raise
doubts about the diagnosis of gout, it is confirmed by the presence of a
yellow-purple deposit on the same man's left elbow. 96 An affliction com-
mon in the experience of the Hippocratic writers, 97 gout is cited by Plato
along with fever and ophthalmia as a typical disease. 98 Yet it seems to have
been unknown to the physicians of Pharaonic Egypt. No mention is made
of it on papyri, and physical signs of it have not been detected on mum-
mies of the high period. As long ago as 1910, Elliot Smith and Wood Jones
described a spectacular case of primary generalized gout with subcutaneous
tophi and incrustations within the joints on a mummy found near the
temple at Philae (one urate deposit measures 5 by 10 by 23 mm.), but the
case is of relatively recent date: the gouty old man in question was a
Christian living under the Roman occupation. 99
Another case of gout from Roman times (second century A.D.) was
diagnosed by Calvin Wells on a skeleton from the necropolis of Cirencester
in Gloucestershire. As is to be expected, it is also a man-gout is much
more common in men than in women-of mature age with gout in several
joints and, interestingly enough, a relatively high social status: of the 268
people in this Roman British cemetery, he and another individual are the
only ones to be buried in sarcophagi. 100 Without wanting to take up at
this point the difficult question of the reasons for the spread of gout in the
Roman Empire and for its outbreaks during certain historical periods, I
note that there is paleopathological confirmation of the antiquity of the
gene in just those regions in which this disease will plague the ruling classes
(Byzantium, Christian communities in Africa, and Britain). 101
According to modern medical research, the metabolic processes leading
to an excess of uric acid and its deposit in tissue are basically dependent on
genetic factors. By contrast, environmental factors are responsible for var-
ious concretions that are deposited in the urinary tract. The climate of
coastal Greece and certain dietary habits make probable on a priori grounds
a high incidence of urinary stones. In fact the exceptionally rich historical
documentation of bladder stones in Greece, in the classical age as well as
the Byzantine period and on into the nineteenth century, confirms their
commonness and especially their frequency in children. Paleopathological
methods allow the identification of urinary calculi among the prehistoric
human remains of Egypt and western Europe. 102 A kidney stone from the
Bronze Age was found in Hungary, at the limit of the Greek world, but
no paleopathological example from Greece itself has yet come to light. 103
No doubt the cause of this lack of evidence is the difficulty of finding such
sn1all stones after bodily decomposition.
Specialists in the prehistory of diseas~ in Greece have been more fortu-
nate with gallstones. In 1954 J. L. Angel found several in one of the tombs
of Grave Circle B in Mycenae (Middle Bronze Age). These reddish brown
aggregates, with green patches, several facets, and rounded edges, were
found between the lowest right ribs and the pelvic brim of the skeleton of
a massive man around 55 years old (I3I Myc.). 104
In a man of the classical period (skeleton 65 AI(.) found in the Keramei-
kos in Athens and dated around 450 B.c.) about 40 years of age, osteoar-
chaeological examination has revealed a strange condition of the larynx. It
had become rigid, almost ossified, as the result of calcification of the
thyroid and arytenoid cartilage. 105 The same skeleton shows spinal hyper-
ostosis (with ankylosis of several thoracic vertebrae even to the point of
their fusion with certain ribs) and scattered osteophytosis, which suggest
generalized metabolic disease. This individual also suffered from a large
tooth socket abscess and gum disease. Without X-rays and histological
examination, the diagnosis of systemic disease in this citizen of Periclean
Athens cannot be specified. Nor can the medical historian add anything
to the laconic report in Herodotus concerning the bone anomalies ob-
served during the creation of an ossuary for the victims of the battle of
Plataea in Boeotia (victory of the Greeks over the Persians in 479 B.c.):
''When the corpses were fleshless, they discovered-this was after the Pla-
taeans put all the bones in one place-a skull without any sutures, a single,
continuous bone; also a jawbone, the upper jaw, with all the teeth at-
tached to one another, both the front teeth and the molars all made of
one bone; and also the bones of a man five cubits tall. " 106 What were the
shape and dimensions of this skull "without any sutures"? Since it be-
longed to a warrior, who must have been an adult male without handicap,
the skull could not have been fused into a single bone since birth. Hero-
dotus's summary description, however, is insufficient for a retrospective
diagnosis. As for the strange "dental ankylosis,", we can mention two
famous historical parallels: it is said that Pyrrhus, king of Epirus (ea. 319-
272 B.c.), and Prusias Monodus, son of King Prusias of Bythinia (second
century B.c.), had single, continuous bones by way of teeth for their upper
jaws. These are probably extremely rare forms of dental agenesia, a specific
recessive genetic disorder. The pathological state has only recently been
recognized. 107
As for the giant, his size is given in round numbers. Five cubits corre-
sponds roughly to 7.5 feet (2.3 meters). For a learned 1nan in ancient times
it was not easy to determine the exact height of a man using bones as the
only clue. When discussing another giant from Persia, a military engineer
PALEOPATHOLOGY 75
named Artakhaies,, who "fell ill and died," Herodotus is more precise: he
"measured six royal cubits less four fingers," that is, about 8 feet 2 inches
(2.55 meters) . 108 Relatively proportional gigantism can be a matter of con-
stitution with no accompanying metabolic anomalies. Often, especially in
extreme cases like those mentioned by Herodotus, it is due to endocrine
disorders, in particular to a hyperactive adenohypophysis before the phys-
iological end of growth. An excess of growth hormone after puberty results
in acromegaly. The tokens of this disease (namely, excessive growth of
extremities, facial peculiarities like a projecting jaw and an oversized nose,
lateral and convex spinal curvature, etc.) are visible in several Hellenistic
statuettes. 109 Finally, I mention only in passing the iconographic represen-
tation in antiquity of several other syndromes due to chromosomal anom-
alies or endocrine disorders: achondroplastic dwarfism, Down's syndrome,
hermaphroditism, hydrocephalus, and Cushing's syndrome. 110
Vitamin Deficiencies
From what is known of the diet and climatic conditions in ancient
Greece, it is fair to posit the existence of vitamin A deficiency (which causes
night blindness and xerophthalmia), the rare and sporadic appearance of
vitamin B2 deficiency, osteomalacia, rickets, and scurvy, and, finally, the
absence or extreme rarity of beriberi and pellagra. Only rickets, osteoma-
lacia, and scurvy (two forms of vitamin D and vitamin C deficiency) affect
the bone system. No other vitamin deficiency can be identified by osteoar-
chaeological methods.
Rickets is a systemic disease of bones and cartilage in infants and chil-
dren. By far the most common clinical form of the disease is associated
with a lack of vitamin D 2 ( ergocalciferol) and insufficient exposure to sun-
light, for solar radiation is indispensable to the synthesis in the skin of
vitamin D 2 from previtamins that are not often lacking in the food chain.
It is easy to understand why fog, heavy clothing, and staying indoors
continually are, along with a restricted diet, factors associated with the
development of rickets. The disease's main characteristics are nonmineral-
ized growth and bone deformities. Its diagnosis can be made as easily on
the skeletons of children as on those of adults, thanks to its permanent
consequences. If the vitamin D 2 deficiency arises after growth ceases,. that
is, in adulthood, the result is a bone disease called osteomalacia. Unlike
rickets, it only occurs under conditions of extreme deprivation of vitamin
sources, general malnutrition, and heavy drain on the body's calcium
resources (particularly in pregnant women). That is why, as a rule, osteo-
malacia is much less common than the childhood form of vitamin D
deficiency.
In view of what has just been said about the etiology of rickets, it is not
surprising to learn that almost all cases of the disease observed up to now
on prehistoric, protohistoric, or medieval bones come from the misty

regions of northern Europe and from Siberia. A few fairly dubious identi-
fications have been put forth for pre-Columbian peoples in the Americas,
but all the osteoarchaeological documentation now extant proves that
rickets only became a relatively common ailment along with the urbaniza-
tion of Europe after the Middle Ages. i 11 There is no rickets on early Greek
and Egyptian bones. Harris' lines and striations on dental enamel are
insufficient to sustain a diagnosis of this disease. Nor is the existence of
hunchbacks, which has been confirmed by osteoarchaeology, iconogra-
phy, and literary texts, proof of the antiquity of the disease.
A possible but highly problematic instance of rickets is the case of a
young girl from Centuripe in Sicily (first century B.C.). 112 A thickened
frontal bone and a curved tibia are indeed suggestive of the disease, but
the skeleton in question, which is now housed at the Archaeological Mu-
seum in Syracuse, should be reexamined by a pathologist well versed in
modern paleopathological technique. Two cases have been identified by
Guyla Regoly-Merei on skeletons from the Roman era found at Fazekas-
boda near Pees (in ancient times Sopianae, capital of Lower Pannonia).
The poor wretches were undoubtedly natives of the region, not Roman
soldiers or colonists. 113 Aside from instances of hunchback, whose connec-
tion with vitamin D deficiency is far from unequivocal, the marks of rickets
are missing from archaic and classical iconography. But some have thought
they recognized the disease on certain Hellenistic terra cottas found in Asia
Minor and Italy . 114
The Hippocratic writings contain no allusion whatever to pathological
phenomena that could be interpreted as rickets. And a decisive argument
for the rarity of vitamin D deficiency in classical Greece consists of the
complete absence in gynecological treatises of reflections on pelvic defor-
mations· and their obstetrical significance. The flat pelvis was unknown to
Hippocratic physicians, nor is it mentioned in the medical texts before
Soran us. Osteoarchaeology confirms the rarity of pelvic anomalies and the
relative broadness of the pelvis in Greek females. Not until Roman times
did urban density, changes of diet, and the impoverishment of one part of
the population unite to bring about conditions favorable to rickets and to
making its ravages apparent to medical practitioners. 115
Did endemic bone dystrophy exist in the classical period among certain
neighbors of the Greeks, chiefly the Persians, as Herodotus seems to sug-
gest? On a visit to Pelusium, a harbor on the eastern branch of the Nile
and the site of Cambyses' great victory over the Egyptian army in 525 B.C.,
Herodotus tells us he found the following:
At the place where the battle was fought I saw a very odd thing, which the natives
had told me about. The bones still lay there, those of the Persian dead separate
from those of the Egyptian, just as they were originally divided, and I noticed that
the skulls of the Persians are so thin that the merest touch with a pebble will pierce
PALEOPATHOLOGY 77
them, but those of the Egyptians, on the other hand, are so tough that it is hardly
possible to break them with a blow from a stone. I was told, very credibly, that
the reason was that the Egyptians shave their heads from childhood, so that the
bone of the skull is indurated by the action of the sun-this is also why they hardly
ever go bald, baldness being rarer in Egypt than anywhere else. This, then, explains
the thickness of their skulls; and the thinness of the Persians' skulls rests upon a
similar principle: namely that they have always worn felt skull-caps to guard their
heads from the sun. I also observed the same thing at Papremis, where the Persians
serving under Achaemenes, the son of Darius, were destroyed by Inarus the
Libyan. 116
Travelers reaching Mesopotamia would have encountered cases of

scurvy, a deficiency disease marked by serious hemorrhaging and disorders
of the bones and joints. But Greek physicians did not know of it. 117 Even
during famines, it was never vegetable products containing ascorbic acid
(vitamin C) that were lacking to the inhabitants of Greece and Italy, and
sea travel was still just coastal navigation. The only occasion on which
scurvy could have had a significant impact was the prolonged siege of a
city, but historical accounts of the subject are lacking. 118 It may well be
that scurvy appeared sporadically, for instance, among people who were
bedridden or in prison, even in a society in which the necessary supply of
vitamins was abundant. We cannot exclude some incidence of infantile
scurvy or Barlow's disease, which seriously interferes with bone growth.
Unfortunately, diseases of the newborn were not well observed or differ-
entiated by ancient physicians. Their pathology is historically invisible. In
any case, no osteoarchaeological case of scurvy is known until the Middle
Ages. 119
Rheumatic Ailments
Traces of "rheumatism" are very common on prehistoric and protohis-
toric human bones, and their diagnosis is the most common in osteoar-
chaeology .120 Nevertheless, I take the precaution of putting the name of
this disease in quotes, since the ancient concept of rheumatism has sur-
vived in modern medicine only at the cost of its pathological coherence.
It broke up to form a host of various osteoarthropathies. On a dry bone,
the morphology of rheumatic lesions is relatively uniform-rough joint
surfaces with local compressions and erosions, bony excrescences on the
edge of these surfaces-but that does not mean that their etiology is not
complex, very diverse (wear and tear, traumas, infections, metabolic dis-
orders, autoimmune reactions, genetic defects with delayed aftereffects,
etc.), and, in general, still inadequately understood. 121 So it was to be
expected that such lesions would occur on bones from Greece and that we
would have to confront the difficult problems of differential diagnosis that
they present.
Degenerative joint disease (degenerative arthritis or, better, osteoarthri-
tis) 122 is the most common of all the rheumatic ailments that lend them-
selves to osteoarchaeological diagnosis. It manifests itself in living subjects
as a progressive deterioration of articular cartilage that is, on the one hand,
either a secondary phenomenon resulting from a malformation, a trauma,
an intra-articular deposit, or an attack of fever, or, on the other, a primary
one, without some other pathological state being its "cause." Bony le-
sions are associated with the erosion and ulceration of the cartilage: under-
lying bone tissue thickens in certain places (zones of osteosclerosis) and
makes fissures in others; at the edge of the transformed joint, bony pro-
tuberances called osteophytes begin to grow. The synovial membrane is
either normal or slightly inflamed. In modern man, primary degenerative
joint disease is seen chiefly after age 50, and its frequency increases with
age. Its basic pathogenetic process is wear and tear on joints linked with
mechanical stress and changes due to aging. 123 The disease is often confined
to a single joint or acts in bilateral symmetry. It attacks the hip, the knee,
the shoulder, the wrist, the fingers, the joint of the jaw, and frequently
the spine. Its localization on the spine can be either anterior or posterior.
Degenerative joint disease of the limbs is often disabling, while in the spine
it is usually well tolerated.
Osteophytes grow as the result of ossification under the periosteum or
within the bone. Their presence at the margin of the articular cartilage
together with marginal lipping reinforces a diagnosis of degenerative joint
disease. However, their number or size is not an indicator of the serious-
ness of the degenerative process. The etiology of osteophytes is multiple,
so that it is not easy to detect, especially on a dry bone that is not well
preserved, the pathological event that is at the origin of an isolated growth
of them around a joint without accompanying joint lesions. Divergences
in the statistical data of different paleopathologists could be due to differ-
ences in their diagnostic evaluation of isolated osteophytes.
There are good reasons to think that at the dawn of mankind the food
consumed called for considerable feats of chewing. Two consequences of
such a functional strain are especially common on ancient human remains:
dental abrasion and wear on the joint that is involved in chewing. The
presence of degenerative osteoarthritis in the teniporo-mandibular joint
has been reported on several Neanderthals (Krapina, La Chapelle-aux-
Saints, La Ferrassie, etc.), on some specimens from Taforalt, an epi-
Paleolithic necropolis in western Morocco, and in numerous Neolithic
specimens from central and western Europe. According to W. M. Krog-
man, 9 of no mandibles from Tepe Hissar, an Iranian necropolis dated
4000-2000 B.C., have sure signs of degenerative joint disease on their
condyles. 124 That amounts to a little more than 8 percent; the real percent-
age should be higher, since dubious or mild cases were not taken into
account.
There should be no shortage of similar signs on prehistoric jawbones in
PALEOPATHOLOGY 79
the eastern Mediterranean. Unfortunately, currently available information
provides only a sketchy picture of the frequency of this disorder. These
modifications of maxillary condyles and their sockets in the mandible are
relatively slight and easily go unnoticed. Angel has reported osteoarthritis
of the temporo-mandibular joint in n of 43 individuals at ~atal Hiiyiik
(Neolithic), which is 25.5 percent of those examined (31.5 percent among
men and 21 percent among women). 125 A systematic study of prehistoric
jaws in Anatolia has been undertaken by B. Alpagut, a Turkish researcher,
who, in a preliminary survey, was able to diagnose one serious case of
bilateral temporo-mandibular degenerative joint disease (adult male of the
Early Bronze Age) and three cases in which the condyle is affected unilat-
erally (adults, Neolithic and Chalcolithic). 126 For the Peloponnesus, I can
cite, as an example, a very pronounced case in a pre_historic inhabitant of
Lerna (I25 Ler.), who also suffered from periodontal disease on the same
side. 127 For Attica, I can report an adult lower jaw from the Kitsos Cave
(2/5I8 J(i.) that has signs of degeneration of the joint on the left condyle,
probably associated with bite disorders resulting from a major loss of
molars. 128
Degenerative joint disease in the limbs is not unusual in ancient bones
from Europe. Its role in Greek osteoarchaeology has been stressed by
C. M. Furst (with special reference to specimens from the Middle Bronze
Age found at Asine) 129 and studied in detail by J. L. Angel. The latter
found that on noninfantile bones from the Helladic necropolis at Lerna,
osteoarthritis of the limb joints affected 41 percent of the men and 18
percent of the women. 130 During the classical, Hellenistic, and Roman
periods these frequencies diminish by about half. When degenerative joint
disease affects the joint of a limb, it is often associated with a malformation
or a fracture that has healed in an abnormal position. Their association on
ancient skeletons makes it possible to establish, in some cases, the causal
links that result in a joint's degeneration. Degeneration can also result
from especially harsh and repetitive physical labor. The osteoarchaeological
proof of such labor is sometimes provided by the increase in bone ridges
to which muscles are attached. As an example, there is the female skeleton
137Ler.)which shows, on the one hand, scars in the form of an enlargement
of the scalene muscle insertions on the first right rib, and, on the other,
traces of degenerative joint disease on the surfaces of the right scapulo-
humerus joint and along the biceps groove of the right humerus. It is
probably a case of occupational shoulder osteoarthritis in a weaver. 131
As against the observations of J. L. Angel, Robert P. Charles has only
rarely been able to report the presence of degenerative joint disease on
ancient skeletons from the Argolid. According to him, although signs of
rheumatism are usually very common on the prehistoric bones of western
Europe, he has seen only three cases of it in Argos: some "rheumatic
exostoses" (that is, osteophytosis) on the clavicle of 93 A13. (from a pro-
togeometric tomb), on the humerus of one of the specimens from tomb

IS A,;q.) and on the vertebrae as well as at the level of the left hip joint on
skeleton I6 A,;q. (a man with congenital hip dislocation; see above, p. 71).
A new investigation was able to add only one more case: XXXVI-I Deiras)
the remains of a woman from the Myce~aean era, aged about 35, with mild
spinal osteophytosis. 132
Without doubt, the most common disease diagnosed by paleopatholo-
gists is degenerative osteoarthritis of the spine. 133 In the great majority of
cases, one should not even call it a disease in the usual sense of the word,
since the bone changes in question, although they are morphologically
impressive and plainly pathological, do not result in serious functional
disorders and are generally not experienced by the person affected as a
sickness. In some cases, vertebral osteophytes can become so important
that they form massive bony bridges across intervertebral discs and make
some portions of the spine rigid. Ankylosing vertebral hyperostosis, whose
anatomical and clinical profile has been clarified by Jacques Forestier, is a
severe form of osteophytosis. It is marked by vertical ossifications between
the thoracic and lumbar vertebrae that have the appearance of dripping
wax. Despite its impressive anatomical modifications, this disease is usually
without major clinical symptoms. 134
Paleopathological "rheumatic bone growths" were first discovered in
the prehistoric bear (F.J.K. Mayer, 1854). At first their etiology was thought
to be the cold and dampness of caves, whence the name "cave gout"
coined by Virchow (1895). The hypothesis seemed confirmed by the de-
scription of similar lesions on the vertebrae of Neanderthal men, specifi-
cally Krapina man (D. Gorjanovic-Kramberger, 1906) and the man from
Chapelle-aux-Saints (M. Boule, 1911). But evidence began accumulating
that this disease had also been common among the Neolithic populations
of Europe, who were relatively well protected from the elements, and in
particular among the inhabitants of Pharaonic, Hellenistic, and Roman
Egypt. To the pioneers of paleopathology, nothing could be more aston-
ishing than widespread endemic rheumatism in a dry, hot climate ..To save
the prevailing hypothesis, Wood Jones did not scruple to invoke, as an
etiological factor, "dabbling in the water of the Nile." Then, for a while,
the trend was to explain vertebral osteophytosis as a reaction to a primary
infectious focus. A causal link was suggested between tonsillitis or perio-
dontitis, everyday diseases in all epochs, and proliferating rheumatism of
the spine. In the middle of this century, osteoarchaeology finally drew its
lesson from the fundamental distinction between acute articular rheuma-
tism and degenerative joint disease, and it realized that the common form
of spondylitis was basically independent of climate and of microbes.
In a study of the biological history of the spinal column, 135 I insisted on
the fact that, in man, it has obvious weaknesses. I attempted to account
for them as a result of its partially contradictory functions and of the
PALEOPATHOLOGY 81
adaptation, perforce inadequate, of a structure initially designed for an-

other role. The' present form of the spine is the result of a compromise
among several functions that it has to fill, each of which has a shaping
influence that either complements or competes with the others. On the
one hand, the spine must be both an endoskeleton that serves as a me-
chanical support for the body's system of locomotion and also an exoskel-
eton that must shelter and guard the central nervous system. On the other
hand, the spine is shaped from the beginning like a mechanical support for
a marine animal of soft consistency; stretching out into a horizontal posi-
tion, it developed like the hydrodynamic axis of animals that swim, only
to adapt itself thereafter to locomotion on dry land (first crawling, then
walking on all fours, then walking erect). This series of readaptations is the
source of certain spinal ailments, notably degenerative spondylitis. Un-
known among most animals, this diseas~ appears among the larger saurians
of the Cretaceous and among the crocodilians, then in the cave bear and
prehistoric man. In the former, it is the price of their size and their passage
from water to earth. In the latter, it is the price paid for the vertical
orientation of a relatively heavy trunk. Just as that orientation creates
lowered resistance in the lumbosacral region, the development of the hand,
cephalic flexure, and cephalization define another in the cervical region.
Moreover, degenerative diseases of this kind are promoted by the fact that
the mechanism of natural selection does not work after the conclusion of
the sexual cycle. The spine must function appropriately only until the end
of the reproductive phase. Selection cannot correct genetic errors that
become apparent only during senescence.
The Greek osteoarchaeological record offers a few spectacular examples
of degenerative osteoarthritis of the spine. The most famous is that of a
Middle Bronze Age man from Asine (3 FA.) described by Carl M. Furst in
1930. In this stalwart individual between 40 and 50 years of age, numerous
vertebrae have osteophytes, and six vertebral bodies of the lower thoracic
region are fused into a single block. According to Furst, "This man suf-
fered from a very serious case of arthritis deformans chronica anchylopoe-
tica. " 136The diagnosis of the Swedish anthropologist seems correct; it only
has to be translated into current medical parlance. The man from Asine
had degenerative ankylosis of the lower spine, most likely Forestier's an-
kylosing vertebral hyperostosis. The same diagnosis (Forestier's syndrome
or even Resnick's DISH [diffuse idiopathic skeletal hyperostosis]) can be
considered in the case, mentioned above in a discussion of calcification of
the larynx, of the skeleton 65 AI(. exhumed in the Kerameikos in Ath-
ens.137 On a skeleton from Grave Circle Bat Mycenae (S9 Myc.), one can
see three thoracic vertebrae (T 9-11) fused together through ossification of
the disk portions of the anterior longitudinal ligament and numerous os-
teophytes on the cervical, lower thoracic, and lumbar vertebrae. These are
the remains of a warrior prince who died at about age 50, a particularly fit
and sturdy fellow. 138 Surely some extraordinary stress on the spine (athletic
training and wrestling?) contributed to the development of this ailment.
The man has exostoses at almost every tendon attachment, so that other
etiological factors were certainly involved. Again, the spinal column of the
specimen I8I Ler._,a fairly muscular 40-year-old male, shows pronounced
exostoses of the vertebral bodies. There· are also some indications of spon-
dylolisthesis (forward slippage of one vertebra over the one beneath it). All
of which suggests that heavy work and trauma were at the origins of this
case of spinal joint disease. 139
Among the many vertebral osteophytes found in Greece, some are due
to cases of herniated disk, an exclusively human disease that sometimes
takes the form of severe sciatica. The syndrome in question is described in
the Hippocratic corpus. 140 On prehistoric and protohistoric bones, degen-
erative spondylitis usually occurs in the thoracolumbar region; its second
choice is the cervicothoracic. Furst reported an instructive instance from
Asine (4 FA.), 141 while Angel has stressed the relative frequency of cervical
spondylitis among the ancient inhabitants of Lerna. 142 As for the statistical
frequency of degenerative osteoarthritis of the spine in the eastern Medi-
terranean, J. L. Angel has estimated that this ailment (including all forms
of spinal osteophytosis) occurs in approximately 70 percent of prehistoric
specimens and that the percentage drops to about 40 percent for the
historical periods of antiquity . 143 In subsequent publications, Angel pro-
vided the following figures: 74 percent at Nea Nikomedia in Macedonia
(Neolithic), 61 percent at <;atal Hiiyiik in Anatolia (Neolithic), and 62
percent at Karata~ in Lycia (Early Bronze Age). 144 The same order of mag-
nitude is in force at Lerna, but there is a significant split between the men
and the women. With the former, spinal osteophytosis reaches 75 percent,
while among the latter it is only at 50 percent. 145 To be sure, the majority
of the cases counted are mild and of no import clinically. But the statistic
remains very interesting to the historian because of the role played by
stress, especially physical labor, in the etiology of degenerative osteoarthri-
tis: the numbers given show that a clear improvement in conditions of life,
above all in working conditions, took place during the archaic period and
at the beginning of the classical period. The statistical effect of this change
of conditions in the social environment is all the more significant since a
concurrent increase in the average life expectancy should have influenced
the numbers in the opposite direction from the one they take.
Nevertheless, it would be incorrect to conclude that reduction in the
frequency of this ailment is a continuous and regular historical phenome-
non. For example, spinal osteophytosis was rare among the Hittites of
central Anatolia prior to the time it was common in Greece; it was much
more widespread in medieval Byzantium than in Greece during the classi-
cal period. The difference between masculine and feminine morbidity is
essentially due to social factors, such as sexual discrimination in occupa-
PALEOPATHOLOGY
tion, not to bioiogical differentiation. On bones from Kalinkaya, a proto-

Hittite necropolis in central Anatolia dated 4000-3000 B.c., spinal osteo-
phytosis occurs in 50 percent of the women and 33 percent of the men. 146
Wedge-shaped deformation of certain vertebral bodies and the existence
of Schmorl's nodules (hernias of the pulpy nucleus in the spongy matter
of the vertebral body) make likely (e.g., in 69 Ler.) 182 Ler.) 196 Ler.) and 62
Myc. 147) the diagnosis of Scheuermann's disease, an affection of both bone
and cartilage during growth. Its etiology is unknown, and it occurs as
kyphosis in adolescents, often evolving into degenerative joint disease of
the spine. According to G. Bergmark, this disease is represented on an
ancient statue of Antinoos of Bithynia, Hadrian's lover. 148
Differential osteoarchaeological diagnosis is especially difficult in the case
of arthritis. 149 A joint's inflammatory reaction can be. septic, that is, due
to a direct microbial influence, or rheumatic, that is, not linked to the
local presence of microbes. Septic arthritis in the strict sense of the term,
particularly suppurating joints brought on by nonspecific pyogenic bacteria
and tuberculosis of the bones and joints, will be touched upon elsewhere.
The limited resources of osteoarchaeology do not allow for the diagnosis
of oculo-urethro-synovial syndrome (RS or Reiter's syndrome or, more
correctly, the Fiessinger-Leroy-Reiter syndrome), whose infectious origins
have recently been proven, nor for certain kinds of multiple joint arthritis
whose etiology is still unknown (Felty's syndrome, lupus polyarthritis,
psoriatic rheumatism).
Unfortunately, the same is true of the most important of all the rheu-
matic diseases, rheumatic fever or Bouillaud's disease. This form of multi-
ple joint arthritis appears after a streptococcal sore throat or an outburst of
festering infection in some other focus-the bacteria act at a distance,
without local infection of the joints affected. The disease especially strikes
children and young adults, and it often brings with it serious cardiac
complications. A very common disease in Europe and America before the
antibiotic era, it was surely present in the Greek world from prehistoric
times. According to the oldest Greek medical texts, red sore throat with
pus (streptococcal tonsillitis) was commonplace. The Hippocratic corpus
contains a concise yet readily identifiable description of Bouillaud's
disease. 150
So rheumatic fever existed in ancient times, but since it produc~s no
bone changes we can provide no osteoarchaeological proof of its presence.
By contrast, rheumatoid arthritis, which is now the most common form
of chronic inflammatory arthritis, does leave enduring marks on bones but
perhaps did not yet exist in antiquity. This disease develops in irregular
stages, affecting the limb joints, especially the wrist and the fingers, and is
frequently symmetrical. It generally prefers women to men, and although
it can arrive at any age, it usually begins after age 40. 151 First defined as a
clinical entity in 1800 by A. J. Landre-Beauvais (under the name "goutte
asthenique primitive" [gout that begins with weakness]), rheumatoid ar-

thritis seems to have been an unusual disease before that date. Its clinical
aspect is very striking-the effect on hands can be disabling-and, as was
said above, among present populations it is common. So it is all the more
surprising that clinical descriptions, pictorial representations, and osteoar-
chaeological evidence prior to the nineteenth century that might relate to
this disease are so rare. It is possible to prove its existence in the seven-
teenth century-from a description by Thomas Sydenham and paintings
by P. P. Rubens-and to suspect its existence in the Middle Ages-the case
of the Byzantine emperor Constantine IX, a passage in the eighth book of
the encyclopedic treatise De proprietatibusrerum by Bartholomew Anglicus,
some Flemish paintings going back to about 1400, and some dubious
osteoarchaeological cases-but nothing enables us to follow its history fur-
ther back in time. 152 A diagnosis of the disease was posited by E. G. Smith
for an Egyptian mummy of the Fifth Dynasty, but specialists who have
looked into the case since consider it only an unconvincing possibility. To
explain the paucity of historical information about the existence of rheu-
matoid arthritis, C. L. Short offered the hypothesis that it is a relatively
recent disease that appeared as the result of a genetic transformation of the
factors responsible for ankylosing spondylitis. 153 The cause of rheumatoid
arthritis is still unknown, but certain immunological phenomena that go
with it suggest the intervention of an autosensitization process with an
immunological attack of the organism on its own tissues. The appearance
of the disease at a certain moment in human history and its spread can be
explained as immunogenetic events. Irt support of the etiological role of
genetic factors, we can adduce the familial grouping of victims and the
very high frequency among such groups of the histocompatibility antigen
HLA-Dw4. 154 Nevertheless, the possibility cannot be excluded that rheu-
matoid arthritis is an infectious disease caused by a slow virus. In that case,
the diachronic fluctuations in its prevalence could be still more easily ac-
counted for. 155
A discovery on Greek bones has just come up in the debate on the
origin of rheumatoid arthritis. According to Linda Klepinger, a skeleton
from the Hellenistic era (ea. 300 B.c.) found at Morgantina in Sicily has
joint lesions and osteophytes that resemble no currently known rheumatic
disease. According to Klepinger, it could be an intermediate stage between
ankylosing spondylitis and rheumatoid arthritis, that is, a kind of archaic
rheumatoid arthritis. In that case the skeleton would be a paleopathologi-
cal proof of the hypothesis Short made as the result of medico-historical
research and epidemiological considerations. 156 Eminent rheumatologists
reject any idea of a historical kinship between the two diseases in question.
The paleopathological diagnosis of Linda Klepinger has been strongly crit-
icized, and the debate is still unresolved. 157
As for ankylosing spondylitis itself (pelvispondylitis ossificans, Striimpell-
PALEOP ATHOLOGY 85
Marie's disease, pr Bechterew's disease), the supposed or disputed ancestor
of rheumatoid arthritis, all specialists agree that it goes back to the dawn
of mankind. A disease that is from every point of view comparable to
ankylosing spondylitis has been observed in monkeys. 158 Its prehominid
origin seems likely. It is a chronic disease of unknown etiology that by
preference strikes young adult males. Beginning insidiously with a painful,
ankylosing inflammation of the sacroiliac joints, it then attacks one verte-
bra after another, going from lower to higher. The involvement of periph-
eral joints is morphologically indistinguishable from rheumatoid arthritis.
Genetic predisposition plays an incontestable role. Hereditary transmission
of the disease takes place by an autosomal dominant factor. In more than
9 of 10 victims the histocompatibility antigen HLA-B27 is found, while its
frequency in the general population is on the order of 4 percent. A viral
infection may trigger the disease. For. instance, this form of spondylitis
may follow Reiter's syndrome, whose infectious origin is a certainty. 159
The oldest osteoarchaeological case of ankylosing spondylitis was re-
ported in France on a Neolithic skeleton. 160 According to E. G. Smith,
Wood Jones, M. A. Ruffer, and A. Rietti, this disease was very common
in Pharaonic Egypt. But there is a basic error in the statistical estimates of
these paleopathologists: they thought they saw ankylosing spondylitis
where there was only ankylosing vertebral hyperostosis. 161 It remains true
that in certain cases from Egypt the old diagnosis seems utterly justifiable.
My conclusion is that ankylosing spondylitis existed in Egypt at least since
the Third Dynasty (ankylosed spine of a man named Nefermant), but it
was relatively rare. 162 Among the Macedonian soldiers of Alexander the
Great whose tombs are at Chatby near Alexandria, two doubtful cases of
the disease have been found, though a firmer diagnosis can be made for a
female mummy of the Ptolemaic era. 163 Ankylosing spondylitis has not
been reported, as far as I know, on ancient vertebrae from Greece or Italy.
Hippocrates probably observed it, since he mentions rigid curvature of the
spine as a consequence of old age or ''from giving in to pains.'' 164
Notes on Infectious Diseases

Pathogenic germs have often framed the destiny of individuals and so-
cieties. If humanity succeeded fairly quickly in gaining mastery over the
animals, until the last 100 years it was defenseless against countless living
creatures who were so tiny that they escaped notice and so powerful that
for millennia they were by far the most significant cause of disease and
death. Bony lesions of microbial origin are only a minor component of the
pathology of infectious disease. Even so, they offer direct proof of the
antiquity of pathogenic parasitism, and, for the historian of diseases, they
constitute the only visible portion of a huge iceberg whose submerged
portions are lost in the past.
I have devoted a whole chapter to the commonplace germs that cause

nonspecific festering inflammations in affected tissue, bony or otherwise.
Other chapters of this book are concerned with treponematoses, leprosy,
and tuberculosis, which are the main specific inflammatory diseases. I
found it necessary to include the osteoarchaeology of these diseases in a
medico-historical inquiry of wider scope, one that takes special notice of
the most recent microbiological, epidemiological, and immunological re-
search. A single en demo-epidemic disease caused by Protozoa, malaria,
affected the course of Greek history more than any other pathological
factor. Though it has no direct effect on bones, malaria does leave traces
on them since it promotes certain hereditary defects that affect bone mar-
row. A separate chapter is also devoted to this subject.
What malaria was to Greece in decisive moments of its history, schisto-
somiasis (bilharziasis) was to Egypt. It is a debilitating disease caused by
the parasitic worm Schistosornahaernatobium)a species of fluke whose calci-
fied eggs have been found in the kidneys of Egyptian mummies from the
Twentieth Dynasty. 165 No such discovery can be made in Greece, since
Greeks did not mummify corpses. Schistosomiasis seems to have been
known to the physicians of the classical era, although it was not seen in
Greece itself. Actually, climatic conditions and the aquatic fauna of the
country have not been hospitable to the life cycle of this parasite, in
particular with regard to its intermediary infestation of aquatic gastropods.
Acute viral diseases spare bone tissue. However, it is possible for the
paleopathologist to discover traces of their consequences. Thus one can
imagine that the extremely slender left femur shaft of a 40-year-old male
from Lerna (95Ler.) is the result of a partial paralysis that suggests, in turn,
infection by the poliomyelitis virus. 166 In such cases, the diagnosis is very
hypothetical. If we use it to prove the existence in antiquity of this viral
disease, there is a strong possibility that the argument is circular. Some
scholars think they recognize traces of smallpox on Egyptian mummies. 167
Considering the contagiousness of that virus and the existence of commer-
cial links between all lands bordering the Mediterranean, such traces would
be indirect proof of at least an episodic presence of the scourge of smallpox
in ancient Greece.
The spinal deformation in an Iron Age woman exhumed in Greece may
have been brought on by blastomycosis, 168 a fungal infection of the skin
and internal organs, but a diagnosis of tuberculosis seems more likely.
According to Jean Dastugue, Paleolithic man in Cro-Magnon times was
already the victim of another fungal disease, actinomycosis or lumpy jaw. 169
A new examination of this specimen suggests that its lesion had a nonin-
fectious origin. It seems that 1 Cro-Magnon died a victim of malignant
eosinophilic granuloma (a disease now called histiocytosis X) . 170
Chapter Three
P ALEODEMOGRAPHY
Evidencefrom Ancient Boneson the Conditions
of Daily Life in Greece
To obtain even a vague idea about the health of the ancient inhabitants of
the Greek world, it is useful, if not indispensable, to know beforehand
something of their demographic traits. The pathocoenosis of these peoples
would be completely obscure were it not possible for us to ascertain at
least the order of magnitude for such parameters as the total number of
inhabitants, their dernographic density, their degree of urbanization, their
distribution by gender and age cohort, their fertility, average lifespan, and
average height, and the frequency among them of certain bodily defects.
The absolute values of such parameters at a single time and place concern
us less, however, than their geographic and chronological fluctuations.
This list of demographic traits that can serve to profile a pathocoenosis
leaves out birth and death rates, for the simple reason that in our present
state of knowledge about ancient Greece there is no way to make ade-
quately secure inferences concerning them. As for the parameters I have
listed, osteoarchaeological examination provides information precise
enough to justify, as was said in the previous chapter, the recent flowering
of a specific discipline called paleodemography . 1 Insofar as paleodemogra-
phy is based on the osteological and archaeological record, it is not limited
strictly to historical periods. But when it does concern itself with history,
availing ourselves of written sources can usefully enhance osteoarchaeolog-
ical methods.
Estimates of the number of inhabitants, their demographic density, and
their degree of urbanization are a crucial ingredient in our understanding
of the epidemiology of the past, given that certain infectious diseases take
different forms depending on the size and structure of the populations
they attack, and given also that they invariably pass over populations whose
size and density fall short of a certain threshold. Average height as well as
tooth and bone characteristics yield general information about diet. In
chapter 2 we saw how the percentage of rheumatic lesions allows inferences
about stress from physical labor. All these data, along with estimates of
average lifespan and the proportion of infant to adult skeletons, make
possible a synchronic overview of the health of a population at a given
moment in its history. For diachronic study, such data enable us to take a
global look at the variations in health and bring to light periods in which
the equilibrium of a pathocoenosis is broken.
The Dynamics of the Ancient GreekPathocoenosis

For the sake of clarity, I interrupt for a moment the logical thread of my
exposition to offer a perspective on the dynamics of the Greek pathocoe-
nosis. Fuller argumentation will follow. 2
The transition in the Neolithic period from hunting and gathering to
agriculture and cattle-rearing must have had a profound effect on the state
of health in human groups. 3 Earlier than most, the Balkan peninsula and
the islands of the eastern Mediterranean were the domain of this change-
over from a predatory to a producing economy. The Neolithic revolution
began on several fronts at once. One of the most ancient of these, perhaps
the most ancient in the world, was western Asia. From there the new
civilization spread slowly into Europe, where its first manifestations are
visible in what would later become the Greek world. Humans become
sedentary and settle in villages. They work the land and domesticate ani-
mals without giving up hunting or fishing. Judging by carbon 14 dating of
a sample from Nea Nikomedia, the oldest villages in this region date from
6200 to 4500 B.C.
Each Paleolithic human required approximately 5 to ro square kilometers
of land to sustain himself. In the Neolithic period, a smaller lot sufficed to
ensure a more varied and abundant diet, and the resultant increase in
resources facilitated an extraordinary demographic expansion that had
complex repercussions for general health. Relative overpopulation tends to
slow improvements in the food supply, or even to reverse here and there
an initial trend toward abundance. Loss of mobility and concentration in
villages inevitably brings outbreal(s of new infectious diseases as well as the
revival of certain old ones. With humans and domesticated animals living
under the same roof, certain infectious diseases developed that were com-
mon to both, and a particular group of viral diseases arose. Moreover,
communal life in a fixed locale caused serious problems with the supply of
drinking water and the disposal of waste and excrement. Cycles of reinfec-
tion began establishing themselves, allowing several parasitic diseases caused
by bacteria, protozoa, and intestinal worms to become endemic. The
PALEODEMOGRAPHY
settling of villages in fertile areas near bodies of water and the irrigation of
fields fertilized by manure favored the spread of several especially noxious
diseases. In addition to these pathogenic factors, there were changes in
climate as well as the beginnings of the division of labor and social stratifi-
cation. As a result, silicosis appeared among stonecutters, lead poisoning
among potters, mercury poisoning among the producers of cinnabar (no-
tably at Vinca), higher mortality among women than men owing to their
servitude at hard labor, and, generally speaking, the first differentiations of
pathology in relation to social class. The first irreversible transformations
of the environment, like deforestation and pasturing, were to have long-
term public health consequences. Most Neolithic sites were small, com-
prising scarcely more than a few hundred souls, but recent archaeological
excavations have brought to light the existence of important communities
that have almost the same proportions as true cities.
The ease and speed of sea travel has played as significant a role in the
history of diseases as demographic expansion. Just as stone tools were
transported throughout Greece (the inhabitants of the Franchthi Cave
were already using obsidian from Melos in the Mesolithic period), so were
disease germs carried quickly and easily from one place to another through-
out the region. As a result, most places in the eastern Mediterranean
during the Neolithic period came to share a pool of infectious diseases.
Time would only reinforce their unity in this regard.
The conditions of daily life during Neolithic times were perhaps not as
bleak as the exclusive consideration of new pathogenic factors makes them
out to be. The biological advantages of a sedentary life in a protected
habitat, with essential needs for survival more or less assured, probably
won out over the health problems, or at least compensated for them. That
seems especially t1ue for Greece-more so than for Asia-where the sun,
the climate, the fauna, and other special features of the ecology were not
favorable to the great killers from Africa and Asia, except for tuberculosis,
malaria, and typhoid. Although the evidence leads us to believe that cer-
tain viral diseases arose in Mediterranean lands after the domestication of
animals and once the population had surpassed a certain critical density, it
does not seem that their origins can be located in the Mediterranean. Even
in Hippocrates' time, Greeks still did not know of smallpox, measles, or
plague, nor did they ever face scourges like bilharziasis or yellow fever.
And yet average age and various other characteristics of Neolithic bone
remains point to a generally poor biological state. These indicators do not
change significantly during the whole prehistoric period, except in relation
to the salubriousness of particular sites. Even in the Early Bronze Age they
remain about the same, which amounts to saying that the conditions of
life neither improved nor worsened during the long series of centuries
stretching from the seventh to the third millennium B.c. The apogee of
Minoan civilization and the beginning of Mycenaean civilization in the
Middle Bronze Age (ea. 1900-1600 B.c.) and the flourishing of the Myce-
naean sea empire in the Late Bronze Age (ea. 1600-1200 B.c.) go hand in
hand with a clear improvement on the paleodemographic plane in physi-
cal well-being. 4 For instance, lifespans lengthen perceptibly, and the
frequency of porotic hyperostosis decreases to a spectacular degree.
Though the latter phenomenon (the sign of a drop in endemic malignant
malaria, or perhaps of improvement in the nutrition of children) begins
in the Neolithic period, the jump in longevity takes place all of a sud-
den. It is probably as much a result of changes in physical surroundings
and social structures as of the arrival of a mass of new genes. There is
no way for us to discern with certainty the role ascribable to each of these
factors.
It is certain, however, that the inhabitants of Greece in the Bronze Age
are no longer pure biological descendants of the tribes whose skeletons fill
the Neolithic cemeteries. From the end of the third millennium B.C.,
several waves of Inda-Europeans invaded the Aegean area, first destroying
the old economic centers, especially in the Argolid, and then building
Mycenaean palaces, fortifications, and villages. Only after the "arrival of
the Greeks" is a true physical improvement noticeable. I wish to stress
particularly the biological importance of the extraordinary genetic variety
of the mixture that is at the origins of the Greek people. This "racial
impurity," this heterogeneity, gave the Greeks their vigor, adaptability,
and resistance to the often noxious forces in their environment. So on the
one hand, chance encounters and the need to survive at any price made
the phusis (nature) of the Greeks a rich and versatile one, especially adept
at surviving all kinds of biological and social adversities. On the other
hand, this physical well-being could never have been realized without the
rapid development of a "civilized" society that knew how to control a
number of pathogenic factors. For instance, the bookkeeping that went
on in the Mycenaean palaces reflects a fussy organization of the production
and distribution of victuals. There is archaeological evidence to suggest a
concern for cleanliness and, at least among certain classes, an interest in
personal hygiene and fashion in clothing. And there is striking proof of
technical progress in supplying water and constructing sewers. In any case
the growth of the population and the crowding of people into palaces and
feudalistic settlements bespeaks the need for sanitary arrangements. 5
Opulent, fortified centers had one great enemy: overpopulation. Avail-
able resources dictated the size of the towns that utilized them. Emigration
and colonization were insufficient to overcome such constraints. From
time to time people must have suffered subsistence crises, and by the Late
Bronze Age the Aegean population had attained the density necessary for
the outbreak of catastrophic epidemics. Paleodemography reveals to us a
clear worsening in the condition of women, and the higher mortality rate
PALEO DEMOGRAPHY 91
for them than for men does not cease climbing until the Golden Age of
the classical peri6'd.
The archaeological record signals the material decline and collapse of the
bureaucratic states organized around the palaces during the period from
1200 to noo B.C. Then the Dark Ages of Greek history began, marked by
the Dorian invasion and the arrival of a civilization based on iron. 6 By the
ninth century, the demographic shock seems to have been absorbed. A
new type of society emerged with even more anthropologically variegated
populations that suffered from infighting. There was still a tendency to-
ward unity of culture and biological homogeneity. The general improve-
ment in health that began in the Bronze Age halted or even reversed itself
during the Iron Age, only to take off again with renewed and surprising
vigor beginning around 800 B.c. 7
Several generations before the classical period (strictly speaking) begins,
the height of the inhabitants of Greece increases, and their physical aspect
changes (e.g., the relative length of long bones and the roundness of their
shafts, or the depth of the pelvic inlet). The frequency of lesions from
chronic wear and tear as well as infantile mortality both diminish, while
the average lifespan reaches limits that it will not attain again until the
twentieth century. In fact, the supply of food, the fit between human and
habitat, the physical condition of people, and general well-being were
never better in the West during antiquity. 8 To be sure, there were a
number of shadows across the landscape. Warfare was unceasing, work was
often harsh and exhausting, and the variety of diseases was greater than
ever. The heaviest threat was still overpopulation. Well-being generates a
strong demographic surge, and that, in turn, forces colonial expansion,
the export of human beings, which was particularly intensive from the
eighth to the sixth century B.C.
Beginning in the sixth century a profound transformation seems to have
taken place in Greece. From an essentially agricultural land, it became, at
least in some regions, an important industrial power with outside trade
substantial enough to permit population growth without an increase in
agricultural productivity. The import of grain and other foods made colo-
nization less pressing. The age when communities begin to live not just
from agriculture but also from the products of their artisans and from trade
marks the real debut of urbanization in Greece-one can cite, for example,
the concentration of people and wealth in Athens and Corinth-- even
though the p6lis is of modest proportions compared with certain urban
centers in Asia and with what the city of Rome would become. 9
By 600-500 B.C. the pathocoenosis of Greece was in a rarely attained
state of equilibrium. Humanity seems to have adapted well enough to the
forms of parasitism native to Mediterranean agriculture, and endemic ma-
laria seems to have abated. Thus, at the dawn of the classical period, the
inhabitants of the Aegean region enjoyed exceptional health, but that

situation would not last for long. The change took place over the fifth
century B.C., not abruptly, but still profoundly and inexorably. The flour-
ishing of "scientific" medicine under the name of Hippocrates could not
halt the decline; from the perspective .of public health, its contribution
was merely psychological. In fact scientific Greek medicine arose at the
time when the pathocoenotic equilibrium was crumbling; as it developed,
the health of the Greeks was deteriorating. Some might consider that a
kind of paradox.
The success of the Greeks in adapting themselves to their environment
and in mastering pathogenic factors in it was in great part due to their
cultural development. Good health fed upon the Greek miracle and added
to its ampleness. However, the new forms of civilized life carried within
themselves the seeds of misfortune. Without knowing of the paleodemo-
graphic arguments at our disposal nowadays, the great historian Jacob
Burckhardt (1818-97) once remarked that the fifth-century Greeks had a
heightened need to consult their doctors. They were more sickly, said
Burckhardt, not because of an easier life but from their unbridled passions,
"from an unhealthy activity throughout the city. " 10 While Burckhardt
and several other historians of the nineteenth century saw only moral
causes, current medical concepts force us to seek out material ones. The
decline in public health was still understated during the Peloponnesian
War, apart from the "plague of Athens" and its aftereffects, but it became
obvious during the fourth century B.C. According to Angel, "It was a
response to social conditions before it was a response to disease organisms.
The direct causes of the increase in such diseases as malaria, typhoid, and
tuberculosis must have been: (1) rural dislocation through constant petty
warfare, with inevitable breakdown of proper swamp drainage and irriga-
tion; (2) urban overcrowding and inadequate expenditure on sewers or
sanitary inspectors; (3) wandering of war refugees." 11 Just at this time the
shores of the Mediterranean were forming, as was said above, a shared
pool of infectious diseases. Near the end of the Hellenistic era and in
Roman times, this pathocoenosis would again be seriously disturbed by
exchanges of microbes with pools originating in the Far East and Africa. 12
GeographicUni-tyand Genetic Varie-ty

among the Inhabitants of Greece
Historically, Greece consists of the promontory of the Balkan peninsula

(mainland Greece), some islands of the eastern Mediterranean, and the
west coast of Asia Minor. Outside this Aegean region, the Greeks colo-
nized some adjacent lands, for instance, southern Italy, Sicily, Syria, and
the coast of Egypt. But their colonial expansion never went beyond the
PALEODEMOGRAPHY 93
banks of the Mepiterranean. At no point does the Greek world leave the
temperate zone behind. Wherever Greeks settled on a long-term basis,
they were able to introduce a way of life that they had devised in perfect
accord with the geographical conditions of their civilization's birthplace. 13
In Greece, climate varies perceptibly from one region to the next, al-
though some traits ascribable to the preponderant influence of the sea
persist everywhere. "Nothing to excess"-this maxim of the Greek wise
men mirrors the profile of their natural environment. Situated between
the cold of northern Europe and the warmth of the African continent,
maritime Greece revels in its temperate heat, etesian winds, the regularity
of its seasons, its dry summers and wet winters. As a whole, the climate is
remarkably healthy, hospitable to a free-spirited life favoring productive
labor, sports, and intellectual activity. 14 ·
By virtue of its position at the crossroads of the continents, the Greek

world forms a bridge between Asia, the northern coast of Africa, and
western Europe. 15 It served as a link between the Mediterranean basin and
central Europe, and the great' routes of infectious disease went straight
through it. 16 Not only did the first great historical pestilences pass through
Greece on their way north and west, but it also saw the slow, insidious
penetration of endemic disease (like tuberculosis and malaria in the distant
past and, more recently, leprosy). On the other hand, its temperate climate
acted as a barrier against the diffusion into Europe of the so-called tropical
diseases whose vectors or germs could survive only under specific physical
or biological conditions. Sheltered from yellow fever, schistosomiasis,
sleeping sickness, filariasis, and perhaps acute cutaneous treponema infec-
tion as well, to cite only a few of the large number of diseases that ravaged
lands with less merciful climates, the Greek world nonetheless suffered
terribly from one African pestilential disease: malaria. I will discuss it in
detail elsewhere, but it deserves mention here since, as its name suggests,
it represents a disease closely tied to ecological factors. The '' malarial com-
plex,'' as defined by Max Sorre, existed in Greece from prehistoric times
in its most typical form. A significant portion of Greece is marshland (I
recall in this regard Aristotle's statements that the swamps were a cause of
the depopulation of certain parts of the Argolid). 17
Greek soil is on the poor side. Even in our time less than a fifth of it is
fertile, and since the Greeks practiced biennial crop rotation, at most only
a tenth of the land was harvested in any given year. Deforestation from
protohistoric times and especially in the Hellenistic era increased the
amount of ploughland but brought with it changes in microclimate that
were unhealthy for both air and water. To the modern observer, Greece
proper is a small country. Its total surface area is approximately 132,000
square kilometers, while Attica has only about 2,800 (more like 2,600 in
the classical period) and the Peloponnesus about 21,500. These relatively
modest dimensions are surprising, considering the role that this land and
its inhabitants have played in the political and cultural history of mankind.
Greece lacks great plains, which tend to favor huge concentrations of
humanity. So to a certain extent its political partitioning in the archaic and
classical periods reflects its topographic compartmentalization. Plato tells
us that citizens actually left Crete and _the Peloponnesus because in the
cities the population had "surpassed the number that the land can feed." 18
The links between physical surroundings and the constitution of those
who inhabit them, not to speak of their political institutions, did not
escape Greek intellectuals. In fact, it is the central topic of the treatise Airs)
Waters) and Places) one of the most charming texts in the Hippocratic
collection. For many historians of medicine, it is the most authentic work
of the great master from Cos. Whatever the case may be, its dating to the
fifth century B.C. and its Coan origin can hardly be doubted. The first
section of the treatise lays down the foundations of medical geography: it
tries to ascertain the effects of winds, waters, and the placement of cities
on the health of their inhabitants. In the second part, there is an explana-
tion in ecological terms of the differences between Greeks, Scythians, and
Asiatics with respect to certain physical and psychological traits and their
political consequences. 19 This treatise "places mankind under a strict de-
terminism" that is, as R. Joly judiciously remarks, "very exaggerated and
too often incorrect. " 20 But all the reservations one can have about the
correctness of the etiological positions taken by the author of this treatise
do not detract from its value as a descriptive witness to the salient nosolog-
ical elements of the pathocoenosis of the Greek city-states near the end of
the fifth century B. c.
I have spoken elsewhere of the work's epidemiological content. 21 Here
I will cite a few passages from its second part that have to do with climatic
changes and their effect on the general constitution of Hippocrates' con-
temporaries. After speaking of the Scythians, the author portrays the rest
of the Europeans (meaning, as far as he is concerned, the Greeks) as
follows:
The other people of Europe differ from one another both in stature and in shape,
because of the changes of the seasons, which are violent and frequent, while there
are severe heat waves, severe winters, copious rains and then long droughts, and
winds, causing many changes of various kinds . . . It is for this reason, I think,
that the physique of Europeans varies more than that of Asiatics, and that their
stature differs very widely in each city . . . For the frequent shocks to the mind
impart wildness, destroying tameness and gentleness. For this reason, I think,
Europeans are more courageous than Asiatics. 22
The extraordinary variety in the inhabitants of archaic and classical Greece

is fully confirmed by the anthropological study of their skeletons. The
variety is due above all to their genetic polymorphism, to the wealth of
their gene pool. The climate has something to do with this state of affairs,
but in ways not foreseen by the Hippocratic writer.
PALEO DEMOGRAPHY 95
DemographicI&ensity
The study of the demographic history of the ancient eastern Mediterra-
nean seems relatively secure as regards the determination of general trends
in the evolution of populations, but we encounter insurmountable obsta-
cles when we come to the degree of precision in the actual numbers it
supplies. For prehistoric periods, we can make hypotheses that stand up
to criticism only if they are limited to a relatively imprecise order of mag-
nitude. For the classical period, we have to be content with the vague
information given by ancient historians and topographers. Manipulation
of this information does not provide incontestable results.
The total number of hunters scattered over the European biotope dur-
ing the Paleolithic period was probably, at any single· moment in their
history, fewer than 100,000. In the Mesolithic period, after the last Ice
Age, the number increased perceptibly, but it was not until the beginning
of the Neolithic that, by virtue of a sedentary way of life and a broader
variety of means of subsistence, the overall population of Europe reached
or even surpassed I million. For centuries, these primordial farmers lived
in a natural environment so empty that the average density was equal to
or less than I inhabitant for each square kilometer. Over the long term,
this figure increased slowly but steadily. At the beginning of the Bronze
Age, the population of Europe rose to a few million, and by 1000 B.c. it
had reached about 10 million. Speculation on the actual density is useless,
since the distribution of people was extremely uneven. Most Europeans
kept to the south, where the climate was more forgiving. According to
Colin McEvedy and Richard Jones, ''By the end of the Bronze Age in
1000 B.C., the density of population was higher than the European average
by a factor of three in Greece and more than two in Italy. This is the
demographic background to the emergence of classical society. Greece set
the pace. Between 1000 and 400 B.c., the population of Europe doubled,
increasing from 10 to 20 million; in the same period the population of
Greece tripled, reaching a final total of three million, an amazing figure for
the era. " 23
Neolithic civilization entered Europe by way of the Balkans. That is
where it bore first fruit on the continent, and where it initiated the Euro-
pean demographic upheaval_. In 5000 B.C. about a quarter of a million
farmers were already living on the Balkan peninsula. At the apogee of
Mycenaean civilization, their number reached about 2 million, with almost
half of them living in continental and insular Greece.
Of all the guesses that have been made about the number of inhabitants
in prehistoric Greece, the one most worthy of respect at the moment is by
Colin Renfrew. It is based on a scrupulous analysis of the number and
importance of the archaeological sites, especially necropolises, in the Ae-
gean region. 24 The first villages in this geographic area were small. Nea
Nikomedia or (:atal Hiiyiik counted several hundred inhabitants, but most

other villages had scarcely a hundred. Houses were scattered here and there
without concern for common, fortified protection. The concentration of
persons was very low, about 200 per hectare of village land. The demo-
graphic density was approximately I inhabitant per square kilometer. By
Renfrew's calculations, it ranged from 0.85/krn. 2 in Laconia and central
Macedonia to 1.77/km. 2 for Messenia, with the Cyclades at 1.20/km. 2 and
Crete at 1.53/km.2
The history of the Aegean population from the beginning of the N eo-
lithic to the end of the Bronze Age is typified by almost exponential
growth (that is, it was constantly accelerating) in some regions, notably
Crete and Messenia, and by growth of the same general intensity but with
a marked decline during the Early Bronze Age in other regions, for in-
stance, in Laconia or the Cyclades. The first type of growth can be consid-
ered "natural" or in conformity with a relatively favorable biological state
of affairs and free from dire external disturbances. The second type of
growth shows that a catastrophe, or rather a series of them, intervened
and struck the population of certain parts of Greece between 2500 and
1900 B.C. In this case the catastrophe was certainly the invasions by Indo-
European peoples and the disturbances resulting from them, but the in-
formation available does not allow us to say to what extent this prolonged
disaster was caused by political events and to what extent biological factors,
including pathology, contributed to the situation. In any case we can rule
out pestilential diseases, since they would not have spared the most pop-
ulous areas.
Here are Renfrew's estimates, in round numbers, for the demographic
density of several regions during the Bronze Age-the three numbers in-
dicating, successively, the situation during the Early, Middle, and Late
Bronze Ages-Crete: 9, 26, and 31per square kilometer; Messenia: 8, 46,
and 63; Cyclades: 14, 8, and 12; Laconia: 7 .5, 8, and 14; and central Mace-
donia: 2.5, 2.5, and 4.5. The population of Crete increased from 75,000 in
the Early Bronze Age to 250,000 in the Late; Messenia jumped from
23,000 to 178,000, while Laconia went from 26,000 to 50,000 and the
Cyclades decreased from 34,000 to 29,000. 25 For the eastern Mediterra-
nean in general, J. L. Angel maintains that the demographic densities for
all three Bronze Ages were, respectively, 10, 18, and 30 inhabitants per
square kilometer. 26 To be sure, all these statistics are conjectures that can
only serve as general indicators.
During the Middle Bronze Age, fortified towns proliferated and villages
grew. On average, they had a total of 500 inhabitants. So, for instance,
Lerna in the Argolid was at that time a town of about 120 dwellings, each
occupied by a relatively small family (5-7 members). At the height of
Mycenaean civilization its total population approached 800, according to
Angel. The state of public symphyses among its female inhabitants shows
PALEO DEMOGRAPHY 97
an average of 5 childbirths
., per adult (equivalent to 5.5 per fertile woman,
assuming a sterility rate of 10 percent). With a birthrate of about 44-45
and a death rate slightly higher than 40 per 1,000 inhabitants, there was an
annual growth of 0.4 percent, which is high for a preindustrial society. 27
All these figures were arrived at from the results of osteoarchaeological
analysis, and despite the exiguousness of the sample studied and uncer-
tainty about some basic values, they seem to us relatively secure, at least
in their order of magnitude.
Examination of the female pelvis, including the state of pubic symphy-
ses, makes it possible to estimate fecundity, here defined as the number of
childbirths per woman beyond the age of sexual maturity. Over the land
of Greece, fecundity varies between 4 and 5, with a rise during the Neo-
lithic and the Middle Bronze Age and a fairly clear drop during the Early
Bronze and the Iron Age. The number of children per adult woman may
seem small, given the likely absence of contraception, but it is explicable
in terms of the shortness of life and thus of female genital availability: on
average, a woman's period of fecundity did not exceed a dozen years. 28
When the classical period opens, the Greeks are in the midst of demo-
graphic expansion. The population of the eastern Mediterranean grows
without respite from the ninth to the fifth century B.c., despite emigra-
tions and a slowing birthrate: pubic scars on female adult skeletons of the
archaic and classical periods suggest an average of 4.3 childbirths. In the
fifth century B.C., in the time of Pericles and Hippocrates, the total num-
ber of inhabitants of Greece was not much more than 3 million, 2 million
of whom lived on continental Greece (surface area 56,000 km. 2 ), with
another 800,000 on the Peloponnesus (surface area 21,500 km. 2 ) and some
400,000 on the islands. So the demographic density, excluding the islands,
amounts to 36 inhabitants per square kilometer. 29
In fact, as Pierre Salmon has shown, we should distinguish between
different regions: those of low demographic density, from 10 to 30, like
Aetolia, Achaea, and Thessaly; those of average density, from 30 to 100,
like Laconia (33), the Argolid (36), and Boeotia (60); and those of high
density, like Corinth with no and Attica with 160 inhabitants per square
kilometer. The enormous density of people in the isthmus and in Attica is
ascribable to urbanization, itself a consequence of the development of an
artisan class, of sea traffic, and of the slave trade. 30
Without any doubt, the figures supplied are conjectures, since no real
census of the Greek population exists from before the modern era. 31 To
arrive at the values cited above, it was necessary to use evidence from
historians on the number of citizens in assemblies and on the number of
persons a city could put in the field in case of general mobilization, the
lists of ephebes, information on wheat consumption, what the archaeolog-
ical record reveals about the surface area of towns, and so on. Thucydides'
testimony on the military force of Athens around 431 B.C. 32 has inspired
very close analysis and informative debate about the population of Attica. 33
The upshot is that there were plainly more than 300,000 inhabitants in
Attica at that time, probably as many as 420,000. The uncertainty stems
from the difficulty in knowing the quantitative relationship between citi-
zens, resident aliens, and slaves. These _lastseem to have been more than
half of the total population. So the community consisting of Athens and
the Peiraeus had, at the time Thucydides considers the high-water mark of
Athenian imperialism, a population surpassing 200,000 souls. For the con-
ditions of life in Greece, this concentration of human beings was an ex-
traordinary break with centuries of custom and habit. In the classical
period, Greeks lived in the countryside. Thucydides tells us that the ma-
jority of Athenians themselves "had only known life in the countryside"
and that being cooped up within the city's walls in wartime for protection
was to them a painful ordeal. 34
During the fourth and part of the fifth century B.c., the population of
Greece remained almost unchanged overall, though the social and geo-
graphic distribution of persons did not. On the one hand, the population
of slaves increased perceptibly compared with that of free persons; on the
other, urban centers became more populous at the expense of the coun-
tryside. 35 By the middle of the second century B.C., the demographic
decline is striking. 36 Inscriptions confirm the accuracy of this withering
glance by the historian Polybius: '' All of Greece suffers from a halt in
reproduction and a dearth of persons; the cities are being depopulated.
The problem is that men nowadays, in their love of splendor, money, and
idleness, too, no longer want to get married, or if they do, to raise a
family. It's all they can do to let one or two children enjoy their wealth
and grow up in luxury." 37
Knowledge of demographic realities is useful, even indispensable, for
any epidemiological investigation. It is certainly no coincidence that the
"plague of Athens," the first catastrophic epidemic in the West whose
symptoms and progress are really known, broke out just at the time and
place of the greatest human concentration in Greece. At the beginning of
this chapter, I said that certain infectious diseases cannot subsist in popu-
lations that fall short of a certain threshold in size and density. That is
especially true for certain viral diseases that either destroy their host or
confer lasting immunity, are transmitted from one person to the next
without any animal reservoir, and have a relatively brief acute phase; the
most important of these are measles, smallpox, and mumps. Because of
their epidemiological traits, specifically their extreme contagiousness, their
viral aggressiveness, and their capacity to immunize, they manifest them-
selves most often as acute fevers in children. Unfortunately, we do not
know the exact values of the demographic thresholds each of these viral
diseases requires. 38 According to studies recently n1ade among the Ama-
zonian Indians, and according to epidemiological observations of peoples
PALEODEMOGRAPHY 99
on certain isolatc;d islands, demographic densities comparable to those in
Neolithic Greece are inadequate to support measles. 39 It therefore seems
that that disease had no chance of sustaining itself in Greece until the
Middle Bronze Age. Its premature introduction from a more densely pop-
ulated region outside Europe could have caused a deadly, but short-lived,
epidemic tide. It is perhaps worth noting that Neolithic bones from <;atal
Hiiyiik suggest a relatively low infant mortality rate. It may well be that
the situation changed with the arrival of Mycenaean civilization. It surely
did so at the beginning of historical times, growing worse and worse all
through the classical and Hellenistic periods.
Osteoarchaeowgical
Data on Lifespan
Longevity is an essential feature of a pathocoenosis. We now have at
our disposal sufficiently sophisticated ways of recognizing the approximate
age of an individual from skeletal remains and, as a consequence, a way of
calculating the average age of persons buried in a necropolis. 4° From such
osteoarchaeological data, we can draw some conclusions as to the average
lifespan in a larger prehistoric or historic population. The value of those
conclusions depends, on the one hand, on the size and representativeness
of the sample, and, on the other, on the validity of some demographic
hypotheses. The numerical estimates reached in this way are valid only
until the next research effort modifies them.
In 1947, John L. Angel attempted an osteoarchaeological demographic
study for ancient Greece. 41 Here are the main results of his investigation:
over the whole period in question, the mean age at death was 35.4 years
(men, 39.8; women, 31); for the Neolithic and the Early Bronze Age, 32.1
(men, 34.7; women 29.6); for the Middle Bronze Age, 34.7 (men, 39.3;
women, 30.1); for the classical period, 38.1 (men, 42.6; women, 33.7); for
the Roman era, 36.8 (men, 42.1; women 31.6); and for the Byzantine pe-
riod, 33.7 (men, 36.5; women, 31). On about fifty adult male skulls from
the classical period (that is, in Angel's study, from 650 to 150 B.c.), signs
of senility occurred on about 10 percent, while on a group of about thirty
adult female skulls of the same era, they were apparent on only 3.7 percent.
Angel was fully aware of the provisional character of this pioneering
study and of problems in the numbers he produced. His sample consisted
of only 384 adult skulls spread out over the period from 3500 B.c. to A.D.
1300. Using just the state of cranial sutures as a criterion of aging, the
average age of this same sample was appreciably lower: 27 .4 years overall,
and 29.1 for the classical period, including both sexes. Intermediary figures
were obtained if age was determined for a few dozen skeletons as a function
of the ossification of the pubic symphysis: 31.9 years (men, 35.5; women,
27) for all the periods of antiquity together. 42 In the osteological material
examined at the time by Angel, there was a shortage of skulls of children
IOO DISEASES IN THE ANCIENT GREEK WORLD
and old people, a fact attributable at least in part to the selective destruc-
tion of their bones. It was out of the question to establish average longev-
ity, that is, the life expectancy of an individual at birth.
Even so, three demographic peculiarities arise clearly from these data: (1)
the shorter lifespan of women than of qien; (2) the rarity of the tokens of
senility; and (3) the increase in average lifespan until the classical period
and its progressive diminution after it (at least for persons who reached
puberty). Between the Neolithic and the classical period the gain in lon-
gevity was, according to Angel's statistics, 6 years, or 8 years for men and
only 4 for women.
Infant mortality was assuredly very high, but it is difficult to arrive at
figures for it. Pertinent information can be obtained from the numerical
relationship between children's and adult's tombs as provided in excava-
tion reports. For instance, in the classical necropolis of Olynthus, the
proportion is 290:294, which means that death overtook 49.7 percent of
this particular population before it reached adulthood (infants, 28.3 per-
cent; children, 21.4 percent). 43 In six cemeteries of the Mycenaean age the
percentage of nonadults increases to at least 55 percent. Furthermore, an
examination of 30 prepubescent skeletons of various dates allowed Angel
to state that more than half of these young persons did not live beyond
the age of 5.44
For the last thirty years, research on this subject has been continuous.
Angel has revised his figures on the basis of an ever-larger number of bone
specimens. 45 Progress has been made by restricting statistical analysis to
samples that are better defined in time and space. The order of magnitude
of Angel's initial results has not been changed, but relatively subtle modi-
fications have been made, lowering the figures slightly for very ancient
times and raising them a little for historical times. The result is an even
clearer rise in longevity at the time that Hellenic civilization came into
flower. Also, estimates of infant mortality are slowly becoming more exact.
Since the work of Henri-Victor V allois, 46 we know that inhabitants of
the European continent during the Paleolithic and Mesolithic died most
frequently during childhood and between the ages of 21 and 40. Infant
mortality was tremendous. There was a drop between the ages of n and
20, then a strong rise into the forties, beyond which most people did not
live. This overall picture conforms with the one given by Jean Noel Biraben
after minute study of skulls from the epi-paleolithic necropolis of Colum-
nata in western Algeria (ea. 6000 B.c.). According to his reconstruction of
the table of survival, the average lifespan was only around 21 to 22 years,
which presupposes, for a stationary population, a raw death rate of 46 to
47 per thousand. However, it is important to state that Biraben used a
method, classification by age groups according to the state of cranial su-
tures, that produces lower results than the reality it targets. Moreover,
Biraben himself thinks that, all things considered, his osteoarchaeological
PALEODEMOGRAPHY IOI
data suggest not,a stationary population with a very short lifespan, but a
different demographic model: an average lifespan of 25, an annual growth
rate of 0.5 percent, and the periodic occurrence of catastrophes. 47
In demographic studies based on the state of fossilized bones from the
Greek world, there has been general reluctance to calculate average lifespan
in the strict sense because of lack of confidence in the data on prepubescent
mortality. All of Angel's numbers on this provide only the average age at
the moment of death of persons who have reached adulthood. He tries to
determine the length of adult life, not what modern demographic statistics
call the average lifespan or life expectancy at birth (e0 ). In modern life
tables, his numbers correspond approximately to life expectancy at age 15
(e15)•
According to Angel's revised (1972) paleodemographic picture, the av-
erage length of adult life reached 32 for men and 24.9 for women in the
Mesolithic period (based on 71 specimens), 33.6 for men and 29.8 for women
in the Early Neolithic period (259 specimens), and 35.7 for men as against
28.2 for women in the Late Neolithic period (40 specimens). The Meso-
lithic bones come from various parts of Europe, the Early Neolithic ones
from C.::atalHiiyuk and Nea Nikomedia, and the Late Neolithic ones from
Kephala (the island of Kea). 48 The most representative and homogeneous
sample of bones, relatively speaking, for these very early periods is the
series of almost 300 skeletons unearthed by James Mellaart at C.::atalHiiyiik
in Anatolia. 49 The adults whose bones were found in this Neolithic ne-
cropolis (ea. 6500-5700 B.c.) died on the average at age 31.5 (men, 34.3;
women, 29.8). The numerical proportion of infants' (ages 0-4) to chil-
dren's (ages 5-14) to adults' (15 years or more) tombs is 7:4:10, which
bespeaks considerable infant mortality. Angel stresses the demographic
improvement since the Mesolithic period. The increase in the average
duration of life is not very large, but even so it reveals a change in the
conditions of life, especially since women benefit more than men. It seems
reasonable to attribute this enhancement of survival to the greater security
offered by sedentary village life. 50
From the classification into age groups established by Angel, 51 I have
constructed a table of survival and calculated that, assuming a demographic
situation fairly close to a stationary state, the average lifespan (e0 ) for the
Neolithic population of C.::atalHiiyiik might have been around 23 or 24
years (both sexes). By this hypothesis, the raw death rate exceeds 40 per
thousand. If we return to Angel's table, the average age of deceased adults
in the Early Bronze Age was 33.5 for men and 29.6 for women (estimated
on the basis of 400 specimens); in the Middle Bronze Age it reached 36.7
for men and 31 for women (183 specimens); in the Late Bronze Age, 39.3
for men and 32 for women (286 specimens); and in the Iron Age, or more
exactly near the end of the twelfth century B.C., 38.8 for men and 30.4 for
women (164 specimens). 52
The best osteoarchaeological series for these periods is one of about 500
skeletons from the necropolis of Karata~ in Lycia, the burial ground of the
inhabitants of a fairly typical village of the Early Bronze Age (ea. 3000-
2000 B.c.). The adults in this sample died on average at age 31.7 (men, 34;
women, 30). The proportion of infant~' to children's to adults' tombs
seems to be 6:5:10. Actually, skeletons of small children are rare and the
proportion given is an extrapolation on the basis of finds in a well-preserved
portion of the necropolis. 53 According to more recent excavations, the
proportion of infantile (ages 2-14) to adult deaths is 4:10, which is the
Hiiyiik and better than in the Mycenaean tombs at Lerna. 54
same as at <_::atal
Relatively speaking, the situation elsewhere was by no means as good. So,
for instance, the length of life was appreciably shorter among the inhabi-
tants of Aghios Cosmas (Attica) at the same period. 55
Excavations in the Mycenaean parts of Lerna in the Argolid uncovered
234 skeletons, of which 35 percent are less than 5 years old, 21 percent are
children from 5 to 15 years old, and 44 percent are adults (proportionately,
8:5:10). By comparison with the populations of Neolithic hunters and
villagers, this proportion suggests an increase in the mortality of little
children and at the same time a decrease in the mortality of bigger ones.
As for the adult inhabitants ofLerna in the Middle Bronze Age (ea. 2000-
1600 B.c.), death overtook them on the average at age 34 (men, 37; women,
31). As was said above in a discussion of the number and density of this
population, each adult woman had an average of 5 childbirths. According
to the demographic model suggested by Angel on the basis of these data,
the average period of fertility for a woman lasted 12 years, and the birthrate
seems to have been a little less than 45 per 1,000, while the death rate was
slightly higher than 40 per 1,000. Despite significant infant mortality (of
1,000 births, more than 300 died before the age of 5), this Helladic popu-
lation sustained noteworthy growth. 56
If we insist on figuring the average lifespan according to principles that
are in force for populations whose essential parameters are precisely known,
for the ancient inhabitants of Lerna we would obtain an extremely low
figure for e0 because the very high death rate for children has a powerful
effect on the average. It would create an essentially false impression of an
overall decline in longevity between the Neolithic period and the Middle
Bronze Age. Even though, strictly speaking, life expectancy at birth prob-
ably did decline because of the accumulated risks of the early years (risks
that were probably due to infectious diseases), life expectancy is on the
upswing once the age of 5 is attained.
The royal graves in the two circles of tombs at Mycenae (ea. 1650-1450
B.c.) offer an especially welcome opportunity to compare the length of life
for common people who lived in the Mycenaean Argo lid, the people of
Lerna, with analogous data for aristocrats of the same region toward the
end of the same era. Judging from the bones of the princely families that
PALEO DEMOGRAPHY 103
are in the Archa~ological Museum at Nauplia, the average age at time of

death was not more than 35.9 for the men of this ruling class, while women
treated to the same funeral honors reached age 36.1. Deeds of warfare
exposed these chieftains to mortal danger so often and so early that their
average lifespan was equal to that of their subjects, if not shorter. As for
the ladies of the Mycenaean aristocracy, they profited from their material
well-being to the extent of living longer than both their lower-class coun-
terparts and the men whose social preeminence they shared. Nevertheless,
it is important to realize that to the 22 male skeletons of this princely
sample, there correspond only 5 female skeletons, which is far from a
representative number. 57
Robert P. Charles, combining his own conclusions about the age of
Helladic skulls in Argos with the older observations of Furst, arrived at
results that favor a much rosier picture for the Middle Bronze Age than
Angel's for the analogous osteoarchaeological series at Lerna. Of 44 speci-
mens examined, a little more than a quarter died before puberty. The
death rate slows considerably between 15 and 30 years of age, only to
accelerate vigorously thereafter, leaving few the chance to live beyond their
forties. Longevity is greater, Charles writes, in the men than in the women:
the maximum death rate is about 40 years of age for the former and about
35 for the latter. 58
According to Charles's osteoarchaeological investigations in Crete, the
average lifespan there was markedly higher than in continental Greece, at
least in Minoan times. Among the Cretans of the Middle Bronze Age, the
highest mortality among adults is at age 50 or thereabouts (between 50 and
59 in men, and between 40 and 49 in women). That tends to associate,
Charles asserts, "the state of health of the island's populations with those
of the Middle East and to distance it even more than Greece from the
deplorable conditions in western Europe. " 59 Although it is altogether
likely that life was not only more pleasant but also, on average, longer in
Crete than on the Peloponnesus or in Attica, the figures proposed by
Charles seem to me too optimistic. Study of a Bronze Age population in
Rumania yields an average lifespan of 25, with 30 percent dead before
adulthood and only I percent surviving to age 60. 60 In Anatolia, research
by M. S. Senyiirek fixes the average age of deceased adults in the Bronze
Age at 35 for men and 28 for v.romen. 61
For Greece in classical times, Angel's work (1973) gives 41. 7 years as the
average age of adults at the moment of death, specifically, 45 years for men
and 36.2 for women (based on 146 specimens), or in a recent report (1983),
44.1 years for men and 36.8 for women (with the number of specimens
increased to 230). The proportion of infants' to children's to adults' tombs
is 5:3:10. Most of the adult bone specimens come from Athens and Cor-
inth, while the calculation of the proportion between children's deaths
and adults' is based mainly on results of excavations at Olynthus. 62 These
figures show a general improvement of health conditions in continental

Greece around the seventh century B.C. The average lifespan has length-
ened, but impressively so only for men. From the Neolithic period to the
Middle Bronze Age, males reaching adulthood died about 4-6 years later
than females. By the end of the Bron~e Age and in archaic times, the
difference became even greater, around 7-8 years, and it did not diminish
in the classical period. This higher mortality of women between the ages
of 15 and 40 was essentially due to the harmful results of pregnancy, child-
birth, and breastfeeding. Variations in it attest to the influence of social
factors on the biological risks inherent in female reproductive functions. 63
In our present state of knowledge, available osteoarchaeological materi-
als provide only an approximate overall estimate of life expectancy at birth
in the classical Greek world. In my opinion, the results cited above support
a hypothesis that the Greek populations living in these times should be
placed between level 15 and level 20 on the typical life tables established by
the United Nations. 64 Accordingly, the life expectancy varied between 27
and 30 years. It is very likely that the latter figure was reached and surpassed
around 600-500 B.c., the time when general health was at its best in the
history of ancient Greek society. At that time the death rate probably went
down to 30-33 per 1,000. But in "natural" biological conditions, when
there is no possibility of a truly effective struggle against the most deadly
diseases, a demographic situation of this sort could occur only at intervals
and for relatively short stretches of time.
The decline in average lifespan probably began in the fifth century B.C.
and became apparent in the fourth century. For comparison's sake, I can
cite data for Hellenistic and Roman times that were gathered from Greek
bones by the same techniques as were used for more ancient periods.
Around the fourth and third centuries B.c., the average age at death for
adults is only 42.4 in men and 36.5 in women; by the second century A.D.,
it is about 38 years of age (men, 40.2; women, 34.6). It is worth noting
that the decline, especially in its early stages, affects female longevity much
less than male. This demographic drop is a lasting one. Not until the
second half of the nineteenth century do demographic indices reach the
levels of the classical age.
The osteoarchaeological methods used to arrive at average lifespans of
the past are open to criticism. In the first place, the number of specimens
examined is still relatively small. They are scattered in time and space, and
by looking at individuals born over a relatively long time-span as though
they were a single generation, we open ourselves to error. Furthermore,
statistical calculations presuppose a fairly stable and closed population,
which was certainly not always the case. Determination of age by the state
of cranial sutures and wear and tear on teeth has risks, especially toward
underestimation. Finally, there is no guarantee that the population of
corpses in a ceme.,.teryis always an accurate statistical reflection of the living

population that uses it.
All these criticisms, and many more as well, have a good deal of weight
to them. To counter them and for reassurance, I can point to the conver-
gence of results and insist on the reality of the trends rather than on the
values themselves. And for historical times, it becomes very interesting to
compare data obtained in this way with data available from written
documentation.
Determination of Average Lifespan

from InscriptionalEvidence
From the writings of historians, lexicographers, and biographers, infor-

mation can be culled about the longevity of numerous Greek personages
from the classical, Hellenistic, and Roman eras. Although this information
constitutes an impressive mass of pertinent data, it is of no value whatever
for determining average longevity. The process of selection operating in
the writing down and transmission of the data renders the resultant statis-
tical sample unrepresentative of the population as a whole. At most, we
could try to use it to get a notion of the death rate among the oldest
members of the most privileged classes.
By consulting biographical documents, we remain prisoners of the an-
ecdotal approach and are deprived of a demographic perspective. So his-
torians of the ancient world have tried another approach that is at first
sight more promising: statistical analysis of epitaphs. Very elaborate dem-
ographic studies have been made of Latin funerary inscriptions, but rela-
tively little has been done with this aspect of Greek epigraphy. The reason
is that Latin epitaphs generally include the age of the deceased, while
Greek ones do so only randomly. Age specification became customary only
under Roman influence. For the Greek material, the monograph of Bessie
Ellen Richardson published in 1933 is still the main reference work. 65 She
collected the data on 2,022 Greek tombstones that specify the age of the
deceased. The series of epitaphs stretches from the classical period to the
beginning of the Middle Ages, with a clear majority from Hellenistic and
Roman times. B. E. Richardson distributed the cases into five-year age
brackets and calculated first the percentage belonging to each bracket, then
what she calls "average expectancy of life." According to her, this expec-
tancy is as high as 29.4 years in her sample. 66 This figure was arrived at by
dividing the sum total of ages by the number of individuals. So it is the
arith1netic average of the ages at death and should approximately corre-
spond to life expectancy at birth. Richardson did not provide separate
statistics for the two sexes, perhaps because that would have made still
more plain the strange distribution of frequencies in her sample. Her book
also has an appendix containing a catalogue of all the inscriptions used,
which enables us to complement the transverse study she made with an
attempt at a longitudinal one. In order to be able to compare this sample
with the results of osteoarchaeological _research, I first submitted it to the
method by which Angel calculated his averages. In Richardson's sample,
the average age of adults at the moment of their death is only 36.5. This is
a lower number than the one obtained from the osteoarchaeological rec-
ord. I have tried to concoct life tables; the estimate of average life expec-
tancy at birth (e0) by this method is 27.4 years.
With this e0 value, the tables of male and female mortality fall into level
15 of the United Nations typical life tables. However, comparison of the
mortality quotients of the epigraphic series and of the typical tables brings
to light various serious anomalies-in fact, they are demographic impossi-
bilities. In Richardson's sample, the death rate is extraordinarily high be-
tween 10 and 20 years of age, too high between 20 and 35, and much too
low before 5 years of age and after 55. No real population can provide a
curve like this one, especially if it is supposed to represent death rate as a
function of age over a period of time long enough to temper the effects of
catastrophic events.
Our calculations presuppose that all the individuals in this series belong
to the same generation of a relatively closed and stable population. Even
though it is a false assumption, the results of the calculation of death rate
quotients could not be so patently atypical unless the sample itself were
utterly unrepresentative of the real populations from which it was selected.
Numerous factors could be the cause of such deviation: the inadequate
number of available inscriptions; their excessive scatter over time and space,
which in turn mixes data about populations with differing demographic
characteristics (several epitaphs in Richardson's catalogue are in Greek but
contain the names of barbarians or persons from non-Greek lands, espe-
cially Italy and Egypt); funerary beliefs and customs (such as relative indif-
ference at the death of infants, or the need to erect a status symbol at the
death of a young wife); the social selectivity of the sample; and so on.
Since Richardson's monograph appeared, there has been an effort under
way to mitigate the heterogeneity of her sample by grouping epigraphic
data and analyzing them according to better-defined geographical catego-
ries and shorter historical periods. For example, funerary inscriptions of
the fifth and fourth centuries B.c. put the average age at death between 29
and 30 years of age, 67 while those of Greece under Roman occupation put
it at about 29 years 68 and those of Greco-Roman Egypt at about 24 or, if
analyzed by locale, between 22 and 40. 69 For the Roman Empire, age at
death varies from province to province between 21 and 47 years, with the
lowest figure in Rome itself and the highest ones in the African colonies. 70
Seasoned demographers have recently denied that these results are rep-
resentative of ancient demographic realities, and they characterize any at-

tempt to determi'ne the death rate from funerary inscriptions as illusory.
Experience has shown that even under conditions much more favorable
than those prevailing in Greek and Roman epigraphy-for instance, when
a relatively high number of inscriptions from a single cemetery is to hand-
the results arrived at are generally absurd. The distribution of ages inscribed
on tombstones, according to Louis Henry, teaches us more about how
funerary honors vary with the age of the deceased than about the prevailing
death rate in a given era. 71
With this methodological obstacle in mind, let us return to the classical
Greek world and two literary documents. The first is Aristotle's advice that
the ideal age for marriage is 37 for a husband and 18 for a wife, a difference
of 19 years that is desirable ''in order to synchronize the cessation of sexual
activity. " 72 The second is Thucydides' assertion about the age structure of
the Athenian populace on the eve of the Peloponnesian War: "The Ath-
enian youth was more numerous than at any other time in history. " 73
These young people were then decimated by the war with Sparta, the
plague, and the Sicilian expedition. Was the situation much different a
century later? The question is all the more interesting since for the last
quarter of the fourth century B.C. lists of adolescents (epheboi)and arbitra-
tors (diaiteto{))that is, of hoplites at the first and last phases of their service,
at ages 18 and 60 respectively, have come down to us. Around 330-325 B.C.,
there were approximately 500 adolescents and 100 arbitrators. 74 According
to the typical life tables, in a population where for every five men between
the ages of 18 and 19, only 1 survives to age 60, the life expectancy at birth
is extremely low, around 22 or 23 years. Admittedly, one should hesitate
to generalize from such a picture, since the sample in question is a portion
of the inhabitants of Athens who were subject to high risk in the prime of
their lives.
Old Peopleand Maximum Longrnty

Average longevity, otherwise known as life expectancy at birth, is the
average age reached by a group of persons born in a given period and
living in a given environment. I have noted the way it varies over the
history of the Greek world, interpreting its fluctuations as an important in-
dicator of changes in health conditions. As such it is an ecologicallongevity
(Bodenheimer, 1938) that does not coincide with physiologicallongevityor
maximum longevity. 75
Ecological longevity is determined above all by the
death rate prior to the beginning of aging. In ancient populations it is
almost unaffected by death rate quotients in age brackets above the fifties.
Although old people were relatively rare in the past, they were often
remarkably active. In our present state of knowledge, their scarcity forbids
any study of the biological and social impact of old age in prehistoric
societies. In the classical and archaic periods, great respect was shown to
men in their sixties. According to the Iliad) in Nestor's time "two gener-
ations of mortal men had perished, / those who had grown up with him
and they who had been born to / these in sacred Pylos, and he was king in
the third age. '' 76 This septuagenarian king fights alongside young warriors
beneath the walls of Troy, and although he no longer takes part in athletic
contests ("my limbs," Homer has him say, "are no longer steady ... nor
my feet, neither / do my arms, as once they did, swing light from my
shoulders"),7 7 he excels in counsel and in moderating impetuous behavior.
He has the role of a symbol in the epic. In the political system of Greek
city-states, true gerontocracies are not uncommon. It is enough simply to
recall Solon at Sparta and the institution of thegerous{a(the senate, literally
the council of elders) as well as the role played by the diaiteto{in the public
life of Athens. In a famous speech that Thucydides puts in his mouth,
Pericles at age 60 makes the perernptory assertion that he is "still in the
prime of life. " 78 Nicias, a sage in his sixties, declares before the Athenian
Assembly that Alcibiades at 36 or 37 is too young to command the expe-
dition to Sicily79 -this at a time when, by Thucydides' own account, the
population of young people at Athens was larger than ever. 80 The existence
of flourishing old people is also well-attested in the theater and in classical
iconography.
I define physiological longevity as the length of life usually attained
when the principal causes of death other than aging are ruled out. Its limit
is reached when death from old age wins out over premature or accidental
death. Modern statistical and physiopathological observations place phys-
iological longevity at around age 80. 81 It is a more biological than social
parameter that does not seem to have changed since protohistorical times.
The Bible considers 70-80 years as the normal span of human life. 82 Ac-
cording to Herodotus, for the Persians in the time of Cambyses II, ''The
longest life a man can hope for does not exceed 80 years,'' and although
the Ethiopians, as Cambyses' spies explained, live to age 120, that is only
due to the miraculous effects of a fountain of youth that existed in their
part of eastern Africa. 83 Finally, the assertion that the human lifespan is an
ideal number, nine squared, is attributed to Plato. 84 To be sure, experience
taught that under exceptional circumstances one could live beyond one's
eightieth or even ninetieth birthday. Prodigious ages were attributed to
Thales, Pythagoras, Democritus, Xenophanes, Hippocrates, and, perhaps,
advisedly so, to Gorgias. 85 Does a truly insuperable limit exist, and if so,
what is it? According to modern demographic statistics, maximum longev-
ity, which is the highest age actually attained, does not exceed n2 years. It
is highly likely that man never has and, except by techniques as yet un-
known, never will surpass a limit that demographic calculations put at
age n5.86
Some authors of the Empire, such as Pliny and Censorinus on the

Roman side or Phlegon Trallianus and Pseudo-Lucian on the Greek, in-
quired into the maximum longevity of men in their own time and in the
past of their civilizations. By our lights they are all very credulous, espe-
cially Pliny, although, strange to say, not more so than some of their
modern successors. In any case, these ancient authors deserve our esteem
for having collected interesting gerontological documents, especially in the
treatise Vitarum auctio that passes for the work of Lucian of Samosata
(second century A.D.). To demonstrate the existence of famous nonagen-
arians in classical Greece, it is enough to mention the rhetorician Isocrates
(98 years of age), the poet Philemon, the philosopher Cleanthes, and, not
least, the tragedian Sophocles. There were famous centenarians as well:
the musician Xenophilos, perhaps the historian Hieronymus of Cardia,
the writer Alexis ofThurioi, and most likely the sophist Gorgias of Leontini
(who lived 105 or perhaps even 109 years). On Greek funerary inscriptions
the oldest age attested is 110. 87 Egyptian storytellers in Pharaonic times
were in the habit of saying that human life could not be prolonged beyond
110 years except by means of magic, and even then for only a single addi-
tional year. 8 ~ The astrologer Epigenes placed the limit of human life at 112
years. 89
Variations in Height and Some Consequencesof Malnutrition

In prehistoric times prior to the introduction of agriculture, the average
height of inhabitants of the eastern Mediterranean was relatively great,
exceeding 170 cm. in men and 158 cm. in women. It begins dropping in
the first villages of the early Neolithic period: the men at ~atal Hiiyiik
were about 169 cm. tall and those at Nea Nikomedia about 168 cm. In the
Late Neolithic period, average height stabilizes between 166 and 167 cm. in
men and between 153and 156 cm. in women. These figures, arrived at from
the dimensions of long bones, do not change again until the classical
period. By the seventh century B.C. a discernible increase has begun (men,
170 cm.; women, 156.3 cm.) that continues into the Hellenistic age (men,
171.8 cm.; women, 156.6 cm.). 90
Height is a phenotypic expression of the genetic program. As a result,
average height is conditioned as much by hereditary racial traits as by the
influence of environment during an individual's growth. Among external
factors the most important is nutrition-the quality, quantity, and regular-
ity of food-especially in early childhood. The height of the inhabitants of
the Greek world seems to have been determined above all by external
factors. It should be stressed, for instance, that the Inda-European inva-
sions had no noteworthy effect on the average height of the populations
derived from the mix of newcomers and autochthons. However, it did
modulate negatively at the time of the changeover from hunting and gath-
ering to agriculture, and positively when the quality of life improved as a
result of the development of industry and trade.
The stability of average height should not blind us to the importance of
individual variations. The gap between shortest and tallest was remarkably
large, especially from the Bronze Age on. It reflects both the genetic
polymorphism of Greek populations as well as strong social differentia-
tions. As an example, we can look at the situation in the Middle Bronze
Age. The inhabitants of Lerna at the time averaged 166.3 cm. (men) and
154.2 cm. (women) in height, those of Pylos 167 cm. (men) and 152 cm.
(women), and those of Kato Zakro (Crete) 167 cm. (men) and 157.5 cm.
(women). 91 Although these figures correspond in order of magnitude to
the body size of the people of Mycenae, the same is not true for the size
of the aristocrats of that citadel. The 14 persons of male gender buried in
the royal tombs at Mycenae had an average height of171.5 cm., and their 3
female companions exceeded 160 cm. Despite the small size of the sample,
the difference of 5 cm. in the height of the princes as against the common
people appears to be significant. 92
The inhabitants of Greece in the Mycenaean, archaic, and classical pe-
riods were thickset and sturdy, with relatively short lower limbs. The
image of their general appearance that one obtains from osteoarchaeologi-
cal evidence does not coincide with the idealized representation of the
human body in Greek sculpture. 93 Although the average man had neither
the slender grace of a statue of Apollo nor the force concentrated in one
of Heracles, he did have a supple, vigorous body attuned to the grind of
daily life.
In Greece as elsewhere, the historical variations in average height were
largely dependent on nutritional customs and possibilities. I limit myself
to a few remarks on this theme, since nutrition and its social and medical
implications are such a huge, complex subject that they merit a monograph
in themselves of no mean proportions. 94 A radical change in nutritional
customs took place during the agricultural revolution of the Neolithic
period. The new system in itself does not seem to have been superior to
the one it displaced. A decline in average height ,suggests that there was
even a deterioration in the physical conditions linked to nutrition. Actu-
ally, hunting and gathering is a very successful way of life for relatively
small human groups, 95 though it cannot adequately maintain populations
that exceed a certain density. The changeover to agriculture took place
under the pressures of a serious nutritional crisis caused by demographic
expansion. 96
On the other hand, a sedentary life reduces the variety of nutrients, and
that in itself can give rise to deficiency diseases, particularly those caused
by lack of a regular supply of proteins, vitan1ins, and certain minerals. 97
Moreover, dependence on agricultural products allows for the periodic
PALEO DEMOGRAPHY III
appearance of faqi.ine. Both the climate and the flora of Greece seem to
have protected its inhabitants from serious vitamin deficiencies like rickets,
osteomalacia, scurvy and beriberi, or the harsher forms of proteino-caloric
malnutrition like kwashiorkor. But that does not mean they did not suffer
often from lesser vitamin deficiencies and lack of essential amino acids,
conditions that markedly reduce resistance to infectious disease.
From Neolithic times, people grew wheat, barley, lentils, and peas in
Greece and raised goats, sheep, pigs, and cattle there. As early as the
Bronze Age they were cultivating olive trees, figs, and the grape, as well as
raising farmyard animals, which eventually included bees. Meat was be-
coming an ever more precious foodstuff, and most mortals did not partake
of it daily. Although game was becoming scarce, technological improve-
ments in fishing fostered growing exploitation of the wealth of the sea. So
the necessary nutritional supply of proteins came from milk, cheese, eggs,
and fish; that of carbohydrates from bread, honey, and dried figs; and that
of fats from olive oil and, less dependably, pork products.
A person's basic caloric intake consisted of bread and wine. I-Iomer and
Plato knew what they were talking about when they praised them. The
wily chieftain Odysseus so addresses hotheaded Achilles:
Rather tell the men of Achaia here by their swift ships, to
take bread and wine, since these make fighting fury and war-craft.
For a man will not have strength to fight his way forward all day
long until the sun goes down if he is starved for food. Even
though in his heart he be very passionate for the battle,
yet without his knowing it his limbs will go heavy, and hunger
and thirst will catch up with him and cumber his knees as he moves on. 98
And here is how Plato imagines the life of the inhabitants of "the first
city":
First of all, then, let us consider what will be the manner of life of men thus
provided. Will they not make bread and wine and garments and shoes? And they
will build themselves houses and carry on their work in summer for the most part
unclad and unshod and in winter clothed and shod sufficiently. And for their
nourishment they will provide meal from their barley and flour from their wheat,
and kneading and cooking these they will serve noble cakes and loaves on some
arrangement of reeds or clean leaves. And, reclined on rustic beds strewed with
bryony and myrtle, they will feast with their children, drinking of their wine
thereto, garlanded and singing hymns to the gods in pleasant fellowship; not
begetting offspring beyond their means lest they fall into poverty or war. 99
After this speech from the mouth of Socrates, Glaucon undertakes to

praise the refinements of civilization; he proffers a less harsh diet as the
requisite of perfect health:
True, I forgot that they will also have relishes-salt, of course, and olives and
cheese, and onions and greens, the sort of things they boil in the country, they
will boil up together. But for dessert we will serve them figs and chick-peas and
II2 DISEASES IN THE ANCIENT GREEK WORLD
beans, and they will toast myrtle berries and acorns before the fire, washing them
down with moderate potations. And so, living in peace and health, they will
probably die in old age and hand on a like life to their offspring. 100
In this system of reference, Plato faults the luxury of his contemporaries,

a luxury that promotes idleness and too rich a diet, as the main cause of
the city's diseases "to which the clever sons of Asclepius have attached
names like flatulence and flux. " 101
The history of medicine can accept this line of argument only with
reservations that deprive it of all its vitality: though it is true that sybaritic
luxury and culinary novelties can be blamed for an increase in some dis-
eases, especially gout and vascular failures caused by obesity and arterio-
sclerosis, the dominant pathology is one of deficiency even at the height
of the economic boom of the Greek city-states. The frequency of bladder
stones in Greek children of the fourth and fifth centuries B.C. is relevant
to this assertion. 102 Idiopathic bladder stone disease in children is a differ-
ent disease from urolithiasis in adults. It is rare in modern societies, except
those rife with poverty and nutritional deficiencies. Reports presented to
the symposium on idiopathic bladder stone disease organized by the World
Health Organization in Bangkok in 1972 stress that the incidence of the
disease in children is a direct reflection of a country's socioeconomic state
and a close correlate of malnutrition. 103
Though it has almost disappeared in western Europe and North Amer-
ica, idiopathic bladder stone disease is endemic in Turkey, Egypt, Iran,
India, and several Southeast Asian countries. Its etiology remains obscure.
It is known, however, that its cause must lie in a person's nutrition during
the first three years of life and in factors determining water metabolism.
Possibly vitamin A deficiency favors the formation of bladder stones as well
as a hot climate, but the decisive etiological factor is precocious weaning
coupled with a relatively dry and protein-poor diet. This usually occurs
when breastfeeding is disturbed because of protein deficiencies in the
mother or wet-nurse. 104
It emerges from several Hippocratic texts that idiopathic bladder stone
disease was a common ailment affecting children, especially boys, in an-
cient Greece. 105 Its symptoms are well described, artd its surgical treatment
is known and forbidden in the famous Hippocratic oath. The author of
the fourth book of the treatise Diseaseseven says that "the principle of this
disease is milk, when a child suckles impure milk. And the milk of a
mother or wet-nurse is impure when her diet consists of mucous food,
impure food and drink. " 106 Hippocratic physicians explained the forma-
tion of bladder stones as a process of sedimentation of impurities contained
in milk or water. Though false in their detail, such explanations do point
to a fundamental truth, namely, the pathological results of faulty breast-
feeding and of the improper nutrition of the wet-nurse.
The natural resources of Greek soil were mediocre at best. Aridity placed
PALEO DEMOGRAPHY II3
limits on the cultivation of grains. According to the calculations of

A. Jarde, annual consumption of wheat was about 3 hectoliters per person,
while the average output of the soil was about 10 hectoliters per hectare.
From this Jarde deduces that ''the food supply became inadequate as soon
as the population density surpassed 36 inhabitants per square kilometer. " 107
In the classical age, most of the continental portions of Greece main-
tained an equilibrium between the production and the consumption of
wheat. Only Thessaly had a surplus, since it was a relatively sparsely pop-
ulated (about 30 inhabitants per square kilometer) breadbasket. The inad-
equately supplied areas were chiefly Corinth, the Megarid, Attica, and the
island of Aegina. Their inhabitants survived thanks only to the industry
and commerce that made possible the importation of sustenance. Along
with imported grain, a new disease reached Greece from time to time and
in waves, namely, ergotism, a kind of food poisoning caused by a parasite
in rye. Rye was not cultivated in Greece itself.
Traces of malnutrition on Greek bones have not been studied suffi-
ciently. A first desideratum is the radiological determination of Harris'
lines on the long bones from a Greek necropolis. Microanalytic determi-
nation of the minute variations in the elementary composition of bones,
specifically the relation between the amounts of strontium and calcium, is
beginning to give us clues about the level of meat consumption among
ancient Mediterranean populations (S. C. Bisel, 1980; G. Fornaciari et al.,
1982).
Paleo-odontowgy
Abrasion of the hard tissues is a very common special feature on the

anatomical crown of teeth. It is seen among adults of every human popu-
lation, though there are important statistical variations with respect to the
earliness of its first appearance, the swiftness of its progress, and the seri-
ousness of the final lesions. It can range from mild wear and tear on the
enamel, hardly more than a physiological phenomenon, to the total abra-
sion of tooth enamel on the occlusive surface with partial destruction of
the dentine and, in the most serious cases, disclosure and secondary infec-
tion of tooth pulp. Its etiology is essentially mechanical, a function of the
hardness of the substances that the teeth grind up. So the degree and the
frequency of dental abrasion can offer dues about the kind of foods that a
given population eats.
According to several authorities, dental abrasion is chiefly attributable
to the presence of particles of sand in food, especially in flour milled by
stone and foods ground up in stone mortars. However, nowadays it seems
more likely that this factor, although assuredly not negligible, is less im-
portant than just the slow, steady way people chew food of a certain firm
or rubbery consistency. In this kind of chewing, the teeth themselves wear

each other away.
Dental abrasion, which is very pronounced in Paleolithic teeth, does
not appear to decrease in the Mesolithic period or at the beginning of the
Neolithic period. 108 For these remote times, a true statistical analysis of the
frequency of this lesion is not possible, at least at present, since our knowl-
edge is based on such small samples. As for more recent periods, the
number of specimens examined is becoming significant, although a diffi-
culty remains. How can the figures given by different authorities be com-
pared when we do not yet really know how to measure the parameter in
question, with the result that we are basically dependent on subjective
estimates?
Dental abrasion is very marked on Neolithic human teeth from Iran and
in Europe, among the Sumerians, and among the inhabitants of predy-
nastic and Pharaonic Egypt. In the latter, considerable abrasion occurs
throughout the social ladder. M. J. Becker has reported very marked ab-
rasion on anatomical crowns in 74 persons exhumed in 1973 at Kato Zakro
(a necropolis dating from Middle Minoan times). According to him, such
extensive wear and tear is due to the use of stone mills and the consump-
tion of coarse grains. Furthermore, this Cretan populace, whose cultural
development was relatively high, suffered from tooth decay-in fact, al-
most everyone over the age of 40 was toothless. 109 Robert P. Charles has
described a male skeleton aged 30-35 found in a protogeometric tomb (ea.
900-850 B.C.) from the Argive necropolis (I6 A13.). This person had a
complete and perfectly preserved set of teeth, free from decay, but all of
them, both uppers and lowers, were "very worn. "ll 0 I cite these two
reports as examples. They are instructive, but, statistically speaking, in
themselves they are not significant. The authors who tell us that dental
abrasion was ''very marked'' in Minoan and archaic peoples rarely produce
figures and fail to specify whether the level of abrasion they speak of is in
comparison with Neolithic teeth, those of classical antiquity, or those of
modern times. They are probably right if their frame of reference is the
dentition of the current inhabitants of western Europe. But if we keep the
phenomenon in its own historical context, our judgment will be more
cautious: dental abrasion in the inhabitants of Bronze Age Greece is plainly
less than that of the first farmers of the Neolithic period, and it is probably
still less than that of Greeks in the classical period. At least, that is the
upshot of John Lawrence Angel's studies, and his statistics have the advan-
tage of being based on the judgment of one and the same researcher, a
factor that eliminates differences in classification criteria and neutralizes
subjectivity in their use.
Angel distinguishes between five stages of dental abrasion: absent, slight,
medium, pronounced, and very pronounced. According to him, 50 per-
cent of adult skeletons at Nea Nikomedia (Early Neolithic period) fall into
the latter two ca~egories, while the proportion is 30.8 percent at Karata~
(Early Bronze Age), 28 percent at Lerna (Middle Bronze Age), and 40
percent overall for the sum-total of archaeological sites in classical Greece. 111
However, this picture is complicated to some extent by the fact that study
of the skeletal remains at (;atal Hiiyiik (Early Neolithic period) finds only
30.8 percent of the skulls with pronounced or very pronounced dental
abrasion. 112 As for the figures on classical Greece, they derive from the
examination of the remains of 87 individuals without common origins.
Given the interest we have in any token of the evolution of living condi-
tions from the seventh to the fourth century B.C., at some future time it
would be useful to undertake a detailed analysis of the classical evidence
that would take account of differences in the age and provenience of each
specimen. It is hard to accept the notion that the sophisticated citizenry
of Athens, Corinth, or Sybaris at their height suffered from dental abra-
sion. In the Middle Bronze Age, the aristocrats at Mycenae were much
less prone to it than the common people. Scarcely 15 percent of the skele-
tons from the royal tombs evince pronounced or very pronounced abra-
sion, while the number rises to 28 percent among the rural population of
the Argolid. 113 Did a similar difference exist between the Athenians' teeth
and those of the Spartans?
Whatever the case may be, the jump from 28 percent among the Myce-
naean inhabitants of Lerna to 40 percent in the Greeks of the classical
period does not actually signify what it seems. Across the sample from
Lerna, strong dental abrasion is much more common in men (38 percent)
than in women (15 percent). This is partly due to the greater force of the
chewing muscles in men and partly to the fact that the average age of the
women in the sample is 6 years less than that of the men. 114 Probably the
classical sample favors its masculine component over the feminine one.
Dental abrasion increases with age, and the only valid comparison is be-
tween equivalent age groups. However, the average age in the Mycenaean
sample (35 years) is 6 years younger than that of the classical one (41). 115
The bread eaten in the Bronze Age and in classical times must have been
much less hard on dental enamel than the grain eaten in the Neolithic
period. Around the sixteenth century a new change in diet in Europe
affected the way people chew: from then on, dental abrasion subsides, but
dental caries increases. The fl!ral population of Greece has lagged behind
in this overall change. Angel states that about a third of modern adult
Greek males (average age, 38 years) have n1arkedly worn down the grinding
surfaces of their teeth.
It is often-but falsely-said that dental caries is a "disease of civiliza-
tion.'' The presence of its typical lesions can be confirmed on the teeth of
Australopithecine man. 116 There is no human population that has com-
pletely escaped it. Nonetheless, it must be admitted that dental caries is
much more frequent nowadays than it was in the past. In Europe, its
frequency begins mounting in medieval times and markedly accelerates in

the modern period. 117 There is no doubt that the fundamental cause for
this increase is progressive change in eating habits. Probably sugar and
white bread are the single most important factors in the current high rates
of caries, since they actually stimulate the action of the microbes in dental
plaque. But so simple an explanation cannot possibly suffice. The etiology
of tooth decay is very complex and for the most part still unexplained.
The foods eaten are significant at various levels, since they supply the
proteins, vitamins, and minerals necessary for the proper formation of
dental tissues, maintain metabolic equilibrium, determine the mouth's
acidity and the creation of dental plaque, elicit local mechanical activity,
and so forth. It is clear that poor people's food and rich people's food
both have their advantages and disadvantages. In any case, given a mini-
m um supply, an excess of food seems to have more damaging conse-
quences for dental tissue than a lack of it. As a rule, the well-to-do have
more decayed teeth than the poor.
Factors other than diet are implicated in the pathogenesis of dental
caries, such as the genetic determination of dental microstructure, the
physiopathology of salivation, climatic factors, the mineral content of
drinking water, the evolution of microbial flora, social and individual hab-
its regarding dental hygiene, and so forth. This plurality of factors favoring
or curtailing dental caries means that variations in its frequency from one
population to the next are difficult, if not impossible, to reduce to a
common model.
Among the inhabitants of ancient Greece, the incidence of caries is
distinctly greater than in analogous populations of western Europe and in
Egypt. For the totality of European and North African populations in the
Mesolithic period, its frequency nears 7 percent of the teeth examined. For
the necropolis of Taforalt, the figure is around 6 percent. Later on, the
incidence of this disease falls considerably in the European, North African,
and Asiatic populations studied, but not in Greece, where it appears to
increase in the Early Neolithic period, only to subside and then remain
steady until the classical period at about 6 decayed teeth for every roo
examined. 118
In comparing published results, it is essential to take into account differ-
ences in reference systems. Some rates refer to the number of teeth actually
examined, others to the number of teeth likely to have existed on extant
jawbones and individuals. The rate based on teeth actually examined un-
derestimates the real frequency of the disease, since it neglects lost and
destroyed teeth. Accordingly, the analysis of teeth must be complemented
by an analysis of jawbones. The sum-total of decayed and missing teeth,
expressed either as a percentage of "possible" teeth or as the average
number of dental lesions per mouth, overestimates the frequency of dental
canes.
PALEODEMOGRAPHY 117
For ancient G_reece, in round numbers, the rate of decayed teeth as

against teeth actually examined is a little more than 6 percent; the rate of
decayed and lost teeth as against "possible" teeth from extant tooth sock-
ets is as high as 20 percent. This latter number corresponds to a figure of
6.5 lesions per mouth. But there were significant local variations. The
aristocrats at Mycenae had only r.3 lesions per mouth, of which o.8 were
actually instances of caries, while their contemporaries at Lerna had close
to 7 lesions per mouth, more than 2 of which were actually caries. 119
According to C. M. Furst, dental caries was common in skulls from Argos,
but R. P. Charles reports its rarity at the same site. 120 The disease was
unequivocally common among the Cretans of the Minoan era, especially
at Knossos (9 percent caries in a survey of 1,500 teeth actually examined)
and at Kato Zakro. 121
These local variations are upsetting insofar as an undeniable improve-
ment in the conditions of life and especially in diet increased the incidence
of dental caries in the palaces of Crete but drove it down in spectacular
fashion among the warrior princes of Mycenae. In their case it is tempting
to postulate the existence of hereditary factors. Contrary to what one
might expect, the urban way of life of the classical period brought on no
deterioration in Greek teeth. In fact the frequency of dental caries was less
than before, though it remained relatively high compared with the situa-
tion elsewhere in Europe as a whole. Angel's initial research on Greek
teeth in this era resulted in the follo\ving statistics: 5 percent of the teeth
were actually decayed or lost, which corresponds to 4.5 lesions per mouth.
New paleo-odontological specimens have brought down this latter figure
to 4 lesions per mouth, with just o.8 of the teeth in it actually decayed. 122
In Bronze Age Cypriot tombs typical cases of dental caries have been found
(at times resulting in osteitis of the jawbone, as in 9A J(ition) as well as very
marked dental abrasion, to the extent of laying bare tooth pulp. 123
It is hardly likely that the relatively high frequency of caries among the
Greeks is attributable to racial traits. For instance, there is no caries on any
of the numerous human teeth from the Greek-Etruscan necropolis of
Spina (fifth to third centuries B.C.) in northern Italy, nor on those of the
Greek colonists at Pithekussai (eighth to seventh centuries B.c.) in south-
ern Italy . 124 Among the external factors that could favor dental caries
among the inhabitants of protohistoric and classical Greece, I should men-
tion the relative abundance ot sugar in their diet. To be sure, we are not
referring to refined sucrose, which was still unknown, but to honey and
fruits high in sugar content, chiefly fresh or dried figs. In asking himself,
''Why do figs that are ripe and sweet i1npair the teeth?'' the author of an
Aristotelian R-oblemwas handing on an old empirical observation. 125
In modern times, it seems that the population of Athens enjoys some
resistance to dental caries. According to recent research, this is due to a
high concentration of fluoride and still more to a weak concentration of
selenium in the teeth. Chemical analyses of several undecayed teeth from

Athenian tombs of the eighth and eleventh centuries B.C. reveal that even
in remote times fluoride was abundant and selenium did not surpass a
critical threshold. Although drinking waters in Athens are not and doubt-
less never were rich in fluoride, a supply of it is guaranteed from sea salt
and the consumption of fish. 126 There is the report of a modern endemic
of fluorosis at Laurium in Attica caused by an excess of fluoride in the
drinking supply. When Henri Duday found a predominance of mottled
enamel (enamel hypoplasia) in the dental pathology at the Kitsos Cave, he
speculated on such a phenomenon as its etiology, given the fact that the
cave itself was situated precisely in the mining district of ancient Laurium.
So a hypoplastic tooth was analyzed by X-ray diffraction, with the result
that the possibility of fluorosis was excluded.
Such linear hypoplasias of tooth crowns are especially interesting for the
paleopathologist. Actually, their significance is analogous to that of Harris'
lines. Appearing as horizontal furrows parallel to the free edge of the
tooth, they represent defects in the generation of enamel, arrested growth
processes. They testify to attacks on the organism in early childhood: acute
infectious diseases or chronic ones with acute crises, such as childhood
fevers, gastrointestinal troubles, and famine. Although it is not possible to
distinguish among their possible causes, these hypoplasias are valuable
indicators of the overall state of health of a given population. So in the
case of the Neolithic inhabitants of the Kitsos Cave, Duday came to the
conclusion that they must have "paid heavily for the illnesses of early
childhood, inasmuch as the skeletons of young people are relatively nu-
merous (10 children of the 18 individuals enumerated) and inasmuch as the
teeth of older persons bear the marks of severe, repeated childhood affiic-
tions. '' 127 Arrested enamel growth is relatively rare among the Neolithic
inhabitants of <_::atalHiiyiik, but abundant in the most recent population
of Lerna. Comparative study of these phenomena is hampered by the lack
of an objective, accepted method for quantifying them.
Chapter EJur
COMMON PURULENT
INFLAMMATIONS
Any wound that breaks the continuity of the skin tends to produce pus
and to be complicated by pathological manifestations in originally un-
harmed body parts or even the whole organism. Such events are caused by
infection, that is, by the penetration of microbes into the body, by their
particular biological activity, and by the reactions of the organism under
attack. Penetration of pathogenic microbes into the human body can also
come about by natural, as opposed to traumatic, pathways: through the
respiratory tract, the digestive tract, and so forth.
Infection is an invisible event in the strict sense of the word, since it
takes place on a microscopic level. For a very long time it was invisible to
the eyes of the intellect as well, because the almost ubiquitous presence of
pyogenic bacteria in the human environment obscured their necessary role
in purulent inflammation. The very banality of infection in its best-known
form was the reason for its elusiveness to medical science. The production
of pus appeared as a "normal" phase in the healing of wounds. 1
The Paleopathologyof NonspecificPurulent Infections

Gram-positive 2 cocci classified in the Micrococcaceae family (in particular,
Streptococcus)Staphylococcus)Enterococcus)Pneumococcus) and M icrococcus),
when found on wounded human body parts or on those with lowered
local resistance, provoke acute inflammatory reactions and the production
of pus. Upon the body's return to clinical health, these microbes continue
living peacefully as saprophytes on the skin or in the mouth, the throat,
or the intestines. The same events can take place with gram-negative bacilli
belonging to the Enterobacteriaceae family (Escherichiacoli) Proteus)J(leb-
119
siella), which are optionally pyogenic bacteria that remain in the human
organism and harm it only in exceptional cases. These bacteria are perfectly
adapted to man, who for his part has at his disposal powerful immunolog-
ical mechanisms to defend himself against their attack. In fact, the forma-
tion of pus is the expression of the mobilization of this defense system.
The mutual adaptation established between the majority of pyogenic germs
and man proves that their coexistence is indeed old.
During the last two decades, paleontologists have proved the existence
of microbial life in the Precambrian era. J. W. Schopf, E. S. Barghoorn,
and several other specialists in the microanalysis of rock have discovered
petrified remains of the cells of various kinds of microbes from more than
2 billion years ago. 3 R. Cameron and F. Morelli even believe that they
have found, deep in the soil of the dry valleys of Antarctica, frozen bacteria
capable of reproducing themselves after living in suspended animation for
several hundred thousand years. It is somewhat surprising to a historian of
biology that the current state of scientific knowledge makes it plausible to
credit such extraordinary contentions. Only a few decades ago experts
would have had difficulty accepting them. I should stress that even if
subsequent research disproves the Precambrian origin of true procaryotes
or the ability of prehistoric bacteria to revive, there can be no doubt about
the presence of many different kinds of bacteria in the biosphere during
geological periods prior to the appearance of the higher animals.
We know neither when nor how certain groups of heterotrophic mi-
crobes were specialized into parasites that disturbed the biological equilib-
rium of Metazoa. Though there are fossil microbes that present the typical
structure of Eubacteria and even correspond in their general aspect to the
Micrococcaceae, the fact that they are found at large in rocks and not
contained in organic tissue prevents us from witnessing their pathogenic
role. In any case, it is absolutely certain that the ancestors of numerous
pyogenic bacteria adapted themselves to parasitic life on higher organisms
before the appearance of mammals and therefore well before the appear-
ance of man on the planet. Man inherited pyogenic germs by vertical
transmission from his ancestors.
Roy L. Moodie described fossil bacteria in the· bones of prehistoric
animals afflicted with osteomyelitis, but doubts have been expressed about
his interpretation of the slides. 4 Even if we cannot accept the existence of
bacteria in Moodie's bone sections, there is no denying his diagnosis of
osteomyelitis, which in and of itself suffices to establish the existence of
pyogenic bacteria-though they cannot be identified with known free-
living fossil bacteria. Among the cases Moodie describes, the oldest is that
of a saurian of the Permian period in which a spine fracture is complicated
by a purulent bone infection. Traces of osteomyelitis have been reported
on a Cretaceous dinosaur, a Pleistocene lion, and so forth.
Among the hominids, an osteoarchaeological proof of the antiquity of
COMMON PURULENT INFLAMMATIONS 121
pyogenic germs is provided by the traces of alveolar pyorrhea on the lower

jaw of Sinanthropus lantianiensis (about 450,000 B.c.). 5 Periodontal ab-
scesses are also present in Neanderthal man and in fact occur in practically
every prehistoric population of Homo sapiens. Such infections arise from
saprophytic microbes in the mouth. Among the Neanderthals of La Fer-
rassie, an abscessed osteitis of the tibia and fibula has been conjectured; in
Chapelle-aux-Saints man, some have recognized the secondary infection
of a crushed phalanx; in Rhodesian man, there is a convincing paleopatho-
logical diagnosis of mastoiditis. 6 From the Neolithic period on, there are
definite traces of osteomyelitis and osteoperiostitis on various European
and African bones. Most often they are localized on the tibia and the
temporal bone.7 But the present rarity of very old and well-preserved bone
specimens makes unwise any statistical study of the possible variations in
frequency and virulence of pyogenic infections during the Stone Age.
Pyogenic germs are transmitted from one human to another directly or
by way of infected objects. They do not require an animal to serve as their
reservoir. Unlike certain viral diseases that are directly transmitted between
humans, purulent inflammations do not demand a minimal population
density for epidemiological continuity. To repeat, this perfect mutual ad-
aptation, which makes the pyogenic microbe a saprophytic companion of
mankind, is the best proof of the antiquity of the common suppurative
diseases. 8
Mastoiditis, a purulent inflammation that usually results from the exten-
sion of an inflammation of the middle ear, was relatively common in
Pharaonic Egypt. Its occurrence on American pre-Columbian skulls, as
reported by E. A. Hooton and H. U. Williams, suggests that the germs
responsible for it, especially the streptococcus and pneumococcus, had
adapted themselves to man before the first colonization of the American
continent. The results of Egyptian paleopathology are not irrelevant to a
study of ancient diseases in the Greek world, for it is plain that a germ
capable of living in equilibrium with man and not requiring special climatic
conditions or vectors or social customs to ensure its survival would of
necessity be present (or, for that matter, lacking) throughout the eastern
Mediterranean. Sir Marc Armand Ruffer proved the existence of gram-
positive cocci in the lung tissue of a twelfth-century B.c. Egyptian
mummy. 9 Since the individual died of pneumonia-his pulmonary tissue
was in the stage of hepatization-it is very likely that the cocci in question
belong to the genus Diplococcus)or at least to the family Micrococcaceae.
This is grounds for suspecting that the inhabitants of archaic Greece were
exposed to lobar pneumonia and therefore to heightened mortality during
cold and wet seasons.
Let us pass on to a survey of actual cases of common pyogenic infections
on ancient bones from the Greek world. In a man from Lerna (88 Ler.)
living in the Middle Bronze Age, traces of periostitis and of ankylosis of
the last joint on the thumb and on the little finger of the right hand are
the result of a typical V-shaped phlegmon: after a wound to the wrist or
palm, infection by pyogenic germ propagated along the synovial bursae
and the flexor tendon sheaths of the hand. Although cured of the acute
purulent inflammation-and also, by t_he way, of a fracture of his right
seventh rib-the subject was left with a slight handicap until the end of his
life (at about age 35). 10 Another man of the same era, whose skeleton was
found at Argos (123A121.))suffered from cranial osteomyelitis: on the occip-
ital bone, to the left of the inion, the external surface is deeply corroded,
and the wall has become very thin (actually, it is perforated, but as the
result of a post-mortem accident). Such damage to this bone can be a
complication from a carbuncle on the nape of the neck, which is generally
caused by a staphylococcus infection. 11 Osteomyelitis was also reported by
V. V. Bobin in 1964 on bones from Neapolis in the Crimea (Greco-Scyth-
ian necropolis, third-first centuries B.C.) . 12
In the relatively complex case of a male child from Lerna (130Ler.)) dead
at age 3, it may well be, as Angel supposes, that a mastoid infection
brought on a thrombosis of the meningeal sinus on the left side, which
caused, in turn, the development of collateral venous channels. 13 If this is
actually what took place, meningitis and intracranial hypertension were
certainly part of the clinical picture. However, other explanations (for
example, the existence of a congenital malformation) might account for
the state of this child's skull.
Traces of a purulent infection on the upper jawbone can be detected on
the skeleton of an athletic, 50-year-old chieftain exhumed from a royal
grave at Mycenae (59 Myc.). A cystic formation in the area of the upper
second molar testifies to the existence of an old abscess. His gum infection
propagated in the maxillary sinus, whose thickened bony walls bear the
marks of an inflammation of the mucoperiosteum-in other words, a
sinusitis. 14
From osteoarchaeological research it is clear that alveolar pyorrhea is a
disease whose commonness among the ancient inhabitants of Greece and
the lands adjacent to it should not be underestimated. The bones of
Aghios Kosmas, Lerna, the Kitsos Cave, and Argos offer a whole series of
instructive examples of it. 15 H. Graham Carr states that up to 80 percent
of the ancient jawbones from Knossos bear signs, however slight, of peri-
odontal disease. 16 To conclude the discussion of this aspect of bone pa-
thology, it is enough to recall one more case whose special feature is that
it belongs to the classical age: some traces of alveolar pyorrhea and a large
abscess of the jawbone are to be found on the skeleton of an Athenian
citizen (65 AI(.) who died around 450 B.C. and who was buried in the
Kerameikos. 17
A separate problem is posed by the pyogenic germs of the Neisseriaceae
family, in particular the gonococcus and the meningococcus. These gram-
negative cocci lask the ability to coexist as peacefully with the human
organism as other pyogenic germs. Unlike the Micrococcaceae and the
Enterococcaceae, the Neisseriaceae do not appear to be parasites that must
have accompanied man since the dawn of humanity and in all societies.
Did gonorrhea and epidemic meningitis flourish in Greece in early antiq-
uity?18 Osteoarchaeology cannot answer this question. Even though the
gonococcus and the meningococcus can leave traces of their pathogenic
activity on bones, such lesions are not specific to them and are indistin-
guishable from those of pyogenic microbes in general.
Ancient Explanations of the RJrmation of Pus

Pus is a thick whitish or yellowish liquid that results from the infection
of the tissues in higher animals by various kinds of bacteria. At the points
at which these bacteria enter into contact with the host organism, a local
inflammatory reaction can be observed, which ancient physicians described
very precisely in terms of its four cardinal symptoms: heat, redness, swell-
ing, and pain. These, in turn, are soon associated with the production of
pus. Microscopically, the pus consists of an exudate swimming with live
and dead bacteria, tissue waste products, and above all numerous degen-
erated polynuclear white blood cells (pyocytes or "pus globules"). Actu-
ally, the production of pus is an immunological defense, and pus is nothing
more than an accumulation of bacteria and cellular ''corpses.''
I have offered the preceding brief summary of the views of modern
medicine on the nature of pus in order to make it clear why in the evolu-
tion of medical ideas, the correct interpretation of the pathological mech-
anism of suppuration and an understanding of its proper biological role
are unimaginable before the advent of cellular pathology, microbiology,
and immunology in the nineteenth century. Although this phenomenon
was inaccessible at the microscopic level, wherein lay the key to its nature,
it was perfectly accessible on the macroscopic level and must have fasci-
nated healers at all times and places. The etymology of the words for pus
in the Inda-European languages confirms the earliness of its existence as a
concept. The Greek word puon or puos)like the Latin word pus and its
homographic descendant in French and English, derives from the same
root as the verbs puthomaiin Greek and puyatiin Sanskrit (meaning ''rot,
putrefy, decompose"). According to Chantraine, this whole family of
words stems from a protestation of disgust, pu or pit.19
In Homer, the verb puthois used on several occasions to designate the
decomposition of corpses, especially their bones, under the earth. Al-
though descriptions of wounds abound in the Homeric epics, pus is never
mentioned. The silence is probably one in1posed by the heroic perspective
and high style of the poetic narrative, which deliberately ignores life's little
woes in favor of great exploits on the field of honor. The term puon is
attested in Empedocles, who claimed that the milk of a pregnant woman

"on the tenth day of the eight month is in the state of whitish puon. " 20
This Presocratic philosopher and physician seems to have taught that all
transformations within an organism, including physiological digestion, the
formation of milk and tears, and the_ pathological modifications of the
humors, derive from a ripening process that he called sepsis.21 This explains
his comparison of foremilk to pus. Some have translated to puon in this
passage as "colostrum," which is usually designated by the Greek term ho
puos with long u., but that cannot be justified in view of Aristotle's testi-
mony on the matter. 22 Here are his remarks: "Milk is blood that has been
cooked to perfection, not putrefied. Empedocles erred or used an inappro-
priate metaphor . . . , for rotting (sapr6tes)is the opposite of cooking
(pepsis).,and pus (puon) is rotten, while milk is a substance that has been
perfectly cooked. " 23 Aristotle, along with the peripatetic philosophers and
most later medical authors of the fourth century B.c., was ,,
indeed careful
to distinguish cooking (pepsis)from rotting (sepsis.,sepedon.,sapr6te). One
term designates ascent to a higher state, the other a process of degradation.
Hippocratic terminology in this regard was not so well defined. To be
sure, one can detect a general tendency toward the use of the pep- family
in a more positive sense than sep- /sap- , but semantic indecisiveness and
inconsistent usages from one treatise to the next affirm that the basic
theories of Greek medicine concerning digestion, assimilation, the forma-
tion of pus, and the elimination of harmful ,,
substances were still in flux.
In the Hippocratic corpus, the verb sepo.,the noun sepsis.,and their deriva-
tives designated decomposition through putrefaction, the deterioration of
the humors, the formation of pus, and at the same time the digestion of
foodstuffs, especially its final stage. 24 The verb pepaino.,the noun pepsis.,and
their derivatives are used by the Hippocratic writers above all to designate
the "maturation" that corrects disturbances in the humors, a kind of
cooking that allows an organism to overcome and eliminate harmful sub-
stances. 25 This process can also produce pus. In a medical observation
collected on the island of Thasos, there is talk of patients "in whom the
disease resulted in suppuration (empuema)or some similar coction (pepas-
m6s)." 26 This pepasm6s.,or cooking of the body ·humors that are in an
abnormal state, appears as well in other medical texts, always in connection
with the formation of pus. There is no need to see a paradox in this: for
the Hippocratic physicians, suppuration was not necessarily an evil process
implying degradation. It could also be a transformation of raw and very
dangerous pathogenic substances into a liquid that was separate from the
rest of the body and ready to be expelled.
The treatise Nutriment teaches that "pus comes from flesh" and that
"the purulent humor comes from blood and liquid. " 27 In short, the
opinions in the Hippocratic corpus are that p1_1sis formed either from
decaying battered flesh or extravasted blood or, more rarely, from the
aqueous humor, ,or even from phlegm. 28 After having remarked on the
abundant suppuration of head wounds caused by blunt, hard, heavy ob-
jects, the author of a surgical treatise explains that "the crushed, wounded
flesh in such cases is inevitably transformed into pus and dissolves.' ' 29
Modern medicine confirms the clinical side of this statement: it is really
true that contusion favors the suppuration of a wound. 30 According to the
Hippocratic writers, wounding implies significant movement of blood,
not just the hemorrhage, but also and even especially an inflammatory
affiux. 31 This engorging of the wound with blood causes suppuration: the
wounds ''suppurate once the blood is modified or warmed until it is
putrefied and passes into the state of pus. " 32 A passage in the fourth book
of Diseasesis the key to understanding how the surgical experience of the
suppuration of wounds served as a model for the Hippocratic explanation
of the formation of pus inside the body in the absence of a violent external
lesion:
If an injury (-trauma) occurs, the flesh is harmed and a wound (helkos)ensues; I call
that a disease (n6sema). If a contusion occurs as the result of a blow, a fall, or
another injury of that kind, and if there is swelling (oidema)) the blood, which has
been forthwith warmed by the violent act, flows back towards the open vessels
and, since it has no means of escape, for all its abundance, it hardens: that is the
origin of the swelling which lasts until evacuation either through the pathways
mentioned or when, at the same place where the swelling occurs, by surgical
intervention or otherwise, the blood finds an exit, whether or not it has changed
into pus with the passage of time. 33
The use of the term n6serna'disease' for a traumatic lesion may surprise the
modern reader. It seems to me very significant for understanding the
ancient Greek physicians' logic. The author of the treatise Fractures even
goes so far as to ask ''if one can say that other diseases (that is, non trau-
matic diseases) are not themselves wounds. " 34 The double meaning of the
word helkos.,at once "wound" and "ulcer," is troubling. Probably the
word originally meant traumatic lesion, 35 though it is important to be
properly cautious in saying so in order not to falsify the archaic way of
thought by introducing later conceptual distinctions. On the level of the
external lesion, the confusion between wound and ulcer must have been
total: in the protohistoric and classical Mediterranean world, every open
wound was infected, so there could have been no distinction between
healing by primary scar formation and the results of secondary infection. 36
This initial confusion is precisely what permitted the extension of the
notion "wound/ulcer" from an external traumatic event to a spontaneous
external or internal lesion, that is, one that appeared in the absence of an
external act of violence.
Extravasated blood discharged in a part of the body that is foreign to it,
for instance, in a cavity in which it is not normally found, changes into
pus. If that is not an absolute truth for the modern physician who knows
126 DISEASES IN THE ANCIENT GREEI( WORLD
well the indispensable role of infection, at least it is an opinion that was

firmly upheld by certain Hippocratic treatises, 37 which could thereby ac-
count for various clinical observations. According to the Hippocratic writ-
ers, there were at least three process by which blood could change into pus
without there being an external lesion on the body: plethora and stagnation
of blood; "spasms" 38 of the flesh with secondary rupture of small blood
vessels; and ''spasms'' of the blood vessels. The pathological mechanism is
described in several passages of the first book of the treatise Diseases)for
instance, in chapter 14 on abscesses in the thoracic cavity and in chapter 17
on abscesses in the abdominal cavity. 39 According to this same treatise, pus
can also come from phlegm: the empyema arises from the rotting of phlegm
that comes from the head and empties into the pleural cavity. 40
The distinction that was made in prognosis between "good" and "bad"
pus had dramatic significance. A suppuration that was white, homoge-
neous, not bloody, and not malodorous was valued as a good omen, while
a purulent discharge with the opposite qualities was considered the reverse. 41
There is some truth, to be sure, in this rule of prognosis, since the quality
of the pus depends on the germ that induces its production, and yellow,
cloudy, fetid pus is generally caused by very harmful bacteria. However, if
one believes that the suppuration of wounds is inevitable, the difference
between "good" and "bad" pus is made in a misleading context. The
"good" kind of purulent inflammation seemed desirable, and instead of
combating it, various means were used to stimulate it. This tragic mistake
was reinforced, in daily practice, by the observation of cases in which the
abscence of a pyogenic response actually was a bad sign due to failure in the
immune system, or again by cases in which suppuration visibly helped the
healing process by the destruction of necrotic tissues in open wounds before
the practice of preventive debridement. From a theoretical point of view,
this false conclusion was facilitated by its elegantly simple integration into
the Hippocratic teachings about the pathology of humors. For these teach-
ings, suppuration and the subsequent evacuation of pus were means by
which an organism's phusis eliminated harmful substances and prevented
relapse: "All suppuration toward the exterior runs counter to a relapse,
since therein lies a coction that is both crisis and abscession. " 42
The First Descriptions of Traumatic Purulent Inflammations
The historian Herodotus tells how an accident slew the Persian king
Cambyses II at the very moment when he wished to rush out and defeat a
usurping Magus:
It was clear to him now that the murder of his brother had been all to no purpose;
he lamented his loss, and at last, in bitterness and anger at the whole miserable set
of circumstances, he leapt upon his horse, meaning to march with all speed to Susa
and attack the Magus. But as he was springing into the saddle, the cap fell off the
sheath of his sword, exposing the blade, which pierced his thigh-just in the spot
where he had previously struck Apis the sacred Egyptian bull. Believing the wound
to be mortal, Cambyses asked what the name of the town was, and was told it was
Ecbatana. There had been a prophecy from the oracle at Buto that he would die
at Ecbatana; and he had supposed that to mean the Median Ecbatana, his capital
city, where he would die in old age. But, as it turned out, the oracle meant
Ecbatana in Syria ... some twenty days later he sent for the leading Persians who
were present ... and when the Persians saw the king in tears, they tore their
clothes, and showed their sympathy by a great deal of crying and groaning. Shortly
afterwards, the bone decayed and the putrefaction spread quickly over his thigh,
and the sickness took away Cambyses son of Cyrus. 43
This took place near the end of the summer in 522 B.c. I pass over the
various magical and religious aspects of Herodotus's narrative (such as the
divine etiology of the wound superimposed on the physical one; the ora-
cle; the prophetic dreams; the king's mental illness; and so forth) to focus
on its medical content. The diagnosis is straightforward: Cambyses died a
victim of traumatic osteomyelitis with septic complications.
A similar misadventure overtook Miltiades. This famous Athenian gen-
eral, the victor at Marathon, had an accident during the siege of Paros in
489 B.C. When he returned to Athens in a sorry state and without having
conquered the town he had besieged for twenty-six days, he was accused
of having deceived the Athenians: "Miltiades, though present in court,
was unable to speak in his own defense because his leg was putrid; he lay
on a couch and his friends spoke for him . . . The popular verdict was to
spare his life, but to fine him fifty talents for his offense. Shortly afterwards
the rot in his thigh grew worse and he died.' ' 44 Herodotus, who reports
this event as an undeniable historical fact, informs us as well concerning
the origin of Miltiades' ailment, but not without some reserve. According
to the Parians, who were the only ones to say so, Miltiades entered the
holy precinct sacred to Demeter that was situated on a hill before the town
of Paros. There, on the very threshold of the sanctuary, he was overcome
by fright and would have returned, except that "in jumping back over the
stone wall he broke his thigh, or, as others say, he smashed his knee. " 45
The infection's point of entry was therefore a compound fracture of the
thigh or perhaps a skinned knee. It could not be just a hip dislocation, as
certain commentators have allowed. Herodotus does not cast doubt on
the reality of the initial wound, and if he speaks of it with rhetorical
circumspection that is only because of the Parians' insistence on Miltiades'
sacrilegious behavior. In this case and that of Cambyses, the serious con-
sequences of an apparently harmless wound inevitably suggest divine
. .
1ntervent1on.
That interpretation was no longer acceptable to the physicians of the
classical era. The Hippocratic writers knew perfectly well the dangers in-
herent in any break in the continuity of the integument. Even without
any conceptual knowledge of infection and the lymphatic system, they had
carefully observed the pathological changes now called nonspecific lym-
phangitis and lymphadenitis. They knew that when a cut is inflamed, the
flesh in the immediate vicinity swells up, and swelling and heat spread
from there along the blood vessels. If a wound is in the leg, tum ors form
in the groin; if it is in the arm, their preferred site is the armpit. A local
injury, warmed by the affiux of humor, makes the whole body febrile.
One can die of it, especially on odd-numbered days. 46
These Hippocratic physicians say that if a wound is deep enough, sup-
puration affects bones as well as flesh. In that case local necrosis can be
observed (sphakelosor gangraina) along with the generalization of symp-
toms. This happens especially in the case of compound fractures. If a leg
bone protrudes through the skin, reduction must not be attempted, says
Hippocrates, since
the patient who undergoes such a reduction succes~fully will surely succumb within
a few days; only a few last more than seven; the spasm 47 kills them; it can also
happen that both leg and foot are attacked by gangrene (gangrainousthai). 48 Should
there be a fracture or a wound, for example of the leg or the thigh, and should the
ligaments or the tendons communicating with these body parts lose their spring,
or should the heel become gangrenous due to an uncorrected bed position, serious
complications will follow. Other than local necrosis, very high and quickly fatal
fevers can occur, along with weeping, trembling, and mental disorders, and also
lividity of large blood vessels which then extravasate blood, and finally gangrene. 49
Among the patients whose histories are recorded in the seven books of
the Epidemics)there are numerous cases of traumatic infection with sup-
puration of flesh and bone and typical sepsis whose likeliest cause is the
ancestor of the Staphylococcusaureus.50 We know today that bony tissue can
also be infected in the absence of any wound by the hematogenous route,
with microbes spreading from a focus inside the organism. These "spon-
taneous'' bone suppurations did not escape the attention of the Hippo-
cratic physicians. The treatise Joints contains this remarkable passage:
"When the joint is dislocated and started from its socket by disease-such
things often happen-if necrosis of the thigh-bone occurs in some of these
cases, chronic abscesses are formed ... and there .is some denudation of
bone. Likewise, both where there is and where there is not necrosis of the
bone, it becomes much shorter, and will not grow correspondingly with
the sound one. " 51
This is without doubt a description of hematogenous staphylococcic
osteomyelitis, a disease well known in modern times and one that was
much feared before antibiotics. 52 It is a juvenile form of osteomyelitis
related to the rate and time of osseous growth. It has a marked predilection
for the long bones of the lower limbs. The importance that the Hippo-
cratic corpus accords to head caries leads me to believe that in Greeks of
the classical period another preferred site of osteomyelitis was the skull. 53
Probably the reas9n for this was the very high frequency of infections in
the area of the mouth, nose, and ears, especially in children.
In ancient times as well as in the not so distant past, tiny skin wounds
or subcutaneous injuries were often followed by purulent reactions that
were generally benign but sometimes were serious or even deadly: furun-
cles, whitlow, phlegmon of the foot, erysipelas, and so forth. The Hip-
pocratic treatises describe cases in which the progress of the infection was
stunning. For instance, an inhabitant of Thasos died the second day after
a phlegmon began forming on his big toe; 54 or a shoemaker succumbed in
three days to a swelling of his thigh caused by an accidental puncture with
one of his tools. 55 In both these cases the infection of an apparently insig-
nificant cut is followed by violent inflammation and fatal sepsis. The like-
liest agent is an especially virulent streptococcus, probably a betahemolytic
Streptococcus pyogenes.
The Hippocratic description of erysipelas is the best proof of the exis-
tence of harmful stocks of this bacterium in classical Greece. To be sure,
the term erus(pelasin ancient Greek medical parlance designates various
diseases that "redden the skin" and also diffuse, purulent inflam1nations
of internal organs, but in its commonest sense it designates a group of skin
diseases with hot, painful, reddish swelling, now thought to be strepto-
coccic dermatitis. 56 The Hippocratic term erus(pelasmay also cover another
disease, erysipeloid, the human form of a disease affecting pigs, but that is
only a guess. By contrast, it is well established that it did cover gas gan-
grene, a particularly serious disease caused by the infection of wounds with
57
certain species of C!.ostridium)Aerogenes)and Bacteroides. These anaerobic
bacteria are found in soil the world over. (Among other possibilities for the
infection of wounds, I should mention in passing tetanus 58 and anthrax. 59 )
Wounds to the abdomen regularly resulted in peritonitis, a disease caused
by various microbes introduced from without or from the intestinal con-
tent. Here is a case observed at the battle of Delos, which took place
around 357 B.C.:
A man from Ainos wounded at Delos by a javelin on the left side in the back
suffered little from the wound itself, but on the third day, he had sharp abdominal
pain. No stools. An enema brought some excrements at night. The pain did not
subside, but became localized in the testicles. On the fourth day, waves of pain in
the pubis and over the whole abdomen. He could not stay still; he vomited dark
bilious matter; eyes like those of people fainting. He died after the fifth day. There
was a slight fever. 60
The Oldest Medical Evidence on Purulent

Inflammation of Internal Organs
There is hardly an organ that cannot be the site of a staphylococcal,
streptococcal, or pneumococcal suppuration. In the Greek medical writ-
ings of the classical period, there are clinical accounts of purulent inflam-
mation in numerous internal organs. In interpreting these descriptions, a
modern physician can say that the following sites were either the common-
est among the sick of those days, or, to give a better account of historical
reality, the most striking to ancient healers: tonsils, the ear and the cranial
sinuses, the meninges, the lungs and pleura, joints, the kidneys and the
perirenal region, the biliary organs, the peritoneum, the rectum, and the
uterus and its annexes. Of all this abundant and varied pathology I can cite
only a few instances, which will offer at least a glimpse of the necessarily
great number of diseases with this bacterial etiology.
Membranous pharyngitis, peritonsillar abscess (Duguet's angina), and
Ludwig's angina were soon isolated as specific clinical entities. 61 Inflam-
mation of the tonsils was thought to promote submaxillary adenitis, to
spread to the lungs, and even to destroy its victim with the typical symp-
toms of sepsis. There is no doubt for a modern physician that sore throats
like these must have often resulted in rheumatic fever with cardiac and
renal manifestations.
Suppurative otitis media was also a well-known disease, and the Hip-
pocratic physicians knew its usual symptoms (flow of pus from the external
auditory meatus, severe pain, hearing loss) as well as those of its most
spectacular complications, like mastoiditis, phlebitis of the lateral sinus, or
meningitis. 62 Here is a clinical history of it told in exemplary fashion:
In Kydis's son, near the time of the winter solstice, shivering and sweating, head-
ache and earache in the right ear. This kind of misery had begun when the boy
was still a baby, and it continued, with fluxions and fistulas and horrible smells.
When he was in this state, it was usually painless, but on this occasion the pain
was terrible, especially in the head. On the second or third day of it, when moving
his bowels, he began vomiting bile that was as viscous as an egg, slightly bilious,
slightly ochre in color. On the night of the fourth day and during the fifth, a little
delirious; the pain in his head and ear were terrible, as was the fever. On the sixth,
his stomach was emptied by the administration of Mercurialis annua; the heat
seemed to have subsided and the pain along with it .. On the seventh day, it was as
if he was cured, but the beating in his temple persisted; he did not sweat at all. On
the eight, he kept down some boiled barley, and in the evening some beets, then
during the night, the pain stopped completely. On th~ ninth day, too, he re-
mained himself until sunset, but during the night the pain in his head and ear
were terrible; and there was also this, that the ear was punilent precisely when the
pain was at its height, just after it started. During the whole night of the ninth
day, as well as during the next day and most of the next night he recognized no
one but did not stop groaning. At daybrealc, he was himself again, the pains
subsided, he was less hot. On the eleventh day, he took some mercurial, and his
excren1ent consisted of foul-s1nelling phlegm and n1ucus. Twelfth and thirteenth
days: in a moderate state. On the fourteenth, he began sweating from tip to toe,
from morning till evening, with n1uch sleep or sleepiness: waking him was a task.
Toward the evening, he awoke, and his body cooled off satisfactorily, but the
beating in his temples persisted. On the fifteenth and sixteenth days, he ate some
boiled food. On the sevent~enth, pain returned to the same places, there was some
delirium, and there was a flow of pus. On the eighteenth, nineteenth, and twen-
tieth days, the pain was unleashed: he shrieked and tried to stand up, but could
no longer control the movement of his head; he stretched out his hands, and kept
on chasing something in the void. On the twenty-first day, a little sweat on the
right side, the chest, and the head. On the twenty-second, his face sweated pro-
fusely; as for his voice, by this time, if he tried very hard, he could succeed in
saying everything he wanted to, but if he let himself go, he only half succeeded;
besides, his mouth was limp and his lips and jaws were always moving, as if he
wanted to say something; his eyes were often moving and glancing, and there was
a tinge to his right eye, like one we call "bleeding" eye; the upper lid was swollen,
and the end of his jaw was reddish; and all the blood vessels in his face stood out;
and as for his eyes, he never closed them and had a fixed stare, and he opened the
lids upward, as when son1ething gets stuck in your eye. When he was drinking, as
it ,vent down through his chest to his stomach, a noise like that observed in the
case of Chartades. Breathing always satisfactory and moderate; tongue, like pa-
tients' with peripneumonia, light yellow. From the beginning and without a letup,
pain in the head; neck always fixed, so that he had to turn his whole body along
with his head; the spine from the neck down was rigid and impossible to bend.
Bed positions as described, and not always on the back. Pus from that time on was
milky white and abundant, a trial to clean up, with an excessive smell. At the end,
completely insensitive to being touched on the feet. 63
Cases of meningitis and meningoencephalitis are brought up in the Hip-

pocratic writings, but the nosological accounts confuse these diseases with
tetanus, cerebral abscess, and even intracranial hemorrhage. 64 As for un-
specific inflammations of the thoracic organs, Greek physicians of the clas-
sical era often came across lobar pneumonia, bronchopneumonia, and
serofibrinous pleurisy, as well as empyema and abscess of the lung. 65 The
Hippocratic disease called peripleumonfeis in most cases what we call lobar
pneumonia. Besides a form of it that is assuredly caused by a pneumococcal
infection, the author of Diseases)III, described one clinical variant that
corresponds exactly to streptococcal or staphylococcal pneumonia. 66 Simi-
larly, Hippocratic pleurttis is actually most often either bacterial or viral
pneumonia. 67 Pleuritis in the modern sense of the word had no name as
yet. The author of Diseases)II, mentions typical serofibrinous pleurisy as
an ailment that occurs "if the lung collapses against the side. " 68 These
diseases, peripleumonfeand pleurttis) are often said to change into empuos)a
term that includes pulmonary abscess and empyema in the modern sense
of the word (purulent pleurisy). 69 The Hippocratic practitioners had re-
markable familiarity with purulent lung diseases. For instance, they knew
the diagnostic value of clubbing or the use of wet clay to determine the
exact spot with the largest accumulation of pus, and so forth. They prac-
ticed the drainage of empyemas by thoracic paracentesis, and they also
operated on perinephric abscesses, which are common complications of
renal lithiasis and were probably caused by staphylococci. 70
Nontraumatic purulent inflammations of the main abdominal organs
were poorly known. Only with the greatest reservations can one propose
a retrospective diagnosis of cholecystitis (inflammation of the gall bladder)
or appendicitis on the basis of vague and confused descriptions. By con-
trast, the Hippocratic writers were not ignorant of metritis, still less of
puerperal fever. 71 In the two oldest books of the Epidemics)there are no
fewer than five clinical histories with fatal consequences in which a diag-
nosis of puerperal sepsis can be made without the least hesitation. 72
Chapter Five
THE ORIGIN AND SPREAD

OF SYPHILIS
Syphilis, leprosy, and tuberculosis are infectious diseases whose impact on

the life and destiny of human societies cannot be overestimated. Unlike
''plagues,'' acute epidemic fevers that fire the imagination of chroniclers
with their brutality and their ephemeral or exceptional aspects, these three
chronic diseases keep a low profile and, insofar as they are long-term
phenomena, tend to be passed over in silence by contemporary accounts.
Yet they are the great hidden killers.
All three of these diseases share a common property that is especially
important for the historian: they produce lesions on bone tissue that are
specific from the anatomopathological point of view, unlike those pro-
duced by inflammations due to common germs. In other words, in cases
of syphilis, leprosy, and tuberculosis, the morphological analysis of bone
lesions often supports an etiological diagnosis, that is, the identification of
a causal agent. Until now, no trace of syphilis has been found on the
ancient bones exhumed in Greece and its vicinity. As for leprosy and
tuberculosis, no truly certain osteoarchaeological case is yet known. But
caution is necessary in the face of such negative results. They have value
only as indicators, and they must be evaluated in the light of biological as
well as historical considerations. In my opinion, syphilis did not exist in
the ancient Greek world; leprosy may well have been present but only
sporadically, not becoming endemic until the end of antiquity; and tuber-
culosis had flourished there since mankind's arrival, or at least since the
Neolithic Age.
133
134 DISEASES IN THE ANCIENT GREEI< WORLD
The l+oblemof the Unity of the Treponematoses

The question of the origin of European syphilis continues to interest
and divide medical historians. 1 In earlier days, scholarship seemed to
vindicate those who held that syphilis was imported into Europe by
Columbus's sailors as late as the last decade of the fifteenth century. But
now eminent specialists are adducing biological arguments in favor of his-
torical speculations that the disease was distributed throughout the world
at a very early date. At least the microbe that is its cause was so distributed,
though the pathological consequences varied depending on its geographi-
cal and social context.
For those who value opinions founded uniquely on literary sources, the
"American" theory is beguiling. Modern 1nedical-historical investigation
has dismantled little by little the dossier assembled by scholars of the
nineteenth century 2 to support the antiquity of venereal syphilis. 3 Actually
there is no information from written documents or the oral or icono-
graphic traditions that justifies the hypothesis of the existence of this dis-
ease in ancient populations of the Mediterranean. If the question is asked
about venereal syphilis in western and southern Europe during the Middle
Ages, although a negative answer is less certain, there, too, no serious
literary proof has been found for a positive one.
In discussing the existence of syphilis in ancient Europe, I have taken
care not to omit the epithet "venereal," since it is necessary to distinguish
between at least four forms or types of syphilitic infection. First, there is
syphilis in the strict sense of the term, a sporadic disease of urban popula-
tions that is chiefly transmitted by genital contact. Its etiological agent is
the pale spirochete, Spirochaetapallida) discovered in 1905 by F. Schaudinn
and today called Treponemapallidum. The second form is endemic syphilis
(otherwise known as treponarid), which is nonvenereally transmitted, lo-
calized in hot, dry climatic zones, known by various regional names (for
example, bejel in Syria and North Africa, dichuchwa among the Bochi-
mans of Botswana, and scherlievo on the Adriatic coast in the nineteenth
century), and occurring among nomads or poor rural populations. Caused
by what appears to be the same germ as venereal' syphilis, the endemic
form of the disease presents clinical features that distinguish it from the
sporadic one. The third disease of this group is yaws (other names: fram-
besia, pian, and bouba), a serious skin and bone ailment that is transmitted
directly, and usually not venereally, among humans. Yaws flourishes in
warm, wet, forested regions, especially in central Africa. Its etiological
agent, Tr. pertenue) was described by its discoverer, A. Castellani, as a
slightly different germ from Schaudinn's treponeme. The fourth and last
form is pinta (other names: carate, mal de pinto), a relatively benign skin
disease limited to the tropical zone of Latin America. Pinta bears little
SYPHILIS 135
resemblance to ordinary
, syphilis, despite the fact that the latter is a multi-
form disease par excellence.I speak of it here only because its etiological
agent, TI: carateum) is indistinguishable from TI: pallidum.
The three forms of Treponemajust mentioned appeared morphologically
close to one another at the time they were discovered. Bacteriologists took
them for cousins, descendants of one particular branch in the evolution of
microbes. But to our surprise, it has emerged in the present state of knowl-
edge that these pathogenic treponemes do not just resemble each other,
they are absolutely identical, both as to their appearance (even under the
electron microscope) and in their responses to staining and in their anti-
genic structure as well. No serological test can distinguish among them,
and immunity given by one has crossover value for secondary infection by
the others. 4 Are we then faced with a single microbial species, or with
several different species or varieties whose distinctiveness resides only in
their pathological effects? Can the yaws treponeme produce ordinary syph-
ilis in the right conditions? And can that of venereal syphilis be trans-
formed into the germ of bejel and vice versa? In a word, are there several
treponematoses or is there just one treponematosis whose clinical manifes-
tations vary according to environmental conditions?
This taxonomic problem is not devoid of interest for a reconstruction
of the historical past of syphilis. Unfortunately, we do not yet know which
of the two answers to this question is the right one. No definite conclusion
can be drawn from experiments on the specificity of treponemes isolated
from patients with known diseases that were then inoculated into animals.
I have no choice but to flesh out, one after the other, the two contradic-
tory hypotheses.
HudsonJsHypothesis:A SocioculturalExplanation
The identity of the germ and the unicity of treponematosis were hy-
pothesized, and the hypothesis along with its medical-historical conse-
quences were carefully elaborated, by the American hygienist Ellis Herndon
Hudson. 5 According to him venereal syphilis, bejel, yaws, and pinta are
four syndromes of a single disease, treponematosis, which is produced by
a single germ, TI: pallidum. The diversification of these syndromes occurred
as a biological response to the challenge of abrupt changes in climatic _and
sociocultural conditions. The ...four syphilitic syndromes are therefore, ac-
cording to Hudson, epidemiological phases of a single nosological entity,
and they can be converted according to a biological gradient in which
endemic syphilis resides between yaws and venereal syphilis.
Pursuing this idea, Hudson arrives at a series of speculative conclusions:
that the cradle of treponematosis was equatorial Africa, where the disease
may have started in Paleolithic times with clinical manifestations almost
identical to those of yaws. The germ Tr. pallidum was derived from a
saprophytic treponeme. This disease then accompanied primitive hunter-
gatherers as they migrated over the African continent. With them, it
crossed the Mediterranean and arrived in Europe, and it also crossed Ara-
bia to reach as far as Indonesia. Outside hot, wet, forested zones, trepo-
nematosis changed its aspect. Skin eruptions gave way to mucous lesions,
and yaws changed into bejel. The principal cause of this change, according
to Hudson, was the change to an arid climate. Among the inhabitants of
the desert regions that border the tropical rain-forests, the skin is not moist
enough for treponemes, which retreat to the mouth, axilla, and groin. A
sociocultural factor reinforced the effect of climate, namely, the appearance
of villages. In the Mesolithic and Neolithic periods, endemic syphilis found
an especially favorable environment for itself in the Fertile Crescent. So
treponematosis was established in this part of the world as an endemic
childhood disease not linked to sexual activity. In comparison with other
dangers to life and limb in the past, it was relatively benign and socially
obscure. Once it arrived in Ceylon, the endemic form of the disease could
revert to yaws. Leaving India and Indonesia, it spread over the islands of
the Pacific to New Guinea and Australia. In the Neolithic period, again in
its endemic form, it crossed Siberia, Alaska, and North America, ulti-
mately reverting to the tropical climate of its African birthplace in Central
America, which allowed it to transform itself into pinta and yaws.
The final phase of this speculative history, again according to Hudson,
was the appearance of venereal syphilis. He attributes its rise to the urban
way of life. Treponematosis became a sporadic disease affecting adults,
attacking parenchymal internal organs and the central nervous system and
transmitted genitally. Its expansion no longer depended on climate. The
transition from endemic to venereal syphilis took place in many places at
various times wherever rural life changed into urban life. Endemic and
venereal forms must have coexisted in Mesopotamia and Egypt starting
between 6000 and 4000 B.C., and in Greece at about 900 or 800 B.C.
Beginning in the classical period, says Hudson, syphilis was confused with
leprosy, a similar, chronic disease of the skin and bones, which "probably
came out of Africa during the prehistoric migrations." At the latest, en-
demic syphilis spread into Europe during the Roman period and was pres-
ent there, especially in rural areas, throughout the Middle Ages. So the
poor sailors of Columbus are not to be blamed for having introduced
venereal disease into Europe after all, even if we admit that they brought
with them a few more treponemes. Environmental factors, chiefly climate
and social habits, were paramount in determining the character of trepo-
nematosis in each epoch and country. It is worth noting, in conclusion,
that Hudson's putative history is evolutionist, but it insists on the socio-
cultural evolution of the host (man) rather than the biological evolution
of the parasite.
SYPHILIS 137
Hackett)s Hypothesis: A Microbiowgical Explanation

The opposite point of view has been maintained by the British anat-
omopathologist Cecil John Hackett. 6 According to him, mankind suffers
from four treponematoses. Venereal syphilis, bejel, yaws, and pinta are
not four forms of one disease but truly four different nosological catego-
ries. Their etiological agents differ among themselves by their pathogenic
properties and cannot be transformed into each other simply as the result
of environmental changes. Though it is true the four microbes resemble
one another to a great degree, that is because they are very closely related,
all being derived from a common ancestor from which they differ only in
a few details of fine biochemical structure.
According to Hackett, the existing microbe that is the closest to the
primitive stock is Tr. carateum) the etiological agent of pinta. Either it is
the ancestor of the group or a precocious offshoot of it, the derivative of a
treponeme that lives parasitically in animals. To justify his opinion on the
relative antiquity of pinta, Hackett points both to the length of time it is
contagious, a characteristic that permits the disease to maintain itself over
time even in small populations, and, more important, to its geographic
limits-at present it is isolated among Indians living in humid areas of
America. According to Hackett, the first human treponematosis appeared
somewhere in the Old World, at the borders of the three continents,
"perhaps before about 20,000 B.c.," through the mutation of a germ
causing treponematosis in simians. Migrations of primitive hunter-gatherers
spread it throughout the inhabited parts of the earth, except for Europe
and northwestern Asia. The date 20,000 B.C. seems to have been chosen
by Hackett to make the origin of pinta older than the land-bridge across
the Bering Strait (around 15,000-10,000 B.C., according to the geophysical
theory accepted by him), the likeliest point of its passage into America.
But in Hackett's hypothesis it is not clear what determines the tenninus
post quem of the initial mutation.
The wet and hot climate of" Afro-Asia" (sic)favored the most invasive
and destructive stock of Tr. carateum and fostered its mutation into Tr.
pertenue around 10,000 B.C. As a result, yaws invaded all of Africa, the
Middle East, and the Far East, the islands of the Pacific, and Australia, but
without reaching Europe or the Americas before the Bering Strait was
flooded. Wherever it established itself, yaws eliminated pinta. The arid
climate that ruled certain parts of the earth after the Ice Age was responsi-
ble for a new birth, by natural selection of mutants of Tr. pertenue: the
appearance of Tr. pallidum around 7000 B.C. From this time on, yaws
remained the only treponematosis within the tropics, but endemic syphilis
took its place in the northern and southern zones, invading the eastern
Mediterranean and central Asia.
Finally, the change of a non venereal treponematosis in rural children
into venereal syphilis in urban adults took place, in Hackett's view, during
the postglacial climatic optimum around 3000 B.c., in the now arid lands
of the eastern Mediterranean and southwestern Asia. Venereal syphilis,
says Hackett, was probably carried by ship along the Mediterranean coasts,
and later on, especially after the Roman conquests in the first century B.C.,
throughout Europe, ''which was a treponemally uncommitted area.'' The
map accompanying Hackett's study shows clearly that for him, venereal
syphilis must have been present between 3000 and 100 B.c. in the Balkan
peninsula, North Africa, Asia Minor, Italy, and on the French and Spanish
coasts.7
The anthropologist Don Reginald Brothwell basically accepts Hackett's
view of the general history of treponematoses, but he proposes a different
date and locale for the mutation that created Tr. pallidum. By his lights,
syphilis appeared in the Far East "at least two thousand years ago." This
cautious assertion suggests an appreciably later date than the one proposed
by Hackett. If it is correct, the spread of venereal syphilis from Asia into
the Americas via the Pacific islands and into Europe via the expanding
Arab world could not have taken place until the end of the Hellenistic
period. 8
Cockbun1/sViews
Aidan Cockburn, a well-known epidemiologist who is both bold and

concerned with scientific precision, takes Hudson's side. He believes that
the treponematoses "are in fact merely forms or variants of one basic
infection, the differences being due chiefly to modes of transmission, cli-
mate, geography, humidity, etc. " 9 At the very beginning of the patho-
genic treponemes, says Cockburn, there was a free-living form whose
descendants include certain current nonpathogenic protozoa living in the
soil (Tr. zuelzerae). At some very distant moment in the past, well before
the origin of man, free spirochetes began attaching themselves to the
higher animals to live symbiotically in their skin and the orifices of their
bodies. Did man inherit the treponemes from his ancestors, or was he
infected horizontally by an animal species in his ecosystem? Cockburn first
hesitated on this issue, since the distribution of treponen1es in the animal
kingdom is poorly understood (mainly because it is necessary to take into
account the symbiotic treponen1es that are not pathogenic to their hosts).
His most recent opinion is that the original infection was produced several
million years ago in a co1nmon ancestor of man and the higher apes. 10
There are some signs pointing to the existence of a clinically hidden au-
tochthonous treponematosis not unlike yaws in African gorillas and ba-
boons.11 However, the only form of syphilis knovvn to exist in animals
whose microbial agent has been definitely isolated is one that affects not
SYPHILIS 139
apes but rabbits. That microbe, Tr. cuniculi) differs only slightly from Tr.
pallidum. Even their serological reactions are identical. Did man catch the
germ from rabbits, or did they catch it from him, or did both catch it from
another reservoir of treponemes? The domestication of the rabbit was
relatively late and the disease does not seem to affect other Leporidae, so
that the transfer from man to rabbit seems much more likely than the
other way around.
According to Cockburn, Homo sapiens) whether he first appeared in
Africa or Asia, brought with him, when he crossed the Bering Strait, the
germ of a primitive form of treponematosis, a disease older than mankind.
The different modern forms of that disease resulted from geographic seg-
regation and from environmental differences. Thus the peculiar features of
pinta were due in part to the modification of the microbe (Tr. carateum is
the only one that resists all forms of inoculation into animals), and in part
to certain genetic traits of a group of American Indians. This is why the
disease remained restricted despite the possibility of its expansion in mod-
ern times. Yaws, in Cockburn's opinion the form of the disease closest to
the original treponematosis, flourished in climatic zones hospitable to it in
which sociocultural conditions facilitated direct, skin-to-skin contact be-
tween men. In Europe, treponematosis took the form of endemic syphilis.
Until about A.D. 1000, says Cockburn, treponemes cohabited with all of
the world's human populations, either as commensals or as parasites, but
they produced only mild, chronic disturbances since population density
was low. The wearing of clothes hampered treponematosis in temperate
regions, where the improvement of hygienic conditions and increased
prudery brought about the disappearance of endemic syphilis and, through
the survival of strains with a predilection for the genitals, the appearance
of venereal syphilis. The discovery of America and the pandemic of vene-
real syphilis are not causally related but are parallel results of one common
sociocultural factor, namely, the transformation of society due to the Ren-
aissance and the Reformation.
The Osteoarchaeology
of the Treponematoses
The historical and epidemiological reconstructions of Hudson, Hackett,
and Cockburn are based on general biological considerations and the cur-
rent distribution of four syphilitic diseases. They neglect specific historical
information, and all of them face a major difficulty: the osteoarchaeology
of the eastern Mediterranean. They do not take account of the fact that
no trace of syphilis has been discovered on more than 25,000 skeletons and
mummies from ancient Egypt and the Sudan or on several tens of
thousands of prehistoric, ancient, and medieval skeletons exhumed in Eu-
rope and Asia Minor. 12 In particular, it has not been possible to detect this
disease on any ancient bone from the Mediterranean. To be sure, that part
of the globe is not the central preoccupation of the scholars cited, but it is
still an indispensable part of their demonstrations.
Venereal syphilis, endemic syphilis, and pinta produce bone lesions that
can often support a positive diagnosis of treponematosis. 13Still, it is not
possible to distinguish among them just by anatomopathological exami-
nation. This failing of paleopathology led Hudson to the conclusion that
"the use of such bones to prove, or to disprove, theories about the place
of origin of syphilis is not justified,'' 14which is a strange remark. In the
present state of knowledge, not being able to distinguish between the
treponematoses on bones seems to me secondary when what confronts us
is the complete absence of the diseases in question.
Let us recall briefly the present state of osteoarchaeological research on
syphilis. This disease, or another caused by treponemes, has been reported
on several pre-Columbian bone specimens from different regions of the
Americas: Argentina, Peru, Guatemala, Mexico, the mouth of the Ohio
River, Arizona, New Mexico, Florida, and even the Antilles. 15 On the
Mariannas in Oceania the skull and long bones of a 13-year-old child have
been found to bear syphilitic lesions. According to carbon 14 dating, these
particular bones are from the Middle Ages, probably the ninth century.
Stewart and Spoehr offer a diagnosis of yaws, 16 a likely hypothesis, but
it is well to be less precise and say that it is a case of treponematosis.
The disease has also been reported on the remains of Australian aborig-
ines.17 On the other hand, no human remains older than 1500 and
bearing sure signs of a treponematosis have been found in Europe, Africa,
or Asia. 18
To explain the absence of paleopathological proofs of the pre-Colum-
bian existence of treponematoses in the Old World, the "anti-American-
ists'' stress the relative rarity of the remains that have been examined. Such
an argument was perfectly valid when it was still believed that syphilis was
present in the European populations of the protohistoric period and of
antiquity in the form of sporadic cases and small epidemics. Yet the hy-
potheses of both Hudson and Hackett are obliged to assert the presence
of endemic syphilis in the eastern Mediterranean, either alone or in concert
with venereal syphilis. However, the conservation of endemic syphilis de-
mands very high morbidity, so high that it makes no sense for the disease
to have escaped osteoarchaeological detection. Moreover, it seems unbe-
lievable that an endemic disease of such scope could have escaped the
notice of Greek physicians. Another argument of the "anti-Americanists"
does not withstand criticism either. It is said that Old World treponema-
toses were confused with leprosy. The famous medical historian Karl
Sudhoff said the same for the Middle Ages, not without justice, but this
argument has no value if we also apply it, as Hudson does, to classical
antiquity. The osteoarchaeological proofs of the existence of leprosy in
Europe are very late. Physicians of the classical period could not confuse
SYPHILIS 141
the endemic pres~ence of syphilis with leprosy for the simple reason that
they did not know of either one.
Tr. Pallidum: An American Mutant

The biological hypotheses on the origin of syphilis certainly deepen
knowledge and exercise ingenuity, but they must be modified in order to
harmonize with the osteoarchaeological and historical evidence. I think
that the four human treponematoses are caused by three kinds of microbes
that, although they are morphologically alike, are distinguished by biolog-
ical properties and do not transform themselves into each other under the
influence of environment. They are the result of an evolutionary process
that cannot really be reversed. I speak of three and not four kinds because
it seems to me very likely that Tr. pallidum is responsible for endemic as
well as venereal syphilis. Historical studies demonstrate the reality of the
change from one to the other according to sociocultural circumstances.
But history contains no known instance of a parallel conversion for yaws
and pinta. It may well be that an as yet undiscovered technique of bio-
structural analysis will determine the degree of relationship between the
three pathogenic treponemes and make it possible to decide which of them
is the closest to their ancestor.
For the moment a decision on that matter cannot be made on a purely
biological basis. The chief considerations must still be historical and geo-
graphic. Hackett' s hypothesis, which gives the right of primogeniture to
Tr. carateum) seems to me incompatible with his guess on the Afro-Asian
origin of that disease. According to modern taxonomic principles, partic-
ularly according to the cladistic rules of Hennig and Brundin, a species
that is isolated in a territory at the edge of the geographical area in which
its group has expanded should not be considered plesiomorphous or prim-
itive. That is precisely the case with Tr. carateum) if its origin is to be
situated in the center of the Old W odd. If we admit that it is the conser-
vative branch of pathogenic treponemes, its provenance must be the trop-
ical zones of the New World. We can suppose that, in the Americas,
natural selection among mutants produced Tr.pallidum) which can flourish
in temperate regions. Once it had spread over the world, from the fifteenth
century onward, this germ caused endemic syphilis and venereal syphilis.
Only in relatively recent times, when it was restored to a hot, moist climate
in equatorial Africa, did it sire the microbe Tr. pertenue) the specific agent
of yaws. Although this hypothesis runs counter to con1mon notions, no
fact now known really contradicts it.
But I favor a different historical explanation. The original treponemato-
sis was, in my opinion, indeed a disease that mankind inherited from the
ancestors of his species. For the initial phase, I accept Cockburn's hypoth-
esis. But, believing as I do in the specificity of the three extant pathogenic
treponemes, I suggest that the original one produced, in the Old W odd,
a plesiomorphous branch, Tr. pertenue. Yaws, not pin ta, is the ancient
clinical manifestation of this microbial group. From remote times, yaws
flourished in hot, moist, forested regions, but it was never able to reach
the territory of the great Mediterranean civilizations in significant propor-
tions. In America, which it reached via the Bering land-bridge, the original
germ had a different biological evolution. It bifurcated into Tr. carateum
and Tr. pallidum) with the first, more conservative branch confined to
tropical zones, and the second, an apomorphous one, fit to conquer the
world. Introduced from Haiti into Europe by sailors in the fifteenth cen-
tury, Tr. pallidum spread rapidly over the three continents of the Old
World, producing either venereal or endemic syphilis, depending on soci-
ocultural circumstances. Pockets of endemic syphilis existed in the nine-
teenth and twentieth centuries on the Balkan peninsula, notably in Bosnia.
Historical research shows that these pockets are relatively recent, not ear-
lier than the seventeenth century, and that endemic syphilis was preceded
there by sporadic cases of venereal syphilis. 19 In any case, the Greek world
knew of no syphilitic disease, no treponematosis. There might have been
scattered cases brought in by way of contacts with black Africa, specifically
cases of yaws, but that disease could not propagate in an environment so
hostile to the survival of its specific germ.
The Mythology of VenerealPlague

Trustworthy historians like Ackerknecht and Henschen inform us that
both the experience and the concept of the venereal transmission of path-
ological states were acquired at the dawn of human history and incorpo-
rated into archaic myths: the Assyrian poem that tells of the trials of
Gilgamesh, the legend of Dionysus, and, above all, the story in the Bible
(Numbers 25) of the disease that afllicted several thousand Israelites after
their debauchery with the daughters of the Moabites. 20 Yet it is surprising,
if this is truly the rational kernel of certain myths, that philosophers and
physicians of the classical age did not know enough to distinguish it and
lay it bare. The fact is that no author from antiquity expresses clearly and
directly the notion of infection by sexual contact. Were they blinded by
the appeal of theories that left no place for magical defilements of this
kind? Or were they really lacking in actual experience of venereal disease?
And if so, isn't the blindness then ours, when we project an empirical
content onto myths that is based on our own knowledge where once there
was only a notion of punishment for breaking a tabu?
The notion of punishment is not just antiquity's prerogative. Medical
and historical writings of the eighteenth and the first half of the nineteenth
century are still imbued with it. For a goodly number of historians of
medicine prior to Pasteur, Koch, Neisser, and Schaudinn, that is, before
SYPHILIS 143
the discovery of ,the real nature of contagion and when the etiology of
venereal diseases was not yet securely established, the hoary antiquity of
syphilis and other venereal diseases was "proven" indirectly by means that
a modern reader would consider not only unsatisfactory but also inade-
quate, inept, and even stupefying. For example, Julius Rosenbaum (1807-
74), a physician at Halle and a broadly educated humanist, was convinced
he had proved the existence of syphilis in Greco-Roman times by laying
out irrefutable documentation of certain erotic practices and morals that
he considered to be corrupt. I cannot resist the temptation to quote his
methodqlogical justification (I could not summarize it without distorting
its content and losing the flavor):
The reproduction or preservation of a species being a law imposed on the genital
organs, it is not likely that this function, when exercised in conformity with its
purpose, would produce disease in these parts. Indeed, experience shows that, in
a rational marriage whose natural goal is the procreation of children, diseases of
the genital organs are rare or do not exist at all. Therefore we are obliged to admit
that there still exist other kinds of sexual functions beyond the natural goal, or at
least in which that goal becomes secondary; these other types of function have
sensual pleasure as their only purpose, and the use of the genital organs to achieve
this purpose is lust. However, just as all abuse redounds not only to the detriment
of the organ but of the organism as a whole, the same must be true for the sexual
organs. In lust, in the abuse of the pleasures of love-that is wherein we must seek
the principal cause of genital diseases. An exact knowledge of the history of lust
becomes indispensable in order to arrive at the history of diseases of the reproduc-
tive organs. 21
According to Rosenbaum, syphilis derives from the abuse of the genital

organs, which were devised for procreation and not enjoyment. In that
nineteenth-century universe, he is not by any means the only one to
imagine venereal disease as the price humanity must pay for its quest for
sexual pleasure, for the ''excitation of debilitating titillation.'' In this view,
the etiology of syphilis is more moral than medical. Since every venereal
disease was essentially just a form of divine punishment, the historian of
syphilis and all other diseases involving the genitals was obliged to study
in detail the sins of all peoples and all historical periods. This is why, after
a hypocritical sigh (' 'what a sad task it is for the historian to pursue the
degradation of humanity down to its most hideous details"), Rosenbaum
exposes with courage (his term) and not without a certain delight (evident
but not confessed) all the "lubricious" sexual practices that for him are
the cause of syphilis and so represent arguments in favor of the early
occurrence of the disease. He goes so far as to assert the syphilitic nature
of certain "Syrian ulcers," particularly in connection with the masterly
description of diphtheria by Aretaeus of Cappadocia, since for him they
are the fatal consequence of fellatio, just as leuke and aphthae are "just
retribution" for cunnilingus. 22
For the modern historian, debauchery, supposing that this moral term
can be part of a discourse about social behavior, can only be one factor
among many in the propagation of venereal diseases. Their appearance
must be based on the presence of certain microorganisms. The existence
of venereal diseases should be proved either by paleopathological exami-
nations or medical exegesis of the clinical descriptions of the past. Did
diseases transmitted by sexual intercourse really exist in ancient Greece? To
answer this question, I will review the literary evidence on diseases of the
external genitals, and so that nothing significant will be overlooked, I will
not limit my survey to venereal disease in the strict sense.
Gonorrhea) Spermatorrhea) and Leukorrhea
In modern times, surely no venereal disease has been better known and,
at least before antibiotics, more common than gonorrhea. It is a conta-
gious disease caused by a pyogenic bacterium, the gonococcus (Neisseria
gonorrhoeae).As a rule it is transmitted by coitus, and in its acute phase it
causes urethritis in men or vulvovaginitis in women. Frequently the infec-
tion propagates in the genito-urinary tract, extends into adjacent organs,
and becomes chronic. It can produce pathological symptoms in joints,
skin, and eyes.
The inflammatory reaction of tissues in the presence of the gonococcus
is not specific. As a result, diagnosis of gonorrhea is not possible through
osteoarchaeological investigation. The etiological interpretation of ancient
clinical descriptions that may relate to this disease is very uncertain, since
instances of urethritis and vulvovaginitis comparable to those produced by
the gonococcus can also be produced by other germs (staphylococci, intes-
tinal bacteria, and so forth). So it is not surprising that historians do not
agree whether gonorrhea existed in the Greco-Roman world. According to
E. H. Ackerknecht and M. L. Brodny, it was mentioned in the Bible and
described by Hippocrates, Celsus, and Galen. 23 Falke Henschen thinks he
recognizes it in certain passages of the Papyrus Ebers) the Hippocratic cor-
pus, and the work of Aretaeus of Cappadocia. 24 Adalberto Pazzini rejects
this retrospective diagnosis and asserts that gonorrhea was unknown in
antiquity. 25 The same negative view is expressed vigorously by several Brit-
ish doctors who have recently researched the whole issue. 26 The critical
examination of literary evidence does not make it possible to resolve the
debate. 27 On the one hand, the diagnosis of gonorrhea is compatible with
certain ancient descriptions, and on the other, for none of those descrip-
tions is it the sole interpretation possible.
The origin of the germ causing gonorrhea is unknown. It probably came
into being as a result of the mutation of a Neisseriaceae saprophyte, but in
our present state of knowledge that is an undatable event. For Oriel, it
SYPHILIS 145
took place after the end of the Old W odd. 28 I prefer to push back the birth
of the gonococcus to a date prior to the classical era of Mediterranean
civilization, since it seems to me that such a hypothesis takes better ac-
count of the myths of venereal plague, the relatively common references
to urethritis, semen discharge, and vulvovaginitis in the ancient medical
treatises, and, last, the biological properties of the germ itself.
The name of this disease, gonorrhea, in Neo-Latin scientific terminology
and several modern languages is a term that goes back to Greco-
Roman antiquity and that originally meant "semen discharge." Although
the Greek substantive gon6rrhoia occurs only in relatively late authors
(Aretaeus, Galen, Soranus, Oribasius), a verbal formula occurs as early as
Hippocrates. For modern physicians, this term has nothing to do with its
original sense, since it applies to the discharge of pus by the urethra or the
vagina and not to the discharge of sperm. Greek physicians used it to
designate what they saw as a real loss of the masculine or feminine seminal
substances, that is, as spermatorrhea in men and some forms of leukorrhea
in women. When the Hippocratic writers speak of semen discharge in
men, that is actually what they intend: they are referring to pollutions,
frequent acts of intercourse, and pathological spermatorrhea, not the
''gonorrhea'' of modern medicine. 29
The Hippocratic description of ''dorsal phthisis'' (notiasphthfsis)) a dis-
ease that '' especially affects newlyweds and men who have surrendered
themselves to sexual pleasure" and whose chief symptom is abundant sper-
matorrhea following urination or defecation, is evidence of an inflammation
of the seminal vesicles in the context of a complex nosological entity. 30 This
type of vesiculitis may well be due to the gonococcus, but that is only one
among many possibilities. Several ancient cases of spermatorrhea are
doubtless functional, nervous disorders. It remains no less certain that
other such cases were caused by nonspecific pyogenic bacteria or the tu-
bercle bacillus. The predilection of such germs for the genito-urinary tract
is obvious from the importance given in Hippocratic texts to clinical pic-
tures that correspond in modern pathological terminology to pyelonephri-
tis and cystitis. Both are often due to secondary infection of a primary
urinary lithiasis. 31 Thus a Hippocratic text mentions as a common event
the formation of "abscesses in the urethra" that "suppurate and burst. " 32
The third book of the Epidemicscontains a description of the principal
diseases observed on the island of Thasos by a Hippocratic physician,
probably Hippocrates himself, during an especially rainy year in the last
decade of the fifth century B.C. 33 In the spring of that year "erysipelas"
began, abscesses in the throat and on the gums, kausoi ("ardent fevers")
and "phrenitis"; there were also other affections:
Many had aphthae (aphth'1Jdea) and sores (hel~'1Jdea)
in the mouth. Fluxes (rheumata)
about the genitals were copious; sores (helkomata)Jtumors (phumata) external and
34
internal; the swellings which appear in the groin (boubonas).Watery inflammations

of the eyes (ophthalmfai)) chronic and painful; growths on the eyelids, external and
internal, in many cases destroying the sight, and which are called "figs" (suka).
There were also often growths on other sores, particularly in the genitals. 35
When interpreting this Hippocratic tex~ in the light of modern pathologi-

cal knowledge, nineteenth-century physicians separated the eye symptoms
from those of the mouth and genitals. The first, they said, were due to
trachoma. 36 Though it is as vague as can be, the description of genital
symptoms here was considered the best direct proof of the existence of a
venereal disease, either syphilis or gonorrhea, in classical antiquity. 37 How-
ever, in 1937 a Turkish dermatologist, Huhisi Behc;et, identified a particular
syndrome that combines uveitis and hypopyon in the eye with the forma-
tion of aphthae in the mouth and ulceration of the genitals. 38 Behc;et's
disease (or Adamantiades-Behc;et syndrome) also brings with it, aside from
the basic clinical triad, attacks of fever, skin eruptions, neurologic lesions,
and retinal complications that can lead to blindness. The agreement be-
tween Hippocrates' clinical sketch and modern descriptions of this disease
is perfect but for one detail, an outward excrescence of the eyelids. 39 Now-
adays this is a sporadic disease linked to certain geographic areas, with a
clear predilection for the Mediterranean basin, especially the lands that
once constituted the ancient Greek world, and Japan. Men contract it
much more often and suffer from it more grievously than women. A
genetic predisposition to it is confirmed by the high frequency of the tissue
histocompatibility antigen HLA-Bs in people with the disease. Behc;et's
disease results from the action of an exogenous factor on an organism with
a specific immunological predisposition. That the exogenous factor is viral
seems very likely, although this has yet to be really proven.
In the Hippocratic text cited above, there is mention of boubonas'bu-
boes, swellings in the groin.' To be sure, Greek physicians did not know
about lymph nodes and their physiological and pathological role, but that
did not prevent them from noticing causal relationships between suppu-
rating wounds, transient fevers, and swelling in the joints of the arms and
legs. 40 Aside from this text, the medical writers of the classical period
noticed no link between genital ulcerations and buboes. It is true that in
Aristophanes' Lysistrata the herald, who has an erection for all to see, is
asked the question, "Do you have a bubo from all your traveling?" 41 This
is a joke meant to draw attention to the physiological state of the character
who has just come on stage, but a bit of medical-historical information
can be drawn from it: Aristophanes' public knew that travel, with the risk
it at that time entailed of infection to the lower limbs, could produce
swollen glands in the groin. But I have not found the least allusion any-
where in the literature of the classical era-in comedy, erotic texts, or
elsewhere-to swelling in the groin in conjunction with ulceration of the
SYPHILIS 147
penis or urethriti~. Galen, a Greco-Roman physician of the second century

A.D., knew of the link; he explicitly says that swelling in the groin is at
times a consequence of genital ulcers. He is amazed that the ancient prac-
titioners speak of this so rarely, and he tries to explain their silence by
stating that "those with fever caused by swelling in the groin do not
summon a physician but prefer to take care of themselves. " 42
In the gynecological discussions of the Hippocratic corpus, a distinction
is drawn between the discharge of female semen and other common forms
of white or yellowish spotting, but it is difficult, if not impossible, to grasp
the pathological reality on which the distinction is based: ''If the semen
'
flows (en de hog6nosaporrhee)pure and uninterrupted, the woman does not
like to have sex with her husband; she does not become pregnant; her
loins hurt; there is a stubborn fever, weakness, and fainting; sometimes
the uterus is out of position. " 43 This description of "gonorrhea" in the
etymological sense of the term is at once true-to-life and also lacking
in differentiating information. It relates a syndrome that has been ob-
served with admirable precision but that has no specific significance for
modern pathology. It would be useless to wish to provide it with a
modern diagnosis, since a substantial number of different diseases have
these same symptoms. Chronic, continuous, and nonfetid leukorrhea,
dyspareunia, sterility, lower back pain, fever, adynamia-all this can be
found in chronic gonorrhea or, even more readily, in a patient with tuber-
culosis, nonspecific salpingo-oophoritis, hormonal disturbances, and many
other diseases.
Writers of the Imperial period, and Galen in particular, will tell us that
female "gonorrhea" is not a disease of the lower genital tract through
which the semen flows but a functional disorder of the vessels that consti-
tute and concentrate it in the abdomen. 44 The treatise Diseasesof ll7<nnen)
II, subdivides leukorrhea into five types, but this multiplying of the symp-
tomatic categories is done in such a way as to confuse the modern clinician
instead of helping him define the pathological processes behind the ancient
nosological categories. 45 Here is the first of the five:
The discharge is white like a donkey's urine; the face swells, especially below the
eyes; they are watery and look poor; their shine is gone; they are bleary and
eyesight is cloudy. The complexion is sallow, and the skin is covered with vesicles.
The lower abdomen swells. A reddish, small, watery, malignant eruption gradually
spreads over the jaws. The legs get swollen; if you press them with a finger, an
imprint forms as in dough. The mouth fills with saliva. The woman has pain in
her kardfa on an empty stomach, and she vomits some kind of sour water. If she
goes up an incline, she is soon out of breath. She suffocates; her legs get cold; her
knees are wobbly. Aphthae form in her mouth. The uten1s is unnaturally dilated
and weighs like lead on its orifice. There are shooting pains in the thighs. All the
lower limbs grow cold, from the lower abdon1en to the feet; the soles of the feet
are numb, and she cannot walk. In these cases a cure is difficult, since the patient
DISEASES IN THE ANCIENT GREE!( WORLD
is no longer youthful, and the symptoms go from bad to worse, unless by a stroke
of luck the disease spontaneously disappears. 46
According to Paul Diepgen, who was an experienced gynecologist and

medical historian, the description of this first type of leukorrhea "recalls
the clinical picture of chlorosis or pernicious anemia. " 47 M.-T. Fontanille
sees here ''the description of pelvic peritonitis with a generalization of the
infection and serious anemia. '' 48 I would add that a diagnosis of myxedema
is also possible, nor can one exclude the diagnosis of several other systemic
diseases that disturb the metabolism as well as both cardiac and renal
function. Whatever the case may be, it is only indirectly a disease of the
external genital organs. Contrary to opinions expressed in the nineteenth
century by some medical historians and repeated since then only by inertia,
it is hard to admit that this disease has venereal origins.
Generally speaking, the cases of white, yellow, and brown spotting
described in the Hippocratic treatises correspond to common types of
parasitic, mycotic, or bacterial leukorrhea and minor metrorrhagia. Bio-
logical arguments make it very likely that such infestations of the geni-
tal organs with protozoa and fungi are both ancient and ubiquitous.
Greek women in the time of Hippocrates are likely to have suffered from
vaginitis caused by Trichomonasvaginalis and Candida albicansor by ances-
tors of this parasitic species. T. vaginalis; a flagellate discovered in 1836 by
A. Donne "in purulent substances and the products of secretions from
male and female genitals," causes foaming, yellowish or greenish, fetid
leukorrhea that irritates the vulval region. Trichomoniasis is often but not
necessarily transmitted by sexual contact. No classical text describes this
disease as a separate entity, but it is probably included in nosological
descriptions that speak of yellowish, fetid discharges. The microscopic
fungus C. albicansis to blame for several pathological states: a vulvovagi-
nitis marked by leukorrhea; a mycosis localized in the large folds of skin
and the nails; rarely, septicemic complications; and, last, some common
diseases of the mouth, tongue, and bronchi. The most typical of the latter
is thrush, a candidiasis of the mucous membranes of the mouth. Its ap-
pearance is characteristic: erythematous patches with a creamy white mem-
brane in the middle that are scattered throughout the mouth in small
children and malnourished adults. The white membranes of thrush were
the points of departure for the studies in 1841 by F. D. Berg in Sweden
and, independently of him, by D. Gruby and C. Robin in France, that
succeeded in isolating the mycotic agent and demonstrating its pathogenic
role. Ancient accounts of leukorrhea do not permit a differential diagnosis
of vaginal candidiasis, but the existence of its germ is confirmed by the
mention of aphthous eruptions in children and pregnant women. Proba-
bly the aphthae in newborns discussed in the Hippocratic Aphorisms are
related to thrush. 49 According to Galen's commentary to the Aphorisms;
aphthae in the mouths of breast-feeding babies are superficial, inflamed
SYPHILIS 149
ulcerations of ths, soft tissue that come from bad milk or the inability to
digest milk properly. 50
Other Diseases of the External Genitals

Hippocratic aphthae are not localized in the mouth alone. These dis-
crete, whitish, superficial ulcerations can also be found on a woman's
genitals 51 and even in the trachea. 52 According to Jacques J ouanna, the
change in meaning from a mouth ulcer to one on the orifice of the uterus
was facilitated by the fact that both openings are called stoma in Greek. 53
The "aphthous fluxions" in pregnant women 54 are especially reminiscent
of candidiasis, a disease that occurs during pregnancy, especially in the last
trimester, and is marked by copious discharges and whitish ulcerations on
the vaginal walls. The trachea becomes aphthous from tuberculosis. 55 The
genital aphthae in Hippocratic texts probably include candidiasis, aphthosis
in the modern sense (especially Beh~et's disease), lymphogranuloma ve-
nereum (LGV or Nicolas-Favre disease), tuberculous lesions, and vulvo-
vaginal diphtheria, as well as ulcerations due to nonspecific pyogenic
bacteria and several other dermatological diseases.
Herpes eruptions can also appear as aphthae. The presence of genital
herpes simplex (in the modern sense of the term) 56 in patients of the
Hippocratic doctors seems to me to be strongly suggested by a clinical
observation that in cases of the formation of phluktainai 'blisters' associated
with menstrual flow, it happens that "the edges of the [genital] labia are
ulcerated (helk~the). " 57 Though these vesicular or blistery eruptions, when
they are situated on the orifice of the uterus, 58 could be nabothian follicles,
that they are herpetic is practically assured by the statement that they
appear during menstruation and cause erosion on the edges of the labia. I
note that a passage in the sixth book of the Epidemics probably alludes to
herpes labialis (cold sores), which coincide with accesses of fever. 59 This
disease, in its genital and labial forms as well as when it produces aphthae
in the mouth, is assuredly relatively benign, more of a nuisance than a
danger. But that does not diminish the significance of its mention in the
Hippocratic corpus for the medical historian: it constitutes proof of an
ancient pedigree for at least one kind of herpesvirus, a viral agent that is
perfectly adapted to the human being with which it lives, in most cases, as
an almost totally unnoticed parasite. It has recently become known that it
can also produce serious complications, that infection with genital herpes
in the mother can actually be fatal to the newborn child.
As for lymphogranuloma venereum (a sexually transmitted disease typi-
fied by a small, pseudo-herpetic, genital or anal chancre and severe adeno-
pathy in the groin) and nonspecific reappearing urethritis, their existence
in the ancient Greco-Roman world seems to me very likely, given the
recent discovery that they are actually caused by the same microbe, Chia-
mydia wachomatisJas endemic trachoma. Trachoma was the bane of ancient

Egypt, and there is every reason to believe that it did not spare other coasts
of the Mediterranean.
Often in the past, but rarely nowadays, women developed ulcers on
their external genitals after giving birth. Such puerperal ulcers are caused
by the maceration and superinfection of obstetric tears and scratches. Here
is a description of one from the Hippocratic corpus: it begins by a local
inflammation ''resembling an aphtha'' and ends with the formation of a
"mushroom," an excessive scar that closes the two lips. 60 Ancient Greek
doctors also knew of condylomas, which are soft, wartlike growths of viral
origin that can be encumbering. In men, they occur on the glans penis
and the prepuce and are relatively small. In women, they are localized on
the vulva and the external part of the vagina. In both sexes, they can
develop around the anus. The author of the Hippocratic treatis Wounds
says that a plant called parthenion microphyllon removes the thumia from
the prepuce. 61 Celsus explains that for the Greek physicians tumors resem-
bling warts have special names depending on their shape. Thus "that
which is named thumion projects above the surface like a little wart,
narrow near the skin, wider above, hardish and at the top very rough ....
The worst, however, are situated on the genitals, and there they bleed the
most. " 62 For the Latin medical writers, condylomas resemble figs rather
than thyme flowers, whence their Latin names ficus (especially in its adjec-
tival form, ficosus) and marisca. An assumption that this disease is transmit-
ted by sexual congress, especially pederastic, underlies some satiric lines in
Juvenal and Martial. It is possible that the assumption is correct. In the
classical period, condylomas were successfully removed by surgery. 63
One Hippocratic author speaks of ulcers (helkea) that seem particularly
dangerous. 64 According to Paul Diepgen, they sound like cancer. 65 Unfor-
tunately, the Greek text contains no details that would make it possible to
support such a diagnosis. Serious ulcerations of the genital organs are
produced by tuberculosis, chancroid, neoplastic diseases, and some others
already mentioned in this chapter. Nowadays we know that tuberculosis
can be directly transmitted genitally. A man with renal or epididymo-
orchitic tuberculosis can give his female sexual partner a vulval or cervical
tuberculous chancre. A diagnosis of urogenital tuberculosis is very likely in
the suicide of a Roman couple. The husband suffered from a chronic,
ulcerous disease of the private parts. 66 Chancroid or soft chancre, that is,
the venereal disease caused by Haemophilus ducreyiJis not described in the
classical and Hellenistic Greek texts, but its existence has been recognized
by some in the medical encyclopedia of Celsus. 67 There is also a historical
example that is doubly dubious, in that it is both late and embedded in a
moralizing narrative: in Alexandria, in the fourth century A.D., a fallen
eremite is said to be afflicted with a corrosive ulcer on his penis. 68 This is
the moment to recall as well the unfortunate fate of the Alexandrian
SYPHILIS 151
rhetorician Apion (beginning of the first century A.D.): "He was circum-
cised of necessity, to treat an ulcer on his genitals; moreover, the circum-
cision did him no good, as his flesh became gangrenous and he died in
excruciating pain. " 69 Josephus tells this story with the explicit goal of
showing how a traitor to the ancestral laws suffered by divine will a sym-
bolic punishment befitting his impiety. Even if it were historically reliable,
it would be hard to pin down the diagnosis: phagedenic ulcer? Kaposi's
sarcoma? or cancer of the penis? Cancers of the penis or the vulva are not
clearly described until the Imperial period (Celsus, Galen, etc.), but it is
hard to believe that they have not existed since the remotest times in
human history. When the Hippocratic writers speak of discharges resem-
bling beef glaze that are so corrosive that they produce skin ulcerations,
the modern physician immediately thinks of cancer of the uterus. 70 The
diagnosis is very likely, if not certain. Finally, it is difficult to know what
to make of the "disease of the genital organs" contracted by Otanes, a
Persian general of the fifth century B.c. Herodotus's narrative tells us only
that his sickness, in conjunction with a dream, induced him to repopulate
the island of Samos. 71
Chapter Six
LEPROSY
The Gradual Spreadof an Endemic Disease
Most scholars who have taken up the history of leprosy or of tuberculosis

have made the mistake of treating their subject in isolation, without con-
cern for the historical and epidemiological consequences of the kinship
between the microbes of the two diseases. This strong kinship-it consists
of a close genealogical relationship between the bacilli that cause the dis-
eases-demands an approach not unlike that of the paired lives of Plutarch:
a parallel presentation of both diseases, concluding with a confrontation
between them.
Leprosy is a chronic, infectious disease that mainly affects the skin, the
peripheral nervous system, and bones. It is caused by a specific germ,
Mycobacterium leprac)that was discovered by A. Hansen in 1871. Human
beings are the only source of contagion; its manner of transmission, which
can take place through direct contact as well as by the intermediary of
common objects or animal vectors, is still poorly understood. As a general
rule, it seems that a single infection by the microbe is not sufficient:
contamination takes place as the result of repeated and prolonged contact.
The clinical evolution of leprosy is very slow, but in the absence of 1nodern
chemotherapy the progress of the disease is inexorable. It results in defor-
mities and mutilations that provoke communal disgust. Although in the
short run it is less dangerous than an acute fever or a cancer, leprosy
produces such horrifying external effects that it was considered, especially
in the Middle Ages, no less a curse than the plague.
Four clinical forms of the disease have been distinguished, two of which
are fundamental: lepromatous leprosy and tuberculoid leprosy. The third
form is the initial stage of the disease, indeterminate leprosy, which then
152
LEPROSY 153
evolves into one of the two main types or the fourth, intermediary form.
The basic lesion: which enables early diagnosis and which ancient physi-
cians found especially striking, is a hypochromic macule, a nonelevated
spot whose degree of discoloration varies from a slight decrease in pigment
to an ashen color. The initial spots are isolated and not numerous; their
preferred locale is the face, but they can also occur on the buttocks, the
legs, the arms, or the trunk. Contour and size vary greatly. Lepromatous
leprosy is the most serious clinical form of the disease. It is characterized
by (1) papules and nodules (lepromas) that are distributed fairly symmetri-
cally over the face, on the inner surfaces of the limbs, and over the rest of
the body, (2) diffuse mucocutaneous infiltrations, and (3) erythematous or
coppery, polymorphous macules that may contain anesthetized zones.
The skin infiltrations and nodules can give the face an appearance that
evokes the head of a lion (leontiasis or facies leontina): general swelling,
erythema, pronounced wrinkles, ears becoming full of nodules and detach-
ing, hypertrophied lips and nose, loss of beard and eyebrows. Tuberculoid
leprosy is a relatively benign form whose chief skin manifestations are pale
spots that are clearly demarcated and numb. A bilateral attack on the
peripheral nerves causes pain, loss of feeling, muscle weakness, and trophic
disorders (ulcerations, resorption of bones). Intermediary or borderline
leprosy has clinical features that lie between these two polar forms of the
disease.
In order to arrive at a satisfactory explanation of the origin and initial
dissemination of leprosy and, more especially, to formulate a valid hypoth-
esis on the time of its arrival and on its prevalence in the eastern Meditter-
anean, it is necessary to survey and reconcile evidence deriving from five
distinct domains: osteoarchaeology, iconographic representations, the old-
est literary sources, the geographical distribution of the disease in modern
times, and biological considerations about the properties of the germ and
its relationship to the tubercle bacillus.
Leprosy leaves an enduring imprint on bones. The exact morphology of
these lesions was practically unknown until the osteoarchaeological inves-
tigations of Vilhelm Ms6ller-Christensen. Since 1948 this paleopathologist
has studied skeletons from a lepers' cemetery in Naestved, Denmark, and
he has described bone stigmata whose specific character has been con-
firmed clinically and radiologically on living victims of the disease. 1 Begin-
ning with the initial phase of leprous infection, there are observable,
pathognomonic lesions on the bones of the face. Leopold Gliick, a der-
matologist from Sarajevo in Bosnia, identified these lesions as early as 1897
in an anatomopathological and clinical study, but the significance of his
observations was not grasped, probably because of the lack of radiological
evidence (X-rays had just been discovered at the time). 2 So the definition
of a syndrome that is very useful for early diagnosis of leprosy had to await
154 DISEASES IN THE ANCIENT GREEK. WORLD
the work of a specialist in osteoarchaeology. It was not until 1952 that

M~ller-Christensen and his co-workers described the facies leprosa, an
osteoarchaeological entity whose medical equivalent was dubbed Bergen's
syndrome in 1953. This entity is characterized by the following pathological
conditions: atrophy of the anterior nasal spine; erosion of the lateral mar-
gins of the nose; atrophy of the alveolar processes of the maxilla, often
accompanied by loss of the incisors; inflammatory modifications of the
hard palate. The surface of the atrophied parts resembles pumice stone.
Leprosy also leaves characteristic traces on the tibia and the fibula (perios-
titis and striation) and on the small bones of the hand and foot (atrophy
beginning in the distal phalanges and progressing proximally to the meta-
tarsals or metacarpals). 3
After an examination of approximately 20,000 specimens, M~ller-
Christensen was able to confirm the presence of leprous lesions on eight
individuals who lived prior to A.D. 1000. Although they come from three
different regions, all eight cases date from the same, surprisingly late era:
the sixth century A.D. 4 The specimens from Egypt were the first described,
in 1910. They consist of two Coptic mummies found in the same spot (El-
Bigha near Aswan) and dating from the same era (about A.D. 500). One of
these mummies, a male, presents typical bone destruction on his hands
and feet. Its case was recognized as leprosy at the first osteoarchaeological
examination. The second mummy, a female, has the characteristic marks
of facies leprosa. At the time of its first examination, the existence of a
chronic rhinitis was affirmed, but its true nature could not be specified.
During a visit to Cambridge, M~ller-Christensen made a diagnosis of lep-
rosy from the skull and the small bones of this Coptic specimen. 5
I stress the absence of any trace of leprosy on the human remains from
Pharaonic Egypt. However, in a recent publication, the Polish researcher
T. Dzierzykray-Rogalski brings to light a find of truly exceptional signifi-
cance: of 31 skulls found in a necropolis of the second century B.C. in the
Daklah Oasis, 4 had typical leprous stigmata. Although the great majority
of the inhabitants of this oasis were Negroid during the Ptolemaic era, the
skulls in question are not. Situated in the Libyan desert, the Daklah Oasis
is not only the largest but also the most isolated of ,the seven major oases
of ancient Egypt. Accordingly, the Polish paleopathologist suggests that it
may have served as a deportation zone for lepers of the Egyptian elite
during the reign of the Lagides. 6
For France, about 1,000 skeletons dating from the Neolithic era to A.D.
1000 have been examined; a single skull from the Merovingian era (sixth
century) has leprous stigmata. Five cases of leprosy have been described in
bones from Great Britain dating from the period A.D. 550-650, to which
two doubtful cases from the fourth and fifth centuries must be added. 7
No leper has been found among several hundred ancient inhabitants of
the Jordan Valley (cemetery ofLachish, near Jericho); the bones were very
LEPROSY 155
closely scrutinized in the expectation of a diagnosis of the disease. 8 More-
over, all paleopathological research in the New World has failed to produce
evidence for the pre-Columbian existence of leprosy.
In 1970, John L. Angel identified a possible case of leprosy that is of
extraordinary importance, if his diagnosis is accepted. 9 The skeleton in
question is that of a man from the Early Bronze Age found at Karata~ in
Lycia (416 J(a.). The metatarsal bones of this subject, who died at age 40,
are atrophic and pitted on their anterior extremity, in a way that is char-
acteristic but not truly pathognomonic of leprosy. There is a kind of
symmetry in the lesions affecting metatarsals 1, 2, and 3 on the left side and
metatarsals 1, 2, 3, and 4 on the right, a feature that favors a diagnosis of
leprosy. Unfortunately, it cannot be confirmed by typical stigmata on the
facial bones. A nonleprous etiology for this atrophy of the lower extremi-
ties can be devised, such as a serious metabolic, toxic, infectious polyneu-
ropathy, or gangrene following a crushing wound. If this man from Karata~
is leprous, his case is doubly interesting, first, since it is the oldest concrete
evidence for the disease, and second, as proof of the early presence of
Hansen's bacillus in Greece. It is a shame that the ambiguity of the lesions
makes it necessary to consider the case a possible indicator rather than a
sure argument.
At the risk of appearing repetitious, I underline the following cautions:
the absence of traces of a disease on the bones of a region and a period
does not constitute proof of the actual absence of it in the region and
during the period in question. It is only a potent argument against the
endemic presence of the disease in question. The results of osteoarchaeo-
logical investigation do not, therefore, exclude the sporadic existence of
leprosy in the ancient Greek world, even if we refrain from a sure interpre-
tation of specimen 416 J(a.
Artistic Representationsof Leprosy

Leprosy is a disfiguring, mutilating disease. Its appearance is unforget-
table. Might not an artist's eye, like the photographer's lens, have fixed
upon the image of such a horrible disease (in horror there is always an
element of the sacred and bewitching) and might not his hand have repro-
duced it in an iconographic message that modern medicine can decipher?
Historians of disease have pored over ancient artistic representations of the
human body in the hope of discovering in them the deformities that
would justify a retrospective diagnosis of leprosy.
Such a medico-historical process is full of pitfalls. Here is an example:
since the discovery of the golden masks of Mycenae, scholars have tried to
detect the imprint of leprosy on their hieratic faces. The idea elicited
passionate debate at the end of the nineteenth century among certain
German historians. 10 Yet an unprejudiced glance at these splendid prod-
ucts of the Mycenaean goldsmiths' craft is enough to show that the so-
called pathological signs (loss of eyebrows, subsidence of the nose) are
nothing more than the consequences of stylization.
The medical and art historians who believe they have found the ravages
of leprosy on the faces of a series of He_llenistic figurines use more serious
criteria. Thus a small terra cotta statue that probably dates from the third
century B.c. has been described as the representation of an individual
whose facial marks, dress, and overall bearing show that he suffered from
leprosy . 11 This object, once kept in the Polytechnic Institute of Athens, is
now in the National Archaeological Museum there (inv. no. 5871). The
retrospective diagnosis of leprosy seems plausible, but not convincing
enough to make it certain that the figurine's creator actually saw someone
with the disease.
Leprosy has been mentioned in the description of several Hellenistic
clay heads belonging to the collections of the Louvre. One of these stat-
uettes was found at Troy, all the others at Smyrna. After examining them
personally, I disagree with Felix Regnault's appraisal of the Trojan head.
The deformities it presents have nothing to do with the usual stigmata of
leprosy. 12 The terra cotta figurines originating in Smyrna are of two sorts.
In the first, the facial expression is horrifying; according to Simon Besques,
it is "a death's head, or a leper at the last stage of the disease, with teeth
exposed, a nose without flesh, a hole in the forehead that may be a
wound. " 13 In the second, the subject's nose is crushed, its lips thickened,
its eye sockets gnarled, and its face furrowed. 14 In molding the first kind
of head, with a face deprived of flesh and without the infiltrations that
suggest lepromas, the Smyrnan coroplast wanted to create an image of
death. He produced it without being inspired by any special disease and
without necessarily having seen a leper. As for the second type, the diag-
nosis of leprosy is plausible but not at all obligatory. On these heads, most
of which express horrible suffering, there is nothing pathognomonic of
leprosy. That is only one nosological explanation among many others that
are no less likely.
Similar caution is appropriate regarding two Hellenistic figurines in lime-
stone found at Alexandria. Angelique Panayotatou presents them as an
artistic image "of the terrible disease that so ravaged Egypt. " 15 As evidence
for her diagnosis of leprosy, Panayotatou notes ''the absence of eyelashes
and eyebrows, the thickening of the skin, the nose partly destroyed, the
nostrils shut by the thickened skin, the thickening of the lips, the project-
ing upper jaw.'' 16 All this appears convincing indeed, but unfortunately
the description is based on the presumed diagnosis: it evokes the disease
with much greater force than the photographs of the objects themselves.
The rounded eyes, the absence of eyebrows, the nose more swollen than
eaten away, the thick lips, and the mouth wide open actually give the
impression of a mask, of an actor in the theater, not a sick person. This
LEPROSY 157
impression is stre!1gthened once comparison is made with some Alexan-

drian clay figures at the Benaki Museum, where the grotesque, caricatural
nature of the subjects is obvious. 17 I also recall here the possible diagnosis
of leprosy in the case of a Gallic limestone bust found at the source of the
Seine (early Christian era) . 18
Rudolf Virchow and I van Bloch believed that they recognized leprosy
on some pre-Columbian anthropomorphic vases from Peru and Bolivia
that depict facial mutilations and tumors. But the identification has not
carried conviction among specialists, who prefer to see in them uta (New
World leishmaniasis), if not bartonellosis, blastomycosis, or even syphilitic
gummas. However, a Canaanite clay anthropomorphic jar found during
excavations of the Late Bronze Age settlement of Bet She'an in Israel and
dated to about 1300 B.C.-that is, before the Hebrews' conquest of the
Promised Land-shows features resembling facies leontina and strongly
suggests a diagnosis of lepromatous leprosy . 19
The OldestLiteraryEvidenceof Leprosy

Leprosy has long been known in the literature of ancient India. 20 It may
well be that it is alluded to in a text as venerable as the hymns of the
Atharvaveda., but there is some uncertainty. The Vedic term kilasa.,which
is used in the archaic sacred texts and is customarily translated ''leprosy''
in modern Western editions and commentaries, etymologically denotes an
"ashen" disease. According to its usage in classical medical works, it de-
notes a disease of the skin (exclusively) that presents discolorations, des-
quamations, and localized inflammations. 21 From the vantage point of
modern pathology, that is a vague notion covering a host of skin ailments.
In Vedic times, kilasa may have included, in addition to other diseases of
the body surface, true leprosy. However, the semantic evolution of this
term in Indic medical literature suggests that leprosy was not its principal
reference. A "white" disease, leukodermic and psoriatic, that corresponds
to the meaning of the term lipra in the Hippocratic texts (discussed be-
low), instead of its modern sense, was the dominant element in the epi-
demiology of dermatoses in early India.
With the great Sanskrit medical treatises, namely the collections of the
Susruta and Charaka, doubt disappears: Hansen's disease is unimpeachably
described in them under the name kustha. This term of classical Hindu
nosography has two senses: a broad one, in which it designates all kinds of
cutaneous diseases (the Susruta distinguishes nineteen of them), of which
leprosy is just a subset; and a narrower sense, especially when prefixed by
the word maha 'great,' in which it applies to leprosy proper in its various
clinical forms. 22 According to the Susruta, kustha is a contagious disease
transmitted either by sexual activity or by touch and respiration or by the
use of objects previously handled by a diseased person. 23 This astonishingly
precise knowledge of leprosy among the classical Hindu physicians-

sophisticated early diagnosis, correct description of various symptoms in
their proper order of appearance, precise prognosis, and awareness of the
danger due to contagiousness-and the existence of legal measures con-
cerning lepers that go back to the fourt~ century B.c. show that Hansen's
bacillus must have been rampant in India during the first millennium B.c.
and was particularly severe during the last centuries of that era. 24
In China, leprosy is mentioned in the oldest medical treatises that have
come down to us. The famous Nei-ching) a compendium of classical med-
icine, describes a disease that ''swells and ulcerates the flesh ... produces
paresthesias ... , spoils the blood, which becomes cloudy, and results in a
collapse of the nasal structure, an altered skin color, and skin ulcera-
tions.' ' 25 The consensus of the historians of Chinese medicine is that this
passage refers to true leprosy (other identifications would not be impossi-
ble, especially given the chapter in which the passage occurs). These spe-
cialists do not, unfortunately, agree on the date of this text. The edition
we know comes from the eight century A.D., but the essential parts of the
Nei-ching certainly go back much further. The question is, How much?
For traditionalists, the text is a faithful record of the conversation of the
legendary emperor Huang-ti in the third millennium B.C. To modern,
critical historians, its origins cannot precede the fourth century B.C., and
J. Needham does not hesitate to date it only to the second century B.c.
Other Chinese medical texts that speak of leprosy (for instance, the fre-
scriptionsfor Emet;gencyof Ko Hung, a Taoist physician of the fourth century
A.D.) are all clearly later than the beginning of the Christian era. 26
In almost all the handbooks and encyclopedic reference books that treat
of the history of leprosy, it is confidently assumed that this disease raged
in China from the earliest historical periods. This conviction, which has
become general by virtue of its being repeated, actually rests on documen-
tation that is both thin and weak. Its only basis is, first, the orthodox
dating of the Nei-ching (which can no longer be upheld) and, second, the
traditional interpretation of a Confucian anecdote. In the celebrated Lun
Yu (Conversations) that were gathered by Confucius's followers, one of
the disciples, Po-niu, contracted leprosy. The master, visiting him, did not
wish to see him but restricted himself to touching the sick man's hand
through a window; Confucius then exclaimed, "Fate kills him. For such a
man to have such a disease!" 27 The event took place in the sixth century
B.C. There is no doubt of the authenticity and antiquity of this story. Its
source is respectable, but its purpose is moral, and the information it
contains is hardly satisfactory from a medical point of view. The disease of
Po-niu is not described or even specifically named: the hero suffers from
chi 'disease (in general).' The notion of leprosy was introduced by later
commentators like Pao Hsien in the first century, who accounted for
Confucius's behavior on the basis of the unpleasant appearance of his
LEPROSY 159
disciple ("the disease," says the commentator, "was a disfiguring one"). 28

It is possible that leprosy was spread throughout China during the first
millennium B.c., but the silence of the documents forces us to count on
its presence in: that part of the Far East only in the very last centuries of
the period.
Since 1937, as the result of an ingenious and bold conjecture by the
Danish missionary physician and renowned Egyptologist Bendix Ebbell, a
passage concluding an Egyptian medical treatise of the sixteenth century
B.C. (the Papyrus Ebers) has been considered the oldest literary description
of leprosy. However, that conjecture has been strongly criticized, and the
latest edition of the text implicitly rejects any possibility of such an inter-
pretation. 29 Without this evidence, there is no substantial sign of leprosy
in Pharaonic Egypt. Paleopathological research makes its endemic presence
most unlikely.
By contrast, one can reasonably share the opinion that this disease was
known in Mesopotamia as early as the second millennium B.C. Since J.
Oppert and C. W. Belser, pioneers in the decipherment of cuneiform
inscriptions, first proposed translating the Akkadian terms saharsubbu)
isrubu) andgarabu as "leprosy" or at least a disease like leprosy, Assyriol-
ogists have consistently done so. The disease in question is mentioned in
imprecations inscribed on milestones from the Kassite period. It "covers
the whole body,'' and those whom it disfigures are excommunicated,
chased outside the walls to live "like wild-asses." The only additional
information given by Babylonian texts is not so reassuring about the tra-
ditional diagnosis: saharsubbu seems to be associated with dropsy. A disease
called sibtu that is cited in the Code of Hammurabi as a sufficient reason
to cancel a contract for the sale of a slave has been identified, uncertainly
at best, with leprosy. The main argument in its favor is its parallelism with
expressions used in similar Greco-Egyptian contracts. 30 The ancient Baby-
lonians suffered from a chronic, incurable disease with symptoms on the
body surface. They considered it a contagious disease, like a tabu, and
their means of combating it was to isolate the victims of it. The disease is
not necessarily leprosy in the modern sense; the evidence that exists is
inadequate for that diagnosis. But there is more: the key to Mesopotamian
leprology is a Babylonian omen text that was published only in 1957. Here
is a translation of the decisive passage, as interpreted by J. V. Kin~ier
Wilson: "If the skin of a man exhibits white patches (pusu) or is dotted
with nodules (nuqdu), such a man has been rejected by his god and is to
be rejected by mankind. '' 31 Kinnier Wilson sees this text as solid proof of
the existence of leprosy in ancient Mesopotamia and also, even at that
time, of a clinical distinction between the two chief forms of the disease,
the depigmented macular (that is, tuberculoid) type and the nodular (that
is, lepromatous) one. This specialist in the history of Babylonian medicine
realizes that the nosology of this civilization lacks the precision required
by modern medicine and that, as a result, other diseases could be included

in the ''leprosy'' of the past, notably pellagra. This does not negate the
fact that, in his view, Hansen's disease is the major constituent of Meso-
potamian leprosy. Wilson's opinion has been criticized by Stanley G.
Browne, a competent leprologist whose judgment in this case seems con-
ditioned by his previous, strongly held position against the leprous nature
of the Biblical dermatosis. 32 In my opinion, the medical exegesis of the
omen text VAT 7525 must be undertaken by way of a careful review of the
information about leprosy among the Hebrews. The Babylonian evidence
and the Biblical prescriptions have a common origin and elucidate each
other.
Leprosyin Palestineand Persia

The exact meaning of the Hebrew word ~ara cathas been much debated.
It is common knowledge that the Bible declares "lepers" impure and
demands their isolation, or rather, their expulsion from society. According
to Leviticus, any person with a swelling, scab, or shiny spot on the skin is
under suspicion of being afflicted with a lesion of ~aracat and must be
brought before the priest (/when), who will decide if certain signs are pres-
ent or not. The presence of these signs, which are carefully described in
the Biblical text, requires that the person in question be declared impure.
If the first examination reveals questionable signs on the skin, the priest
has the person isolated for seven days. Then a second examination takes
place. If the signs on the skin are unchanged, a second period of isolation-
again, for seven days-is necessary before a definitive declaration of purity
or impurity can be made. If declared impure, the affected person must
withdraw from the community, wear torn garments, and live alone, far
from the camp. 33
To a mind not biased in favor of a purely medical interpretation for any
ancient account of a pathological state, one thing is clear: ~ara cat) the
mark of divine wrath, is not a medical notion but a ritual one. It can be
and is applied in the Bible not only to a person but also to clothing or a
house. Clearly a house can be "leprous" only by metaphorical extension
of a medical notion, while a house can be afflicted with ~ra cat in the
strict, nonmetaphorical sense of the word. The diagnosis of ~ara cat is a
matter for the priest, not the doctor, and the length of the period of
isolation is determined by the ritual significance of the number seven.
Still, the description of the signs of ~ara cat should reflect a medical
reality and refer to a concrete pathology that the historian of medicine can
try to outline as closely as possible. 34 To be sure, several details of the
clinical picture provided in Leviticus do not correspond to true leprosy.
Taken as a whole, the signs of ~ara cat correspond to no disease known to
modern dermatology. The only reasonable conclusion is that the Biblical
LEPROSY 161
text refers to a ss,t of skin diseases. Some of these are relatively benign
(psoriasis, vitiligo, steatoid pityriasis, perhaps favus and some forms of
eczema), but they have been confused with one or even a few diseases that
have serious consequences for the affected person and those around
him. In short, I am in agreement with E. Jeanselme, R. G. Cochrane,
J. G. Andersen, S. G. Browne, E. V. Hulse, and other recent authorities
who reject a pure and simple identification of ~aracat with leprosy. But if
the two terms and the notions they refer to actually belong to two different
semantic fields, and if the pathological substratum of the first cannot
simply be equated with that of the second, it does not therefore follow
that the Biblical prescripts are altogether foreign to leprosy. Leprosy could
well be an essential element of ~aracatwithout being its sole constituent.
A short time ago, E. V. Hulse drew attention to an error in the usual
translation of the Biblical passages that describe the skin of persons suffer-
ing from ~ra cat. The passages usually rendered ''white as snow'' actually
say "like snow," and according to Hulse the simile suits scales that detach
themselves from the skin like snowflakes, not the white color of diseased
skin. 35 On this interpretation ~ra cat is essentially a scaling disease whose
ritual impurity stems precisely from such "waste" or loss of integument. 36
Whatever the case may be, Leviticus speaks explicitly of patches on the
skin whose hair turns white and of lumps or nodules that recall Babylonian
divinational terms as well as the early signs of leprosy. To be sure, Leviticus
is not a medical handbook; the expulsion of' 'impure'' persons is a matter
of tabus, not infections in the medical sense. Still, medicine itself existed
only in the shadow of ritual and without distinction from it. It is hard to
believe that such a radical social rejection of persons infected with a certain
disease is simply the result of mistaken religious ideas about completely
benign symptoms. So I vigorously support the opinion of Julius Preuss in
his masterly summation of several centuries of historical exegesis: leprosy
is the only chronic skin disease whose seriousness justifies the social reme-
dies of the Biblical legislation. 37
The difficulties of a medical analysis of Leviticus stem from the fact that
the purpose of this text is not the complete description of one or more
skin diseases but the schematic specification of rules for an early, differen-
tial diagnosis of a physical state of impurity. Insofar as they are actually
interested in leprosy, the authors of Leviticus are concerned only with its
initial, indeterminate phase. So it is understandable that certain important
signs of leprosy have escaped their attention. If such a practical definition
of ~ara cat also includes some harmless skin diseases, isn't this because it
seemed preferable to err in overestimating its severity rather than risk the
consequences of dangerous diseases? I add here that at least the Talmudic
commentators know that spots of ~ara cat are numb, so there is no doubt
that for them the disease includes true leprosy.
Some recent writers have raised doubts about Biblical knowledge of
\
leprosy on the basis of the negative results of osteoarchaeological research

in Palestine. 38 It is a weak argument, at least given the current state of
knowledge. The absence of leprous remains in a great necropolis like that
of Lachish in the Jordan Valley (600 skeletons examined by V. M9)ller-
Christensen) could be due to segregati<;>nof the diseased. And the sum-
total of pre-Hebraic and Hebraic remains found throughout the Near East
in scattered cites and then subjected to paleopathological tests is still too
small to validate conclusions about the incidence of leprosy.
The priestly code that contains the prescripts against ~ara catis the most
recent section of the Pentateuch. Modern Biblical scholarship dates its
definitive form to the fifth or fourth century B.c., contemporary with the
Hippocratic corpus. However, its content is much older. The Biblical text
only codified laws that the tradition attributes to Moses and dates to the
time of the Exodus from Egypt (thirteenth century B.c.). Is there any
evidence that leprosy was known to the Jews at the time of their enslave-
ment in Egypt, or did it begin to spread only after their arrival in Canaan?
In fact, the anthropomorphic vase of Bet She'an (about 1300 B.c.) and
some Babylonian cuneiform texts make it probable that leprosy existed in
the Promised Land before the arrival of the Hebrews. It was not, then,
the Jewish people who brought it with them, but even so they may have
suffered from it before the thirteenth century. It is plain that the Biblical
remedies for ~aracat are inspired by ancient Babylonian beliefs and prac-
tices, and this influence could as easily precede as follow the Egyptian
period in Jewish history. If the latter, then the Jews first came into contact
with leprosy in Phoenicia/Palestine, where they adopted into their religion
magico-hygienic prescripts against it that the native Canaanite populations
had learned from Babylonian sources. Otherwise, the Jews would have
known of leprosy before the Exodus and probably even before their depar-
ture from Chaldea.
Historical sources favor the second hypothesis and actually affirm it, but
they are late and must be used with great caution. The information in
them may be distorted by indirect transmission, and in any case they are
often contradictory and obscure. The oldest is Manetho, a Heliopolitan
priest of the third century B.c. who wrote, in Greek, a history of Egypt
"according to the sacred books of his land." The work itself is lost, but
the Jewish historian Flavius Josephus cites extracts from it. Just like the
fragments of the Alexandrian writers Chairemon and Lysimachus, who
wrote after Manetho, Josephus's extracts concern "lepers" at the time of
the Exodus.
Manetho tells how a Pharaoh had all the "lepers" and "other impure
persons" of Egypt rounded up-80,000 of them in a single day-to put
them to work in quarries far from the rest of the Egyptians. Among them
were several learned priests, also victims of this horrible disease. With the
priests as their guides, especially one who changed his name to Moses, the
LEPROSY
impure left Egypt,,, joined the shepherd-kings of Jerusalem, and made war
against the Pharaoh. Chairemon tells of 250,000 "polluted" men who
were exiled from Egypt as the result of a prophetic dream that came to
Amenophis. And according to Lysimachus, "the Jewish people, afflicted
with leprosy, scabies, and other diseases" were eliminated from Egypt in
the time of King Bocchoris: the "leprous" and "itchy" were drowned
and the rest were chased into the desert. The accounts of Manetho,
Chairemon, and Lysimachus, which I have given in bare resume, are full
of unlikely events and deliberate slander of the Jews, especially in the case
of Lysimachus. Josephus (first century A.D.) has no difficulty pointing out
their contradictions and inconsistencies. However, he casts no doubt
whatever on the presence of leprosy in Egypt at such an early period. It is
not the presence of the disease among the Jews that offends him but the
statements that Moses was afflicted with it and that only Jews contracted
it. 39 Josephus is aroused because Manetho deliberately "confuses the Jew-
ish people with a collection of leprous or otherwise diseased Egyptians.' ' 40
Several other ancient authors echo the accounts Josephus cites, especially
Tacitus, who repeats the slanders of Lysimachus, Justinus, who does the
same, and Diodorus Siculus, who explains the Exodus as the result of a
plague that broke out of old in Egypt. The Egyptians blamed it on the
Jews, who were chosen as scapegoats because of their strange customs and
ceremonies. 41
In presenting opinions from ancient sources on the causes of the Exo-
dus, I have taken care to put the words "leprous" and "leprosy" in
quotation marks. In the extracts of his work in Josephus, Manetho desig-
nates the disease in question by the Greek word lipra and calls those
afflicted with it hoi lepro{. Lysimachus uses the terms lipra and ps-f».a.
J ustinus translates them into Latin as vitiligo and scabies)while Tacitus,
more descriptively, calls the disease tabesquae corporafoedaret (a corruption
that befouls bodies). Josephus himself translates the Biblical ~ra cat by the
term lipra)42 as does the first translation of the Pentateuch, the Septuagint,
which was produced in Alexandria during the third and second centuries
B.C. This is how the word "leprosy" came into the Vulgate and acquired
its medieval and modern meaning. But the problem for the historian of
diseases lies in the surprising revelation that in strictly medical treatises
from antiquity the term lipra has nothing whatever to do with the disease
we now call leprosy. Was it the wish of these Jewish and Egyptian authors
writing in Greek during the Hellenistic and Roman era-Manetho,
Lysimachus, Josephus, and the authors of the Septuagint-to identify the
Biblical term ~ara cat with the relatively benign skin disease called lipra in
the Hippocratic writings? Or did they use this word for lack of a better
one to designate a disease still without its own name in the current medical
terminology?
A passage in Herodotus, the fifth-century B.C. Greek historian, is en-
lightening in this regard. Of the Persians, he says that "if one of their
fellow-citizens has the scaly-disease (lipra) or the white disease (leuke))he
does not go in town or trade with the other Persians; they explain that he
suffers these ills for having sinned against the Sun; any foreigner with these
diseases they expel from their land. " 43 These Persian customs recall the
Babylonian and Biblical prescripts. The comparison of three families of
historical sources suggests the identity of the ailments Herodotus speaks
of with the nuqdu and the pusu of Babylonian divinational formulas as well
as the ~ara/atof Mosaic law. But why, then, does Herodotus use the word
lipra in this context, a word used by Greek physicians of his time for a
benign disease that is much too banal to justify legal repression? I believe
that an explanation lies in the absence of precise technical terms for
Hansen's disease in the medical vocabulary of the fifth-century Greek.
When forced to inform his audience of pathological states still unknown
in his own land, Herodotus, like all good ethnographers, uses approxima-
tions, words from his own language that express the essential sense of
foreign terms. It is revealing that words like leukeand lipra correspond in
their etymological senses to the Babylonian terms pusu and nuqdu) with
the former referring to white spots and the latter to roughness of the skin.
From Herodotus's choice of words to render the sense of the Persian
nosological terms, one can conjecture that the learned caste of Persia knew
of the two chief forms of leprosy (in the modern sense), a knowledge
surely drawn from the sacred traditions of ancient Babylonia. One of these
two forms, called leuke)is characterized by pale spots, while the other,
lipra) features swollen lumps or nodules. Herodotus's lipra) then, is a
rough, scabrous disease rather than the scaly one modern commentators
have in mind. Later on, in the second century A.D., the historian Justin us
applied exactly the same mental process to the translation of Greek terms
as Herodotus applied to the Persian ones. As mentioned above, he trans-
lates the twofold terminology of leprosy into the Latin words vitiligoand
scabies)with due respect for their original meaning, not for their current,
strictly medical usage.
Aeschylus mentions a disease comparable to leprosy called leikhin.
Orestes is in danger of being afllicted with it for neglecting to avenge his
father. The disease attacks flesh and devours the body, covering it with
white splotches. Coming from Apollo, it marks the man afllicted with it
as proscribed by divine law. 44 But it is a mythic disease with a deliberately
vague,, and emotive profile. In classical Greek medical literature, the word
leikhen designates various benign rashes and has absolutely none of the
seriousness ascribed to it by Aeschylus. Isn't it likely that the tragic poet
had access to the same information as Herodotus on the ancient customs
of the Persians and likewise found himself without a proper medical term?
To my mind, his verses confirm my in tuition that true leprosy was only
rarely observed by practicing physicians in the Greek world.
LEPROSY 165
Lepra, Leuke, Alph6s, and the PhoenicianDiseasein the

HippocraticCorpus
The etymology of the Greek noun lipra poses no problems. It is derived
fron1 an adjective lepr6s(rough, scabrous) whose root is a verb meaning
"peel, strip off bark or skin. " 45 The oldest attestation of this adjective in
the sixth-century poet Hipponax has nothing to do with medicine, but in
the Hippocratic corpus lepr6soccurs with its derivatives in a precise tech-
nical sense. It refers to a particular pathology of the integument and not
the scaly aspect of surfaces in general. The original sense has been displaced
\Vithout being altogether lost, as is clear from Nicander's use of the word
leprunomai in a medical poem to describe the normal state of the skin of
certain snakes. 46
In the Hippocratic texts the disease called lipra comes up often, but
unfortunately it is never the subject of a description precise enough to
ensure its identification with a clinical entity of modern medicine. What
can be gleaned from the medical treatises runs counter to its being a
mutilating disease like the one caused by Hansen's bacillus and even sug-
gests relatively benign skin ailments. If we take account of the word's
etymology as well, the first possibilities that come to mind are psoriasis
and eczema. In classifying the main diseases by their prognosis, the treatise
Diseases) I, places "leprosy" next to "arthritis" and "lichen" among
changes in health without serious consequences, that is, not entailing
death. 47 Aphorisms teaches us that "the diseases of spring are maniacal,
melancholic, and epileptic disorders, bloody flux, quinsy, catarrh, hoarse-
ness, cough, leprosies, lichens, loss of skin pigmentation (alph6s),exanthe-
mata mostly ending in ulcerations, boils, and arthritic affections.' ' 48
Hippocratic "leprosy" first presents itself as a kind of eruption on the
skin's surface, an abscession (ap6stasis)on the skin that results from the
expulsion of humors in disequilibrium. For this reason, Hippocratic phys-
iopathology considers hemorrhoids a guarantee against "leprosy. " 49 But
when the abscession of corrupt humors is not expelled downward, there
is the danger of various pathological phenomena such as "suppurations,
ulcers, exanthemata, peeling of the skin (Mpoi))loss of hair, loss of skin
pigmentation (alph6s))leprosies, or their like-ills that will settle by way of
considerable movement. " 50 As abscessions, some skin diseases such as
ulcers, suppurations, "leprosies," are at times useful for the health of the
whole organism, at times harmful, and at other times without conse-
quence. 51 When all is said and done, the Hippocratic physicians fear the
suppression of a "leprous" eruption more than they fear its initial appear-
ance. Treatment is contemplated only in exceptional cases, and even then
only the mildest remedies are used: no caustics, no revulsive agents, only
the topical application of salt dissolved in vinegar 52 or "quicklime in water,
so as not to produce ulceration. " 53
The Hippocratic writers already had a sense of the difference between

simple surface reactions on the skin and true dermatoses. Their "leprosy"
can be one or the other, depending on the circumstances. 54 If it appears
suddenly in the adult, it is only a passing purificatory eruption. But in its
chronic, evolving form, "leprosy" is considered,,.
a disease in the fullest
sense. According to the J+orrheticon)leikhenes) leprai) and leukai are all
pathological phenomena of the same type, "and if one of these appears in
youth or childhood, or it grows little by little over a long period of time,
the eruption must be considered not an abscession but a disease (n6sema);
on the contrary, it is an abscession when the eruption 1nakes a sudden,
massive appearance." The same paragraph goes on to say that "leprosies
and lichens are atrabilious diseases" and that "the younger the patient and
the tenderer or plumper the affected body part, the easier the cure. " 55
Another Hippocratic text asserts that "certain leprosies and articular ail-
ments provoke itching before a rainfall. " 56 The regular use of the plural
form leprai highlights the absence of the real nosological unity of this
concept within the framework of Hippocratic medicine. Instead, the term
refers to skin and mucous membranes of a particular appearance that may
well correspond to a variety of pathologies. That is why Greek physicians
can speak of a "leprous" disease of the bladder in the case of an inflam-
mation (cystitis) with exfoliation; or of an appearance "comparable to that
of leprosy'' in the case of an Athenian suffering from thickening skin and
general itching. 57 In fact the Hippocratic term is so polysemous that it
even includes phenomena outside the proper domain of pathology. In a
stationary-that is, chronic and not evolving-state, the skin's "leprous"
appearance betokens neither abscession (ap6stasis)nor disease (n6sema)but
merely represents an aesthetic blemish (atskhos).This explains the passage
in the treatise Affections according to which "leprosy is a deformity rather
than a disease. " 58
For Theophrastus, leprais a skin manifestation without consequence for
the overall health of the body. He mentions it along with "scabies" (ps7Jra)
and "lichen" (leikhin). And in the Characters he describes the "disgust-
ing'' individual who neglects to care for his body as repulsive in appearance
because of his rashes and pimples, his black fingernails, his lepra) and so
forth. 59 Clearly, this "leprosy" is no great disease. In another work that
belongs to the Aristotelian tradition, the J+oblemat:a) it is stated that lepra
can pass from a sick person to a healthy one just like scabies because they
are both diseases of the surface. 60 The concern here about the contagious-
ness of "leprosy" is absolutely foreign to the Hippocratic writings, but
even in this new context the word lepradoes not seem to refer to a serious,
mutilating, incurable disease.
In the Greek medical writers of the classical period, leprais often associ-
ated with two other skin ailments, leuke and alph6s. Judging from their
etymologies, these diseases were characterized by local loss of pigmenta-
LEPROSY
tion. The Hippoc~atic texts do not justify the conclusion that leuke and
lipra are affections that often occur together in individual patients but only
that the association of their names is due to a similarity in their nature.
Pausanias tells us that the name of the town Lepreon in the Peloponnesus
implies that its founders suffered from "leprosy." In the town's vicinity,
he writes, was a "grotto of the Anigrides Nymphs. Anyone affiicted with
alph6sor leuke who entered it was first obliged to pray to the Nymphs and
promise them a sacrifice, then to wash his body; then, by crossing a
stream, the worshipper left behind his sickness and emerged completely
cured, his skin healed." In fact archaeological investigation has shown that
a spring existed there and that its water was sulfurous: therapeutic baths in
sulfurous water are actually effective in treating some relatively benign skin
diseases. 61
The Hippocratic usage of lipra and leuke cannot be taken the same way
as the Herodotean, since the two terms do not designate two complemen-
tary forms of a single disease. Hippocrates includes in the same class with
them "lichens," scabies, exfoliations, ulcerations and so forth. If "lep-
rosy" and "lichens" arise from perturbation of the black bile, as the
classical theory of humors avers, leuke) by contrast, is caused by phlegm. 62
Acquired leuke does not develop before puberty, and it must be distin-
guished from congenital leuke.63 Aristotelian texts and the case of Atossa
show that the terms leuke and alph6sapply to diseases that are not especially
serious. 64 However, Aristotle's leuke) a skin disease that whitens all body
hair, does not entirely match the nosological domain of the Hippocratic
term. For the author of I+orrheticon)II, considers "the leukai among the
fatal diseases, like the so-called Phoenician disease. " 65
We are now in a different arena: after benign ailments of the skin that
have more to do with cosmetics than medicine, dread and deadly diseases
confront us. What in the world is this so-called Phoenician disease that is
mentioned in passing and not without some horror in the chapter on skin
disease of a Hippocratic treatise? It is a serious disease, in some way be-
longing to the same clinical category as leuke and lipra) scarcely known in
Greece but apparently thought common among the Phoenicians. That is
all we know with certainty, since the expression nousosphoenikze occurs
only here in the CorpusHippocraticum.
Galen comments on this p~ssage, "Phoenician disease: common. in
Phoenicia, and in other Oriental regions; the disease in question seems to
be elephantiasis. " 66 What Galen means by elephantiasis is without any
doubt what medieval authors call leprosy, that is, Hansen's disease. The
wording of the gloss betrays Galen's embarrassment: his first statement is
banal, his second a guess projecting on the past the epidemiological reality
of Imperial Rome. Though it cannot be proven, Galen's hypothesis still
remains the most plausible explanation of the true nature of this mysteri-
ous disease.
For Greek physicians of the fifth and fourth centuries B.C., true leprosy
seems not to have been a disease they encountered in the normal practice
of their profession. It was only an exotic disease limited to eastern lands or
sporadically arriving in Greece from them. A "pestilential" (loim~des))that
is, contagious, form of leuke is said to have afflicted the Delians. "Their
faces were covered by a leuke) their hair grew white, their necks and chests
swelled, but they felt no fever or great pain, and their lower limbs re-
mained utterly unharmed." It was thought to be a punishment by Apollo
for their having buried a man on the island, a violation of sacred law.
Perhaps this story reflects true leprosy, but what date can we ascribe to it?
It is found in a letter addressed to Philocrates by Aeschines, the famous
orator of the fourth century B.C. Nowadays the letter, which is said to
relate the latter's travails on a journey, is considered a forgery not older
than the second century A.D .. 67
Elephantiasis:The Outbreak of True Leprosyin Europe

In the works that have come down to us, the terms elephasand elephan-
tiasisdo not occur in a nosological sense until the first century B.C. (Lucre-
tius has a Latin translation of them). They then become current in the first
century A.D. (Dioscorides, Celsus, Aretaeus, Scribonius Largus, and Pliny).
However, the fragment of a treatise of Rufus of Ephesus transmitted to us
in the great medical encyclopedia of Oribasius (fourth century) makes it
possible to go back, for the history of this disease, as far as Straton, the
pupil and secretary of the physician Erasistratus, who lived in Alexandria
during the third century B.C. 68 Here is the text in question:
The ancients have taught us nothing about elephantiasis; there is reason to wonder
how such a serious and common disease escaped the notice of men capable of
pondering everything in the tiniest detail; only Straton, the disciple of Erasistratus,
has provided us with some notion of this disease, which he calls cacochymia; for
the book concerned with this disease, which is attributed to Democritus, is plainly
apocryphal. The physicians who lived shortly before us established the types of
this disease: they called it, in its early stage, leontiasis, because those affected with
it take on a bad odor, and their cheeks collapse, and their lips thicken; but when
their eyebrows swell and their cheeks are flushed and they are seized with a desire
for sexual gratification, these physicians call it satyriasi~, which, however, is differ-
ent from the disease of the genitals called by the same name; for the latter has
gotten its name from a continual erection of the genitals, while the former has it
also from its character; when the symptoms invade the whole body, the physicians
in question call it elephantiasis. Now its symptoms are not hidden: they consist in
livid and black embossments that resemble ecchymoses; some on the face, others
on the arms, still others on the legs; many also develop on the back, the chest,
and the stomach; at first, the embossments are not ulcerous; later, they ulcerate in
the most hideous way, since their ulceration is accompanied by a swelling of the
lips and so deep a decay that in some cases the tips of the fingers fall off and the
ulcers never succeed in scarring. So it appears to be a superficial disease, since it
LEPROSY
makes its appearance on the skin; but the difficulty of curing it, a difficulty that
comes close to impdssibility, suggests to us that it has a deeper origin, an origin
not easy to penetrate; it is even as deep as that of carcinoma, by common opinion;
in truth, Praxagoras accepts a deep origin above all for carcinoma. 69
This enumeration of the symptoms of elephantiasis by Rufus of Ephesus

(early second century A.D.) corresponds perfectly with the clinical profile
of low immune resistance leprosy. 70 Straton is said to have provided the
first information on this disease in Alexandria, but only the Greek physi-
cians who lived shortly before Rufus-that is, not until Roman times-get
credit for describing it well, distinguishing among its phases, and naming
them. With a single exception, 71 elephantiasis was not related to or iden-
tified with the Hippocratic lipra or leulie)even though these names con-
tinue to be used in Greek medical literature of the Hellenistic and Roman
periods to denote dermatoses such as eczema, psoriasis, and vitiligo.
The Seventy were victims of a regrettable confusion, unless those erudite
rabbis actually thought that Biblical ~ra/at corresponded more closely to
ancient Hippocratic lipra than to the serious ailment that was gaining the
attention of physicians at that very time and in the very city in which they
were laboring. Whatever the case may be, it is unlikely that these transla-
tors had at their disposal a generally accepted Greek medical term that
signified leprosy. Straton's cacochymia is more a pathogenetic explanation
than a specific name for a disease. And the satyriasis of which Aristotle
speaks in the Generationof Animals is not necessarily leprosy, even though
the diagnosis cannot be excluded with certainty. 72 The novelty, variety,
and vagueness in the terminology prove that the Greek physicians were
grappling with a disease that they perceived as a new arrival and concerning
which they found no adequate information in the great masters of the
past.
To Lucretius, the Roman poet and natural philosopher who died in 55
B.C., elephasmorbus was still a strange and foreign disease confined to the
banks of the Nile. 73 About two generations after him, Celsus can still
affirm that '' elephantia) which the Greeks call elephantiasis)is a chronic
disease hardly known in Italy and very common in certain lands. " 74 Pliny
the Elder expressly states that it is a "new disease," one of those that, like
lichen or mentagra) carbunculum and colum) ''were unknown in past years
not only to Italy but also to almost the ,vhole of Europe. " 75 He continues:
Elephantiasis did not occur in Italy before the time of Pompei us Magnus, and ...
though the plague usually begins on the face, a kind of freckle on the tip of the
nose, yet presently the skin dries up over all the body, covered with spots of
various colours, and uneven, in places thick, in others thin, in others hard as with
rough itch-scab, finally however going black, and pressing the flesh on to the
bones, while the toes and fingers swell up. The plague is native to Egypt. When
the kings were aftlicted, it was a deadly thing for the inhabitants, because the tubs
in the baths used to be prepared with warm human blood for its treatment. This
disease quickly died out in Italy. 76
The evidence I have been citing in fact only concerns the presence of
leprosy in Egypt and Italy. As for Greece proper, the information is some-
what later and less explicit. Aretaeus of Cappadocia, Galen, and Plutarch,
who all speak often of elephantiasis, would certainly have stressed its ab-
sence from the Greek world if such had been the case in the first and in
the beginning of the second century A.D. Plutarch takes up the subject in
a chapter of Quaestionesconviviaksin which he undertakes to consider ''if it
is possible that new and unknown diseases can arise.' ' 77 The question is
debated at a symposium in which the physician Philo, the philosopher
Diogenianus, and Plutarch himself speak one after the other. Philo main-
tains that "elephantiasis has only been known for a short time, since no
ancient physicians speak of it." Plutarch takes an opposing position, prof-
fering the evidence of Athenodorus, the author of a treatise on epidemic
diseases according to whom "not only elephantiasis but also hydrophobia
were first observed in the time of Asclepiades." For the modern reader,
Plutarch's remark runs counter to its stated intention: it confirms the
newness of leprosy for physicians in Greece instead of debunking it, since
Asclepiades of Bithynia lived around the year 100 B.c. and, moreover, his
writings basically reflect the state of affairs in Rome. Plutarch's dinner
guests were divided between two opinions that they found equally surpris-
ing: either such diseases really appeared for the first time and there are,
therefore, changes in nature itself, or all diseases have always existed but
they have not al,vays been noticed by physicians. Although such a blind
spot is astonishing for a disease as great and serious as elephantiasis, most
of the participants in the symposiun1 plump for the second opinion, "not
being able to believe that nature, in the human body as in the city, can
enjoy inventing novelties." Diogenianus makes a peroration to this effect,
emphasizing that psychic phenomena and, a fortiori) somatic ones do not
change as a whole but simply vary within certain limits. The constancy of
causes and the conservative character of the very nature of the living organ-
ism are opposed to fundamental innovation in the domain of nosology.
To explain the arrival of apparently new diseases, one has only to consider
the pathological variations brought on by changes in diet and in bathing
habits. Plutarch cites, only to reject it, the hypothesis of Democritus ac-
cording to which the spillage of extraterrestrial atoms could be the cause
of epidemics and unusual diseases. 78 The rest of the discussion only sec-
ondarily concerns the history of leprosy.
Though the descriptions of elephantiasis ~by Celsus, Pliny, Rufus of
Ephesus, and Galen validate the retrospective diagnosis of lepromatous
leprosy, they leave much to be desired with regard to precise knowledge
of the clinical unfolding of the disease and the possibility of confusing it
with other disorders of lesser proportions. Still, by the first century A.D.,
Aretaeus of Cappadocia, a physician thought to have studied at Alexandria
and to have lived at Rome, had given a thorough description of the clinical
LEPROSY r7r
characteristics of eJephantiasis, a disease that he thinks is as great, fright-

ening, and enduring among diseases as the elephant is among animals. 79
True, Aretaeus has some wrong ideas about the etiology and pathogen-
esis of leprosy, but he know perfectly well its main symptoms, their clinical
course, and their hopeless prognosis. His description of the insidious start
of the disease, the lepromas, the facies leontina, and the final mutilations
is masterful. He still confuses true leprosy with some manifestations of less
serious dermatoses (whence, for instance, his mention of alopecia or local-
ized lichenoid eruptions on the fingers and joints), but these are trifling
objections: the clinical picture he paints has great rigor and is on the whole
very distinct. It guarantees an unequivocal diagnosis. The reason Aretaeus
is ignorant of the anesthetic patches in tuberculoid leprosy must be that
his description is based on direct experience with the low immune resis-
tance form of the disease. Even so, one can detect in his account a certain
knowledge, albeit vague and shadowy, of dimorphic leprosy.
Pliny's conclusion about the rapid extinction of leprosy in Italy was
overly optimistic. Far from being extinguished, henceforth this disease
ceaselessly and surreptitiously spreads throughout the Roman Empire. But
it was to become an important endemic pestilence in Europe only after
the collapse of ancient civilization and the abrupt change in the conditions
of life that marks the beginning of the High Middle Ages.
More and more, leprosy evokes a holy terror that is fostered by the
actual propagation of the disease as well as by the spread ofJudeo-Christian
religion. Aretaeus of Cappadocia gives eloquent voice to the fear that
leprosy inspired in common people:
Who would not wish to escape these diseased people, and who does not turn away
from them in horror, even if one's own son, father, or brother is among them?
We fear the transmission of the disease (nietadosistou kalwu). For this reason, there
are those who abandon their most cherished relatives in the desert and the moun-
tains, from time to time bringing them provisions, or they leave off doing even
that and let them perish. so
Though modern medicine justifies this common fear that ancient physi-
cians were at pains to accept, there is a strange paradox in the situation:
the contagiousness of leprosy was recognized, though it is a disease with a
relatively low risk of infection, while the contagiousness of most acute
fevers was denied. 81 The underlying motivations for flight from lepers and
their forced isolation were at first basically magico-religious in nature.
The first leprosaria in Europe date from the sixth century A.D., if not
the fifth. Gregory of Tours (about 538-594) speaks of "ladreries," leper
houses, in France. The edict of the Lombard king Rothari (643) imposes
social death on lepers and regulates their settlement on the margins of the
community. From the sixth century on, the lepers of Rome were crammed
into a hospice called Saint-Lazarus. 82 Walafrid Strabo (808-849) attributes
to Saint Othmar, abbot of the Benedictine community of Saint Gall, the
founding of "a small hospice, situated not far from the monastery but
removed from the homes of the rest of the poor, and devoted to the
housing of lepers who customarily live apart from other people.'' That
took place around 736. During the same century, Pippin the Short (in 751)
and Charlemagne (in 786) intervened to strengthen the social repression of
leprosy. But none of the measures they took succeeded in stemming the
progress of the disease. 83 There is no agreement among modern historians
as to the effect of the Crusades on the spread of leprosy. It is certain that
the disease was present in the West before they took place, and yet en-
demic leprosy reached its apogee between the eleventh and fourteenth
centunes.
Medieval leper houses were for rounding up the sick and localizing the
social canker they represented, not for providing care. Lepers were thought
to be incurables. Actually, people did not even suppose that complete
isolation of lepers was possible. In the available evidence, there is nothing
to show that lepers were shut in or kept by force in certain places. They
were able to circulate in the region but had to avoid urban areas, to refrain
from entering homes or churches, and to give notice of their presence
from afar. So, for example, in the Republic of Dubrovnik, they had to
wear white. From time to time, in fact very often, the measures taken
against them were relaxed. Then lepers invaded towns and took up their
places in the midst of street crowds, where begging was profitable. The
people of medieval times believed that leprosy was contagious, but that
belief was tempered by a specific restriction: the transmission of the disease
should obey causes of a moral order. To a medieval person, leprosy was
above all a form of divine punishment, more an affliction of the soul than
a disease of the body. 84 Byzantine physicians (first among them, Paul of
Aegina in the seventh century) and Arabic-speaking physicians (notably,
Abul-Qasim, who lived in Andalusia in the tenth century) realized the
importance of the neurological symptoms of leprosy. Was this due to a
greater exactness in their clinical observations, or a real change in the
disease itself? I opt for the latter hypothesis: the disease was really changing
its appearance, since the high immune resistance forms, that is, tubercu-
loid leprosy, were becoming more and more frequent.
Nosological terminology itself also changed in the course of the Middle
Ages. Elephantiasisquam vulgus lepram vocat) says a passage in the Collectio
Salernitana.85 The homilies of Gregory of N azianzus show that in the
fourth century the words elephantiasisand leprawere synonyms, at least for
those not initiated in the medical art. The Second Synod of Orleans (549)
consecrated the legal use of the terms lepra and leprosus.The physicians
bowed to the vox populi) or rather, the vox ecclesiae)and agreed to the
nosological term leprain a sense closer to that of Greek and Latin transla-
tions of the Bible than the Hippocratic tradition. The term elephantiasis
was not abandoned, but instead, and still worse, it was reused to denote a
LEPROSY 173
totally different dis_...ease,

namely a lymphatic ailment accompanied by swell-
ing of the limb affected and roughening of its integument; the main cause
of this disease was infestation by filaria, nematode worms of the warmer
regions.
Origin and WorldwideDistribution of Leprosy

Although leprosy as an endemic disease was extinct in Europe long
before the effective use of chemotherapy against it, it is still rampant on a
global scale. The number of lepers in 1982, according to the World Health
Organization, is surprising: 15 million. The disease has four main epidemic
foci: India and Southeast Asia, China, black Africa and Madagascar, and
Latin America. For the record, one should add the remnants of its medi-
eval endemic in Europe, especially in Greece, southern Italy, Portugal,
and, until recently, Scandinavia. Current estimates are that there are ap-
proximately 1,000 lepers in London and at least as many in France, though
they are imported cases. Little is said of them in public, and people have
no direct experiences of these victims, but that is a conspiracy of silence.
The seal of infamy still rests heavily on this disease. 86
According to some authorities, for instance, D. Zambaco Pacha and
G. Barbezieux, leprosy is a universally distributed disease as old as human-
ity itself. Its geographic limits and historical outbreaks are simply due to
ups and downs in social conditions. 87 This view is no longer shared by
specialists in the historical pathology of leprosy. Though it is true that the
geographic distribution of leprosy is for the most part coterminous with
the third world, this need not mean that in the past social conditions were
not only necessary, as they are today, but sufficient causes for an outbreak
of endemic leprosy in a region. The distribution of leprosy in ancient times
depended not only on appropriate social environments but also on the
special history of Hansen's bacillus and its migrations.
Some foci of leprosy are obviously older than others. The disease was
probably born after the global dispersion of Homo sapiens.However, even
though it is possible to reach general agreement that the introduction of
leprosy on the American continent and in Oceania is relatively recent, the
question of the disease's original focus remains open. It was first brought
to America by Spanish and Portuguese sailors, then by the black slave
traders, and last by Norwegian immigrants. Though it has been invading
the New World since the sixteenth century, it never took root in the
northern continent, which was predominantly populated by Anglo-Saxons.
Eskimos and certain other isolated Amerindian peoples have never known
leprosy. In Oceania, it dates only from the nineteenth century. The cur-
rent situation in western Europe and North America, in which continuous
immigration of lepers and the absence of special measures for the protec-
tion of society against the disease have had no serious epidemiological
consequences, can serve as a model and helps to explain why, or at least

makes it plausible that, leprosy did not exist in an endemic state in the
Greek and Roman world during the archaic and classical periods despite
the introduction from that time on of sporadic cases.
As for the place where leprosy first appeared, it may be black Africa, or
the Far East, or the Middle East. In asking himself whether leprosy origi-
nated in Africa or Asia, R. Chaussinand states that "this problem is part
of a larger problem concerning the origin of the human species. " 88 From
this point of view, the most seductive hypothesis is surely the African one.
According to T. A. Cockburn, leprosy is an infection with a mycobacte-
rium inherited directly from the ancestors of man. 89 The place of origin of
the human race, which recent anthropological research situates in East
Africa, would thus coincide with that of the origin of leprosy. But this
argument does not seem to me convincing. The hypothesis of the vertical
transmission of leprosy from hominids to humans, taking place during the
Pleistocene era along with the evolution of Homo sapiensJis far from proven
and even seems unlikely. It comes up against two difficulties: the absence
of a common leproid affection in existing primates (or at least our igno-
rance of any such) and the nonubiquity of leprosy, a fact that suggests a
hypothesis of the appearance of Hansen's bacillus after the first coloniza-
tion of the New World.
Another argument in favor of African origins is provided by immunol-
ogy. The high frequency of high immune resistance leprosy among the
autochthonous inhabitants of black Africa suggests that this geographical
area is an ancient endemic focus of the disease. For instance, in Senegal,
there are ten times as many lepers as in Mexico (calculating, to be sure, by
thousands of inhabitants), and yet the number of lepromatous cases is
about the same in both countries. Approximately ro percent of the lepers
in Senegal have lepromatous leprosy, while in Mexico 80 percent are evolv-
ing toward the more serious form. 90 It is tempting to interpret these data
as a sign of the greater antiquity of African leprosy in comparison with
American leprosy. However, that is not the only possible explanation, and
in any case the situation is more complex if we compare African morbidity
with that of Asia (and not that of the New World). I doubt that this
argument can be valid when extrapolated into a time as remote as that of
the formation of the germ of leprosy. In the current state of knowledge,
immunology cannot decide the case between Africa and Asia.
Those who suppose that leprosy originated in Asia base their belief on
the antiquity of the literary evidence for the existence of the disease in the
Far East and Mesopotamia. The Indian focus is probably older than the
Chinese one, but there is still no way of deciding whether India, Meso-
potamia, Persia, or East Africa should be considered the cradle of the
disease. These contiguous lands forn1 a vast domain all of which was in-
vaded by leprosy in the beginning of the second millennium B.c. at the
LEPROSY 175
latest, but it remaips in1possible to say where inside it the disease was born
and how it was first spread.
From the critical examination of historical sources undertaken above, it
emerges that endemic leprosy spread from Mesopotamia toward the west
up to the Asian shores of the Mediterranean, that is, up to Phoenicia.
Around the fourteenth century B.C., it was present in Canaan (Phoenicia-
Palestine), brought there either by Babylonians who came fron1 Chaldea
in the eighteenth century B.C., or by Hebrews who also came from there
and mingled with the Canaanite populations (seventeenth to sixteenth
centuries). One Hebraic tribe descended into Egypt, returning from it later
on to take over the Promised Land around the twelfth century B.c. The
Biblical prescripts about ~ara cat date from this period. Egypt, Greece, and
Italy-actually, the whole Mediterranean except the Phoenician coast-
were still free from leprosy around the fifth and the first part of the fourth
century B.c. A certain tradition of much later date than the events of
which it speaks held that the Hebrews, during their sojourn in Egypt,
were infected with leprosy and brought back the disease to their new
homeland. That is possible, but it is more likely that they found it in
Canaan, where they also may have left it in the first place. It remains true
that leprosy did not take root in Egypt at such an early date.
Leprosy respects a certain form of civilized life and does not take up
residence in all societies, even if it is introduced into them sporadically.
Certain sociocultural conditions whose essential parameters are still un-
known, though they are plainly correlate with misery, close quarters, and
a low level of hygiene, are needed for the transformation of sporadic into
endemic leprosy. It is indeed instructive that the disease spared Greece and
Italy for a long time despite the significant contacts these countries had
with Phoenician sailors and Persian soldiers and merchants.
Leprosy did not acquire visibility to Greek medical eyes until around
300 B.C. At that time, the endemic of leprosy in Egypt had already taken
on considerable amplitude. It was especially rife in Alexandria, the new
city. For all that it was the most brilliant instance of ancient urban life and
the center of Hellenistic civilization, Alexandria, like Rome after it, shel-
tered its share of hovels in the shadow of grandiose monuments. As a
result of the concentration in this city of the poor and the sick as well as
the learned, the inevitable finally took place: leprosy was seen and taken
for the special calamity it was.
Modern leprosy specialists maintain that the disease was brought to
Egypt from India by the armies of Alexander the Great on his return from
the great eastern expedition of 327-326 B.C. 91 Indeed, there is nothing to
prevent us from thinking that the wars of Alexander, like the campaigns
of Darius and Artaxerxes before him, contributed to the spread of leprosy
into the West. However, there is no proof at all that it was the first
introduction of leprosy into Northeast Africa, nor that Alexander's sol-
diers caught the disease in India. It is very likely that leprosy had long since
existed in Persia and at Babylon, where Alexander succumbed to an attack
of fever in 323 that was almost certainly malarial.
At the beginning of the Christian era, Roman legions brought leprosy
to the heart of the Occident. If Pliny is.to be believed, it began expanding
into Europe at the time of Pompey, who, once he had conquered Mith-
ridates of Pontus, put down Armenia, and crossed the cities of Greece in
triumph, then disembarked on the coast of Italy at the head of an army
laden with microbes from the East (61 B.c.). There, for the moment at
least, my inquiry into leprosy will end. I will return to it after a glance at
the other panel of this medico-historical diptych: the history of
tuberculosis.
Chapter Seven
TUBERCULOSIS
A Great I(iller
The length of this chapter is not in proportion to the importance of

tuberculosis in the history of humanity in general and the history of Greek
and Roman antiquity in particular. If the exposition is slightly shorter than
that in the previous chapter, which is devoted to leprosy, the reason lies
in the less complex and less ambiguous nature of the sources and, above
all, in the quantity and quality of the monographs on the history of
tuberculosis. 1
The Clinical Ibrms and Paleopathologyof Tuberculosis
Currently, tuberculosis is defined as a disease that results from the in-

flammatory reaction of organic tissue to infection with a specific microbe
discovered in 1882 by Robert Koch (M ycobacteriumtubercuwsis),or by cer-
tain microbial species very close to it. For clinical nosology, tuberculosis is
not, strictly speaking, a disease, since it has no symptomatic homogeneity.
Because it is defined exclusively by its etiology, tuberculosis includes a
quantity of diseases that differ among themselves in the place affected, the
symptoms, and the seriousness of the prognosis. Aside from a specific
microbe, the tuberculous diseases also have in common the histological
appearance of the fundamental lesion, namely the tubercle and the caseous
destruction of cellular structures.
The favored locale for these lesions is, at least in modern times and
developed countries, pulmonary tissue. Pneumonic infiltration, granulo-
matous reaction, and the caseous degeneration that often ends up as sig-
nificant local destruction (cavities), as well as the general poisoning of the
177
organism by the microbe's metabolic products, all result in the classic

profile of consumption: wasting away of the whole body, continuous
fever, fatigue, cough, and blood-spitting. The evolution of the lesions
varies depending on the immune resistance of the patient. Not unlike
leprous infection, tuberculosis can develop in two ways: one is character-
ized by excessive exudation along with ulceration and rapid dissemination,
the other by fibrosis and calcifications that isolate the tuberculous foci. In
inhabitants of industrialized nations, the infection usually evolves in both
ways at once (fibrocaseous tuberculosis), as though there were a sort of
compromise taking place between the two basic forms.
The localization of tuberculosis in the lungs is favored by the abundant
oxygen in that organ and, at least in modern man, by the fact that Koch's
bacilli are usually transmitted in the air we breathe. Among their localiza-
tions, which are numerous and varied, I mention the intestines and mes-
enteric lymph nodes (typical when the bacilli are ingested orally in massive
quantity, as with infected milk), the lymph nodes of the neck (scrofula),
the genito-urinary organs, the meninges, and, last but not least, the bones
and joints. On the skeleton, Koch's bacillus prefers to attack the diarthro-
ses (tuberculous osteoarthritis of the knee, hip, wrist, ankle, elbow, and
so forth), the spine, and the areas of hemopoetic marrow in long bones.
The tuberculous affection can attack the cranial vault. Its most common
skeletal form is Pott's disease, which is a tuberculous spondylodiscitis that
destroys adjacent vertebral bodies. The caseous destruction can lead to a
paravertebral abscess that extends downward and burrows under the sheath
of the psoas major muscle. The collapse of one or several vertebral bodies
produces a sharp hump (angular kyphosis) with a distinctly different ap-
pearance from the round one due to rickets and most other factors that
deform the spinal column.
Though anatomopathological observation of osseous tuberculosis is rel-
atively simple and secure when one can proceed to a microscopic exami-
nation of fresh tissue, the diagnosis of this disease is much less simple when
only dry, even petrified, bones are available. Though pathognomonic in
principle, the tuberculous lesion equivocates within the material con-
straints imposed by osteoarchaeological study. 2 In cases in which the di-
agnosis of tuberculosis can be based only on iconographic documentation,
the chief indicator, at least for prehistoric and ancient art, is provided by
the presence of an angular hump. However, it must be admitted that
several pathological processes that are not tuberculous (for instance, trau-
mas, congenital malformations, and pyogenic osteomyelitis) sometimes
produce kyphoses that at least externally mimic exactly the consequences
of Pott's disease. From modern clinical experience, vertebral tuberculosis
is the most common cause of angular hump, but it would be circular to
project the frequency of such an etiology on a distant past whose epide-
miological conditions are scarcely known. So Pott's disease is only one
TUBERCULOSIS 179
possibility among 9thers that must be invoked with respect to hunchbacks

in ancient art. 3 As for the iconographic representation of emaciated indi-
viduals, the usual diagnosis of "consumption" must be taken loosely.
Overall decline is not necessarily tuberculous in origin. Pictorial informa-
tion alone cannot justify a diagnosis of pulmonary tuberculosis. There is
no question of invoking certain prehistoric rock paintings, Hindu sculp-
tures, Greek statuettes, or Roman mosaics that represent extremely thin,
relatively young persons as evidence of the antiquity of tuberculosis.
Tuberculosis is not exclusively a human disease. However, in the
publications that I have had occasion to consult, no mention is made of
this disease on the bones of prehistoric animals. Likewise, no lesion sus-
pected of being tuberculous has been reported on human bones of the
Paleolithic period. There is no need to draw hasty conclusions: the former
lack may be explained by the insufficiencies of paleopathological research
into animals, and the latter by the very small number of bones in question.
An osteoarchaeological diagnosis of tuberculosis has been proposed for
several cases from the Neolithic period and the Bronze Age in western
Europe (France, Belgium, Germany, Denmark). 4 The best-known exam-
ple is a Neolithic spine (about 5000 B.c.) excavated at the beginning of the
century from a cemetery near Heidelberg and described by Paul Bartels; at
the time of its publication, the find seemed to prove irrefutably the prehis-
toric existence of osseous tuberculosis in western Europe. 5 The case pre-
sented partial destruction of the bodies of the fourth and fifth thoracic
vertebrae with complete fusion and an angular kyphotic deformity, which
is indeed an anatomopathological feature of Port's disease. Some doubts
were expressed as to the solidity of the diagnosis, since, as H. U. Williams
remarked, a healed fracture of the spine might well have produced the
same paleopathological condition. Since the apophyses of the Heidelberg
thoracic vertebrae are intact, the likelihood that the deformation of the
vertebral bodies was due to an act of violence or an accident instead of
tuberculous infection seems to me slight, although it cannot be entirely
ruled out.
It is not absolutely impossible that all the European prehistoric cases
diagnosed as Port's disease are actually traumatic destructions or nonspe-
cific purulent inflammations, and also that all the presumed cases of tuber-
culous osteoarthritis from the Neolithic period (such as the ankle fron1
Aumede in Lozere or the hip from Grenelle in Paris) actually result from
the action of staphylococci, not Koch's bacilli. Still, such an interpretation
of the osteoarchaeological discoveries accumulated up to now seems to me
untenable. Even if a critique can be made of each case, the sum-total of
prehistoric lesions described as tuberculous justifies a strong presumption
in favor of the diagnosis. 6
An unshakable case of osseous tuberculosis was recognized by G. Elliot
Smith and Sir Armand Ruffer on an Egyptian mummy of the Twenty-first
Dynasty (about 1000 B.c.). The mummified corpse of this young man, a
priest of Ammon, presents both an angular kyphosis brought on by the
destruction of the lower thoracic and upper lumbar vertebrae and a huge
abscess in the area of the right psoas muscle. 7 More than 30 other osteoar-
chaeological specimens from Egypt and Nubia have lesions that perfectly
resemble those of vertebral tuberculosis. Even if some of the cases are
debatable, that does not detract from the worth of those that resist criti-
cism and ,vhose evidence is, as a whole, undeniable. 8 The oldest tubercu-
lous specimens go back to 3000 B.C., the time of the First Dynasty. So it
is very likely that in that part of the Mediterranean, osseous tuberculosis
was already established in the Neolithic period, at least in its latest stage.
Several Egyptian figurines represent hunchbacks, and some of them have
the hump at right angles. So they could be referred to Pott's disease. One
of the statuettes is probably predynastic (about 4000 B.C.) and therefore
constitutes an indication in favor of the antiquity of tuberculosis. 9 The
presence of osseous tuberculosis in ancient Egypt allows us to credit the
existence of pulmonary, intestinal, and glandular tuberculosis. However,
the statistical correlation between these various forms cannot be extrapo-
lated for such a distant past. In other words, knowledge of the frequency
of one of these forms does not support specific conclusions as to the
frequency of the others. Certain pleural adherences that might derive from
a tuberculous attack have been noticed on mummies, but it has never been
possible to find tuberculous lesions on the mummified pulmonary tissue. 10
Such examinations are very difficult, since the embalmers removed the
viscera from the cadavers and deposited them in Canopic vases in which
the tissues of parenchymal organs quickly underwent important changes.
The OldestLiterary Evidenceof Tuberculosis
According to Charles Coury, pulmonary tuberculosis was once a rarity

in Egypt, "a hot, dry land whose climate was even in ancient times consid-
ered healthy for consumptives. " 11 The fact is, Egyptian medical texts of
the high period nowhere mention consumption.· Was it present in Meso-
potamia at the dawn of the historical period? Yes, probably, but the affir-
mation rests more on biological considerations that on concrete evidence.
Literary sources, notably the tablets of Assurbanipal, contain incantatory
formulas against coughs and against a disease that for philological reasons
seems to be consumption. In some Assyrian texts, we can recognize the
description of pneumonia and bronchitis, but even so their tuberculous
character cannot be established. A medical historian can find proof of the
frequency of respiratory diseases in this part of the Old World, but he
cannot specify their etiology. 12
In the Far East, pulmonary tuberculosis is a disease that was known very
TUBERCULOSIS 181
early on. In the laws of Manu (ea. 1200 B.c.) consumption is a defect or
impurity that can be transmitted to one's descendants. Ayurvedic medi-
cine dubs it "the royal disease" (rajayaksrna).The Susrutasan:i,hitacontains
an excellent description of its symptoms and stresses its sociodemographic
importance as well as its resistance to medical treatment. 13
The Nei-chingy the basic treatise of Chinese medicine, contains some
allusions to pulmonary disorders whose tuberculous nature is highly likely.
According to the annalists, the prime minister of Emperor Kao-tsung (ea.
1300 B. c.) was afllicted with a h un eh back "as angular as a fish fin." De-
scriptions of chronic coughing, blood-spitting, cervical adenopathy, ''in-
fantile consumption," "osseous fever," and deformities of the spine
abound in the classical medical literature of the Celestial Empire. In the
seventh century A.D., Tswei Che-ti assembled the various forms of tuber-
culosis into a masterful clinical tableau; apparently the disease was becom-
ing more and more devastating. 14
I+otohistoryof Tuberculosisin America and Greece
The existence of this disease among the peoples of pre-Columbian

America has been cast into doubt by specialists as competent as R. Moodie,
A. Hrdlicka, and, more recently, Dan Morse. The numerous representa-
tions of hunchbacks in American Indian art, especially in pottery, do not
of themselves constitute incontestable proof of the presence of Pott's dis-
ease. As was said earlier, other ailments can mimic the tuberculous hump.
To back up the diagnosis, tuberculous stigmata should be identifiable on
pre-Columbian bones. By 1960, 15 osteoarchaeological cases suggesting tu-
berculous disease on the spines of American Indians had been published.
Morse reviewed them all and came to the conclusion that only 4 of them
(a Peruvian mummy, a child's skeleton from Pueblo Bonito, and 2 spines
of adult males from Hopewell Mound and Chucalissa, Memphis) are typ-
ical enough to justify a strong presumption of spinal tuberculosis. In 1 of
these cases, a pre-Columbian date is not certain, and in all 4, according to
Morse, other diseases could have caused the same paleopathological
picture. 15
Morse's hypercritical approach and his rejection of prehistoric tubercu-
losis in American Indians has been contradicted by osteoarchaeological
research carried out in a systematic and sophisticated fashion on human
remains discovered during the last two decades in sites as diverse as Sonoma
(California), Pueblo Bonito (New Mexico), Nazca (Peru), and Chavez Pass
near Tempe, Arizona. Of the 44 skeletons from Sonoma, dating from the
sixth to the third century B.C., 1 case of osseous tuberculosis has been
recorded. On the more recent American Indian bones from Chavez Pass,
which date only from the twelfth to the fourteenth century A.D., Marc
Kelley has been able to identify 3, if not 4, very probable cases of tubercu-
lous infection. 16 The diagnosis of pre.:..Columbian osseous tuberculosis is
supported by the philological interpretation of an oral tradition according
to which the Maya and the Inca of the historical period prior to the
conquistadors used special terms to designate pulmonary consumption. 17
According to Charles Coury, ''Tuberculous affections were decidedly
more rare among the red men and the other autochthonous Indian tribes
before the massive contamination of the American continent by Europe-
ans that began in the seventeenth century." 18 However, this same author-
ity cautiously notes that in order to ascertain the morbidity and mortality
due to tuberculosis in general and to its more serious internal forms in
particular, one cannot proceed by extrapolating from the frequency of
osseous tuberculosis. The sample of human remains thus far subjected to
study is too small and statistically insignificant, and, also, the correlation
index between Pott' s disease and pulmonary tuberculosis is too uncertain.
"The fact that in a set of skeletons the percentage of those with traces of
tuberculosis is 3.5 percent for Neolithic France, .22 percent for ancient
Egypt between 6,000 B.C. and A.D. 600 (M~ller-Christensen), or 4.5 per-
cent in California from the sixth to the second century B.C. (Roney), does
nothing to inform us on the frequency of tuberculosis in general." 19 The
notion that tuberculosis was relatively rare among native Americans before
the conquest is based on observations suggesting a high mortality and
special virulence in the tuberculous conditions among them from the time
of their contact with whites. Was it Europeans who brought into the New
World special strains of the human tuberculous bacillus, fiercer than those
already there, or did they introduce the very first strain of the bacillus,
assuming that the Indians had known only bovine tuberculosis before the
whites' arrival?
Being unable to answer this question, I return to my main subject and
the shores of the eastern Mediterranean. Though the introduction of tu-
berculosis into Greece certainly goes back to prehistoric times, sure os-
teoarchaeological examples of it are still lacking. J. L. Angel has reported a
single, doubtful case, in which the collapse of the vertebral bodies and
their kyphotic fusion in a young girl of the Early Iron Age (Argos, ea. 900
B.C.) might be explained as tuberculosis. 20 The representation of hunch-
backs is fairly. common in Greek art, but the works in question are rela-
tively late. Literary sources confirm the presence of pulmonary, osseous,
and glandular tuberculosis in the Greek city-states during the classical age
and support the hypothesis of an endemic worsening during the Hellenis-
tic and Roman periods. According to Krause, the demographic distur-
bances that followed the great invasions from the fifth to the sixth century
A.D. produced a regression of tuberculosis; then there was a strong resur-
gence of it in cities that were overpopulated and at a generally low level of
hygiene. 21
TUBERCULOSIS
The Ravages of Consumption in the Classical Age

Received opinion nowadays sides with the bold medical historians who
have detected pulmonary tuberculosis in a few vague references in Homer
and the tragic poets to a sluggish, devastating disease without a single other
specific trait. 22 In this matter, I prefer a more reserved judgment. It re-
mains true that tuberculosis sufferers were an important segment of the
clientele of the Hippocratic physicians, who had found out the essential
traits of the clinical picture of pulmonary consumption. 23 I haven't the
slightest doubt about the presence of tuberculosis in archaic Greece, but
in all candor it must be admitted that the myths and ancient poetic works
say nothing substantive about it. The omission is made good fro1n the
earliest medical literature on. The concept of pulmonary consumption
appears in the oldest parts of the Hippocratic corpus, in the most authentic
epidemiological texts of the Coan physicians as well as in passages that go
back to the Cnidian Sentences.
In ancient Greek the word phthfsis 'consumption, phthisis' has a more
general sense than the purely medical meaning of its calques in modern
languages. When Aristotle speaks of the phthfsis of the moon, he simply
means that it is waning. The same word can be applied to the setting sun,
to any kind of extinction, and to any diminution of an object that will
conclude with its disappearance. Phthfsis in its original sense is nothing
more than a state of diminution or withering. The noun is derived from
the verb phthfniJ 'wither, diminish, be consumed,' whose root is Indo-
European. 24 Once part of the technical vocabulary of Greek physicians,
this family of words kept its original meaning: the expressions dorsal phthi-
sis, nephritic phthisis, phthisis of an eye or a limb, are used to designate
not tuberculosis but atrophy of those organs. Galen and Aretaeus use a
similar expression for the shrinking of the pupil of the eye. But that does
not prevent the word phthfsis and its derivatives from appearing in the
classical medical literature without qualification and in a narrower sense,
that is, as the name of a specific nosological entity.
The first attestation of the word phthfsis is in Herodotus. The historian
is telling how, at the time when Xerxes' army was preparing to conquer
Greece, one of the three generals of the Median cavalry, Pharnuches, had
to be '' left behind, sick, at Sardis. '' This took place in or about the year
481 B.C.: "When the army was leaving Sardis, he met with an unhappy
accident: a dog ran under his horse's feet, and the horse, taken by surprise,
reared and threw its rider. As a result of the fall Pharnuches began to spit
blood; and his sickness finally turned into consumption kai es phthfsin
periilthe he nousos).25 According to Bruno Meinecke, phthfsis is here "clearly
mentioned" in the modern sense of pulmonary consumption. 26 That is
not the opinion of Fridolf Kudlien: the last phrase of the passage cited
seems to him so vague that he proposes to translate it simply by the words
DISEASES IN THE ANCIENT GREEI{ WORLD
"the disease became chronic." Pharnuches was left at Sardis in the same
way as Philoctetes was abandoned at Lemnos in the Trojan cycle. In both
cases, the long-term consequences of an accident made the military leader
incapable of exercising his functions. In Herodotus, says Kudlien, the
substantive phthfsis does not yet have the more restricted meaning ''pul-
monary consumption." It only denotes a general decline, which is the
primary characteristic of every serious chronic disease. 27 I tend toward a
more finely shaded interpretation, one that lies between those of Meinecke
and Kudlien. The former is incorrect in identifying Herodotus's phthfsis
with pulmonary tuberculosis; and the later exaggerates, to my mind, by
denying any nosological specificity to this phthfsis. Pharnuches' hemor-
rhage28 was brought on by a trauma, so it seems unlikely that the disease
that followed it was tuberculous. Unlikely-but not excluded, since the
accident could have activated a latent specific bronchopulmonary process.
In any case our retrospective judgment on the pathological reality of this
posttraumatic illness is of only secondary importance for an appreciation
of Herodotus's medical terminology. Pulmonary hemorrhage with com-
plications, hemopneumothorax, abscess, bronchiectasis or caseation-the
Greek historian, or the physician who served the Persian army and whose
diagnosis he relates, did not err in calling it phthfsis: though it did exist as
a clinical entity and not just as a vague idea of decline, as yet it only partly
comprised tuberculous consumption.
In the Hippocratic texts, the concept of phthfsis becomes more limited
and more precise, and it takes on an anatomoclinical meaning at once
richer in details and narrower in its definition of a nosological field. Dis-
eases)I, a normative Hippocratic treatise of exceptional value as a repository
of extensive medical experience, enumerates the diseases that, once they
have declared themselves, inevitably lead to death: at the head of the list
stands phthfsis. In the same treatise, mention is also made of phth6e) which
takes first place in a list of diseases that are necessarily of long duration. 29
According to K. Deichgraber, phth6e is nothing more than the ''Cnidian
term that replaces phthfsis. '' 30 But both words occur in the same chapter
of a single work (Morb.) I, 3) whose Cnidian affiliations are doubtful; and
moreover, the treatises that are generally considered Cnidian commonly
use the term phthfsis. The split Deichgraber proposes seems to me difficult
to accept. It remains true that there are philological arguments in favor of
the synonymy of the two words, for they are just two grades of the same
root.
In the Deftnitiones rnedicae attributed to Galen but perhaps compiled a
little before him, the following explanations occur: "Phthfsis: ulceration
(helkosis) of the lung or the chest or the throat; it brings on coughing and
light fevers, along with wasting of the body. Phthfsis differs from phth6e in
that phthfsis generally is used to designate any thinning down or consump-
tion of the body, while phth6e is used especially for the consumption and
TUBERCULOSIS 185
thinning down thaJ are the result of an ulcer (eph) helkei). " 31 It is not easy
to grasp the clinical significance of this distinction. Attested as it is in a late
text, it is not necessarily valid for the classical period. The term phth6e is
used only twice in the Hippocratic corpus, as against 42 usages of the noun
phthfsis and 30 of the adjective phthin~des (which applies to persons as well
as diseased organs). I will return later to the second mention of phth6e)
noting for the moment that it relates to the clinical description of a pul-
monary disease with pyogenic internal ulcers. 32 A glance at the passages in
which phth{sis and phthin~des occur without qualification is enough to
reveal the duality in their use: in some instances, the original and broad
sense of a consumptive disease is at the forefront, and in others, which are
more numerous, the narrower and more precise one must be understood.
As a whole, the latter corresponds to the Pseudo-Galenic definition that
speaks of a clinical triad (cough, fever, consumption) and derives it from
pulmonary or at least intrathoracic ulceration. The nosological vocabulary
of the Hippocratic corpus is fluid, and the word phth6e appears as a kind
of abortive attempt to eliminate the technical annoyance of the double
meaning of phthfsis. The attempt failed at least in part because some mas-
ters of the Greek medical thought on principle refused subdivision and
rigidity in the nomenclature of diseases, doubtless out of fear that it would
cause a loss of perspective on the individuality of concrete cases.
By the end of the fifth century B.C., the nosological conceptualization
of pulmonary consumption took the form of clinical profiles that fixed the
symptoms of tuberculosis sufferers for centuries to come. Its most striking
example is the three consumptions of the treatise Internal Affections:
The first consumption ... begins with a dull fever and chills; there is pain in
the chest and back; from time to time, an overpowering coughing fit; lots of
aqueous, salty spittle. Those are the signs of the beginning of the disease, but as it
progresses, the body grows thin, all but the lower limbs, which swell up. Finger-
and toenails become curved; the shoulders become thin and weak; the throat is
full of a kind of down and produces whistling sounds, as from a reed flute. For the
duration of the disease, a terrible thirst, and total body weakness. After a year, the
patient succumbs wretchedly to the devastation. 33
The second consumption: it comes from exhaustion. The manifestations are
about the same as in the previous case, but the sickness has more remissions and it
slackens in the summer. The spittle is thicker; the cough is overpowering especially
in the mornings; there is more pain in the chest-it feels like a stone; the back,
too, is painful, the skin sweaty, and with the least exertion the patient pants and
breathes heavily. In this disease, death overtakes the victim usually after three
years. 34
The third consumption ... First, the person becomes black and somewhat
swollen, with yellowish skin below the eyes; the body's vessels are distended and
yellow, but some take on a bright red color; the most visible are those under the
armpits. The spittle is yellow when there is any, the patient chokes and at times
cannot cough even if he wants to; sometimes, too, the choking and the need to
cough bring on abundant vomiting either of bile or phlegm or often also of food,
when the patient has eaten; immediately afterward, the patient feels lightened, but
the improvement does not last long, and the same suffering begins again. The
voice is more shrill than when well. From time to time, there are chills and fevers
accompanied by sweating ... Usually the disease lasts for nine years; after that
time, though in a wasted state, one can last; but few escape, it is so severe. 35
The first type of consumption is described still more vividly in Diseases)II:
When there is lung ailment, the patient spits up slimy substances that are thick,
greenish, and sweet; there is grinding of teeth, pain takes over the chest and back,
a slight whistling can be heard in the throat; the throat is dry, the hollows beneath
the eyes are red, the voice lowers, the feet swell and the nails bend; the upper
body wastes away and the patient grows thin. The slimy spittle is disgusting to the
patient when it is brought up and is in the mouth. He coughs especially in the
morning and the middle of the night, but also at other times. The disease tends to
attack younger women rather than older ones. Then, if hair falls out and the head
is bald as after a sickness, and, when the patient spits on charcoal, if the slimy
substance gives off a strong smell, the end is near and the cause of death will be
diarrhea. 36
The chapter that follows immediately upon this description is entitled

"another phth6e." One can therefore conclude that the preceding pulmo-
nary ailment was indeed, for the author of this treatise or for one of its
ancient editors, a kind of consumption. Despite some differences of detail,
the clinical identity of the pathological states described in Internal Affec-
tions) ro ("the first consumption"), and Diseases)II, 48, is not in doubt.
As for the second phth6e) essentially it corresponds to the third consump-
tion of Internal Affections) but with a fairly important difference in the
presentation of the symptoms. Here is the semiological part of Diseases)II,
49: "The patient coughs; his spittle is abundant and moist; at times,
without effort, he spits up pus like a hailstone; crushed between the
fingers, this pus is hard and malodorous. The voice is clear, and the patient
feels no pain. He is not attacked by fever, but sometimes by heat; above
all, he is weak ... The disease lasts seven or nine years. In this case, if
the patient is treated from the start, he is cured. " 37 Another clinical por-
trait of the same pathological category is drawn in chapter 50 of this
treatise. It speaks of the "aph thous pulmonary flute," "the raucous
voice," slimy spittle that sometimes contains "hard pieces that resemble a
fungous growth from an ulcer," "growing thin," "reddening cheeks,"
"curving nails," as well as rapid death "amidst the spitting of blood and
pus. ''38
To be sure, in the descriptions I have just been citing, the nosological
distinctions are not those of modern medicine. All the various kinds of
consumption have symptoms that may signify pulmonary tuberculosis in
the current sense of the term, but at the same time, each of them goes
beyond the framework of the modern disease. In all the examples, the
TUBERCULOSIS
symptomatology is based on concrete observation of pulmonary tubercu-

losis, but it also includes clinical experience with cavities and suppurating
ailments that are not tuberculous: abscesses, fistulated empyemas, excavat-
ing cancers, bronchiectases, and so forth. The "first" consumption corre-
sponds perfectly to the serious forms of ulcero-caseous tuberculosis in the
adult, but at the same time it also corresponds to the clinical appearances
of a pulmonary abscess or a bronchopulmonary cancer. The picture in
Diseases.,II, 50, is of the typical tuberculosis sufferer as we know it, espe-
cially from descriptions of the first half of the nineteenth century. Tuber-
culosis is no doubt its major component, but some symptoms, especially
the seriousness of the morning cough and the qualities of the sputum, also
recall bronchial dilatation. 39 Black skin coloration and extreme debility
could also have been seen in cases of tuberculous affection of the adrenal
glands (Addison's disease). Strident respiration is a sign of the stenosic
form of tracheobronchial tuberculosis. In the description given by Diseases.,
II, 50, it is also easy to recognize tuberculous pharyngitis, a common
complication of pulmonary tuberculosis. But just as Hippocratic con-
sumption extends beyond pulmonary tuberculosis in the modern sense of
the word, it is also true that some other classical nosological entities, for
instance, empyema and pleurisy, comprise clinical forms due to infection
of the lungs and pleura by Koch's bacillus. It seems to me that the impor-
tance of pleurisies in the writings of the ancient Greek physicians 40 can
only be explained by the predominance of tuberculosis in their etiology.
The modern physician can admire the subtlety of the Hippocratic symp-
tomatology and appreciate the judiciousness of its prognostic. In the de-
scriptions given above, the high points are the observations of secondary
signs in addition to wasting away, hemoptysis, purulent sputum, and
mitigated fever: the curved nails, which are part of the phenomenon now
known as clubbing; 41 the quality of the sputum and the nature of the
cough; the examination of sputum over hot coals; the smell of the pa-
tient's breath; the hair loss and the diarrhea in the terminal phase of
consumption. The existence of passages parallel to these in the form of
aphorisms elsewhere in the Hippocratic corpus shows how much value
was attached to such signs: "In persons affected with consumption, if the
sputa that they cough up have a fetid smell when poured on burning coals,
and if their hair falls out, then it is fatal. " 42 In a similar text, there is the
added detail that "there is a strong smell of burnt meat. " 43 In fact, this
method makes it possible to detect the presence of pulmonary tissue debris
in the sputum.
When consumptives spit in sea water and the pus falls to the bottom, danger is
imminent. The water should be in a copper vase. 44
Consumptives whose hair falls out die from diarrhea; and all consumptives with
diarrhea die. 45
Those who cough up foaming blood without feeling pain beneath the diaphragm
are getting it from their lungs. 46
Foaming blood in the absence of abdominal pain is precisely the difference

between hemoptysis and hematemesis. This particular aphorism suggests
that the Hippocratic physicians saw both tuberculosis of the respiratory
tract and gastric ulcer in daily practice.
According to the medical texts of the classical era, the basic form of
consumption is marked by pathological events whose origins lie in the
lungs. The disease arises from the fluxion of phlegm that descends from
the head to the lungs and brings on suppuration 47 or, rather, from an
accumulation of phlegm and bile. These humors, existing to excess, stay
in the lungs, gather there, and, becoming tainted, form "tumors" or
masses of pus; the evacuation of this pus by expectoration transforms the
masses into cavities, which either can become totally dessicated (after the
patient is cured) or can continue to suppurate (with a fatal result marked
by the dessication of the body and by diarrhea). 48 The pathological mass
in the lungs, which today we call the tubercle, is called phu111ain the
Hippocratic texts. It is a technical term with a very broad sense, encom-
passing "any unnatural tumor that arrives spontaneously" ;49 it is especially
appropriate for suppurating tumors. So the word phu111acorresponds pri-
1narily to our terms "abscess" and "tubercle," but it also functions as the
name for certain forms of cancer and hydatid cyst. In a context that assur-
edly refers to tuberculosis, the author of the treatise Joints mentions intra-
pulmonary phu111atathat are "hard and raw." Impressed by this medical
knowledge, Charles Daremberg exclaimed that "the ancients assimilated
lung tubercles to true abscesses, with periods of being unconcocted and of
coction, perceiving the overall progress of the disease and the nature of the
expectorate that accompanies the suppurated tubercles, while our word
tubercle refers instead to the origin, form, and first stage of an accidental
pathological process. " 50
I will not go into detail on the pathogenetic explanation of consump-
tion, since my task is to make out the reality of diseases and not to follow
the tortuous byways of medical theorizing. Still, I cannot suppress my
amazement at the anatomopathological realism of the nineteenth chapter
of Diseases)I, whose subject is the formation of tubercles and their cavita-
tion. Doesn't such knowledge of facts that are invisible from the outside
of the body imply recourse to autopsy? 51 An alternative, historical expla-
nation seems superior: recourse to analogy with animals from butchers or
sacrifices. In Youth and Old Age Aristotle said of the death of animals in
general, "Animals die without violent pain, and the deliverance of their
souls is utterly insensate. All the diseases that harden the lung, whether by
tubercles (ton pneumona skleron e' phu111asin)or secretions or an excess of
unhealthy heat, as in accesses of fever, make breathing frequent. " 52
TUBERCULOSIS
An EpidemiologjcalObservationand Two Clinical Cases

Book 2 of Diseasesand the treatise Internal Affections are traditionally
associated with the medical community of Cnidus and are said to have
found inspiration in the nosological descriptions recorded in the Cnidian
Sentences)a work with major contributions from Euryphon, Hippocrates'
senior colleague. However, it is worth noting that the same clinical notion
of consumption occurs in works thought to be the most authentic, that
is, those attributed to Hippocrates himself or at least to the physicians of
Cos during the last quarter of the fifth century B.C. Here is an itinerant
Coan physician's account of diseases seen over the course of a year on
Thasos:
Beginning early in the summer, throughout the summer and in winter many of
those who had been ailing for a long time took to their beds in a state of consump-
tion, while many also who had hitherto been doubtful sufferers at this time showed
undoubted symptoms. Some showed the symptoms now for the first time; these
were the ones whose constitution inclined to be consumptive. Many, in fact most
of them, died; of those who took to their beds I do not know one who survived
even for a short time. Death came more promptly than is usual in consumption,
and yet the other complaints, which will be described presently, though longer
and attended with fever, were easily supported and did not prove fatal. For con-
sumption was the worst of the diseases that occurred, and alone was responsible
for the great mortality. In the majority of cases the symptoms were these. Fevers
with shivering, continuous, acute, not completely intermitting, but of the semi-
terian type; remitting during one day they were exacerbated on the next, becoming
on the whole more acute. Sweats were continual, but not all over the body. Severe
chills in the extremities, which with difficulty recovered their warmth. Bowels
disordered, with bilious, scanty, unmixed, thin, smarting stools, causing the pa-
tient to get up often. Urine either thin, colourless, unconcocted and scanty, or
thick and with a slight deposit, not settling favourably, but with a crude and
unfavourable deposit. The patients frequently coughed up small, concocted sputa,
brought up little by little with difficulty. Those exhibiting the symptoms in their
most violent form showed no concoction at all, but continued spitting crude
sputa. In the majority of these cases the throat was throughout painful from the
beginning, being red and inflamed. Fluxes slight, thin, pungent. Patients quickly
wasted away and grew worse, being throughout averse to all food and experiencing
no thirst. Delirium in many cases as death approached. Such were the symptoms
of the consumption. 53
This account is the nosological core of the first katastasisof Thasos. 54 The
observations in question date from around 410 B.C. Their author is not
conveying bookish knowledge but reporting what he actually saw. To be
sure, he saw the clinical events by way of a medical theory. Charles
Daremberg provides an especially insightful commentary on this passage:
"This picture of consumption is strikingly realistic. Even so, it is useful to
note that the disease is here described more as a general ailment than a
chest disorder, and that the description of general symptoms exceeds that
of local ones in terms of the space and significance accorded them by

Hippocrates. " 55 For the historian of diseases, the interest of this first ka-
tastasislies primarily in the fact that it studies consumption as a collective
phenomenon. There is no doubt that it attests to the seasonal exacerbation
of a tuberculous endemic on the island. The limitation of such an endemic
to a single island seems unlikely, and this Hippocratic testimony should
be thought of as a valuable reminder of the prevalence of tuberculous
diseases at the time of the Peloponnesian War. 56
The Hippocratic physicians knew that consumption could be a social
phenomenon, a disease affecting several people at once, but they still did
not believe in infection. 57 To account for the epidemic outbreak of con-
sumption, Hippocrates appealed to the noxious effects of certain ecological
factors and especially to the determining role of climate. However, since
all persons inhabiting the same places were subject to the same climatic
conditions, it was necessary to explain in theoretical terms the empirical
fact that only a portion of those individuals exposed to the unhealthy
factors actually fell ill. Hippocrates' explanation is simple, skillful, and
based on precise clinical observation: the disease results from the coinci-
dence of external factors (meteorological as well as earth-bound condi-
tions-diet, exercise, and so forth) with individual predispositions, with a
personal ''constitution'' that reacts pathologically to changing local
conditions.
For most Greek physicians of the classical period, the constitution of an
organism was nothing more than a global expression of its individual mix-
ture of bodily humors. Since they could not acquire direct experience of
this humoral mixture, Hippocrates and his disciples studied the relations
between the habitus of patients and their diseases in order to deduce the
features of temperament and, in the last analysis, the constitutional weak-
nesses of each person. They knew that predisposition to tuberculosis de-
pended on age, sex, mental state, heredity, and prior way of life. According
to the Hippocratic texts, ''Consumption commonly occurs between age
eighteen and thirty-five. " 58 Women in particular are threatened by it,
especially during pregnancy. 59 The prognosis is very poor when it breaks
out after amenorrhea. 60 Phthisis frequently supervenes after childbirth. 61
The author of the treatise the SacredDiseaseasserts in explicit terms the
decisive role of heredity not just in the case of epilepsy but also in our
disease: "One consumptive is born from another" !62 In I+orrheticon)II,
consumption is mentioned, along with dropsy, gout, and epilepsy, as a
disease whose cure is difficult when there is a genetic predisposition. 63 The
predisposition itself can be recognized from the overall appearance of the
body-at least that was the teaching of the Coan physicians. In an account
of his yearlong observations of the diseases prevailing in Thasos at some
point in the last quarter of the fourth century B.C., Hippocrates or a
physician close to him noted that "consumption affiicted men with smooth
TUBERCULOSIS 191
skin, whitish, wit,h freckles, ruddy, blue-eyed, with soft, puffy flesh and
shoulder blades shaped like wings. " 64 A late Hippocratic text adds to this
profile a flattened thoracic cage. 65 The notion of habitus phthisicus caught
on, and Hippocrates' description, especially his mention of the winged
shoulder blade, was repeated by Celsus and Aretaeus of Cappadocia. It
continued to influence the diagnosis and prognosis of pulmonary tuber-
culosis up to our time. The modern physician can confirm, with admira-
tion, the acuity of the ancient Greek practitioners' clinical eye, though he
repudiates their theoretical conclusions, which go beyond observed
correlations.
For physicians at the end of the nineteenth century, at the apogee of
medical bacteriology, "this habitus is not the cause, it is the sign of the
disease. " 66 Around the middle of the twentieth century, some famous
experts returned to the classical opinion that the habitus was, if not com-
pletely, at least in some of its essential components, the hereditary consti-
tutional condition of pulmonary consumption. The modern notion of
"germ" does not supplant the ancient concept of "terrain"; it only opens
a new dialogue between the acquired and the inborn.
I leave aside these problems in medical theory in order to look more
closely at the concrete nosographic experience of the Greek practitioners,
at the bedside reality of consumption. Here is a typical description that
gives, in condensed form, information about a sequence of morbid events
whose dramatic aspects do not accord with the cold style of the narrative: 67
"The wife of Simos, shaken during childbirth, had a pain in the chest and
the side; coughing, bouts of fever, slightly purulent expectorates; 68 con-
sumption, then things returned to normal; 69 then, for six months, bouts
of fever and constant diarrhea; at the end, the fever stopped; then the
bowels tightened up; seven days later she died." In this case, consumption
is enunciated as a symptom among others and not as a generic term for
the whole clinical picture. The initial stages are sketched rapidly. The
author wishes to make plain the pathogenic role of childbirth or, to state
his opinion more precisely, that of succussion during it. The reference is
to a therapeutic procedure whose use is actually recommended in some
gynecological treatises in the same collection. 70 Although this procedure,
which is undisguisedly condemned in the text cited above, can have no
justification in the eyes of a modern physician, it still cannot be accused of
having caused the consumption that ended the life of the wife of Simos.
On the other hand, pregnancy and the stress of childbirth surely aggra-
vated a heretofore latent tuberculous condition in her.
I now cite another case history that is reported in greater detail by the
same Hippocratic writer from the fourth century B.C. : 71
In the wife of Polycrates, around the dog days of summer, fever; difficulty breath-
ing, less so in the morning, worse from midday on, when it became a little more
rapid; coughing and immediately sputum that was purulent from the start; within,
along the throat and windpipe, husky wheezing; good coloring on the face; red
cheeks, not dark red but, on the contrary, fairly bright. As time went on, voice
becoming hoarse and body wasted, scabs on the loins, and bowel movement on
the watery side. The seventieth day: the fever grows very cold externally; on the
temples, no throbbing; 72 but the breathing becomes more and more rapid. After
this respite, the breathlessness is so heightened that the patient remains seated
until the moment of death. In the windpipe, there was a lot of noise; also, terrible
sweats; looks full of understanding up to the final moment. Once, her fever
lightened for more than five days. After the first few days the patient did not cease
spitting up purulent substances.
This clinical picture is grippingly real, and except for its silence about blood
in the sputum, corresponds completely to the unfolding of pulmonary
consumption with a specifically concomitant laryngitis. Nevertheless, one
should realize that this retrospective diagnosis is not altogether certain:
though the sequence of symptoms corresponds exactly to that which
laryngeal and pulmonary tuberculosis can provide, it is also compatible
with several other serious diseases of the respiratory organs. One can imag-
ine a diagnosis of cancer of the larynx, trachea, or bronchi, which would
produce respiratory wheezing as well as dyspnea turning into orthopnea
along with cancerous cachexia. Still, the striking redness of the cheeks tells
against malignant consumption and in favor of tuberculous infection.
Notes on Consumption in the Hellenistic and Roman Eras
The Roman encyclopedist Celsus (first century A.D.) tells us that the
Greek physicians distinguish three kinds of consumption (tabes), the first
two of which, atrophia and kachexia) were thought to be the result of
nutritional insufficiencies or of the body's inability to profit from nutrients
even when abundant. "The third and most dangerous kind is the one the
Greeks call phthfsis)·it usually begins by attacking the head, then it attacks
the lung, where it produces an ulcer that is accompanied by a small, dull
fever that comes and goes. The patient coughs a lot, spits up pus and
sometimes blood. " 73 It would be hard to find a more succinct definition
of pulmonary tuberculosis. But what Celsus describes simply and compen-
diously, an Alexandrian physician of the same era, Aretaeus of Cappadocia,
can relate in minute and judicious detail: his chapter on consumption is
the high point of medical symptomatology in antiquity. 74 Even a physician
of today could learn by reflecting upon his text. Yet I cannot stop to
analyze it, since the chronological framework of my study forces me just
to mention it along with the passages on consumption scattered in the
wordy oeuvreof Galen as well as the finicky account in Caelius Aurelianus. 75
Knowledge of consumption advanced substantially from Hippocrates to
the beginning of the Christian era. Is this due to an increase in the number
TUBERCULOSIS 193
of consumptives during the Alexandrian period? We do not know, nor is

such an inference inevitable. On the other hand, it is certain that at no
time during the classical, Hellenistic, and Roman periods was consump-
tion a rare disease. All post-Hippocratic medical writers consider it at once
a serious but not necessarily mortal disease that is banal, ubiquitous, and
well known. Aretaeus says that even an idi1Jtes) an average person without
professional medical knowledge, can recognize the disease effectively when
he sees a pale and worn-out invalid who coughs. 76 This statement uninten-
tionally suggests the commonness of a disease that can supposedly be
diagnosed by everyone and anyone one may come across in daily life. But
relatively few famous persons of antiquity are said to die of consumption.
That is not a proof of the rarity of the disease, since it tends to favor the
young, who have not lived long enough to acquire notoriety. Actually,
pulmonary tuberculosis appeared rarer among the old than it really was.
Though easy to recognize in young people, it was mistaken for senile
cachexia or peripneumonfa in the old. Perhaps one can understand the
relative rarity of consumption in the pathological histories of the famous
as a sign of the disease's social selectivity, since it attacked primarily the
poor. Still, it had its victims: the Greek general Aratus of Sicyon (ea. 271-
213B.c.) and the Macedonian king Antigonos II Doson (ea. 263-220 B.c.).
The former died as the result of hemoptysis and a wasting away that he
himself thought were signs of poisoning; 77 the latter likewise expired after
spitting blood. 78 The dietitian Herodicus of Selymbria and the comic au-
thor Theopompus of Athens (fifth century B.c.) were consumptives-the
second claiming that he was cured by the divine intervention of Asclepius. 79
In the Athenian sanctuary of the god of medicine there was an ex voto
statue of Parian marble of a severely en1aciated Theopompus reclining on
his bed. 80 Votive stelae at Epidaurus relate the miraculous story of Ther-
sandrus of Halieis, who, down with consumption, was healed by the
sacred snake of Asclepius (second half of the fourth century B.c.). 81
In such sanctuaries, the cure for consumption was a matter at once of
diet, climate, and divine intervention. For this particular disease, such an
approach was certainly superior to strictly medical treatment. Celsus only
repeats an old adage when he advises consumptives to ''take long cruises,
change your climate, and move to a thicker air than the one you leave
behind." The dry climate of Egypt was especially prized: consumptives
get better, says Celsus, "passing from Italy to Alexandria. " 82 Following
this recommendation, Pliny the Younger sent his freedman Zosimus to
Egypt to treat his consumption. When he returned from Africa to Rome,
Zosimus had a relapse and Pliny dispatched him for a stay in subalpine
Forum Iulii (Cividale). 83 No doubt, cli1natic therapy along with rest and
good food was the most effective means at the time in the struggle against
tuberculosis on an individual level. From a public health standpoint, such
trips for consumptives had the disadvantage of spreading the contagion

outside of Rome, a city that had become one of its havens since the
Hellenistic period or earlier.
It is worth noticing that Pliny the Younger and his circle at Rome
believed in the contagiousness of the dise_ase.The fact emerges clearly from
the moving eulogy of Fannia, a young, dedicated noblewoman who was
caring for the Vestal Virgin Junia and caught a disease with the typical
symptoms of pulmonary tuberculosis: "continuous bouts of fever, wors-
ening cough, emaciation, great weakness. " 84 For Fannia's fellow citizens,
the diagnosis was beyond doubt; for a modern physician reading Pliny's
epistles, it is very likely.
Extrapulmonary Tuberculoses
Medical texts confirm the presence in the classical Greek world of certain
extrapulmonary forms of tuberculosis. They confirm rather than disclose
it, since the documentation of pulmonary consumption suffices in itself to
postulate the existence of other localizations for tuberculous infection.
Although the documents in question are explicit enough to leave no doubt
about the retrospective diagnosis, they unfortunately provide no informa-
tion on the relative frequency of tuberculous attack on various organs.
That frequency may well have undergone historical fluctuations. Even
though we can extrapolate from the current situation about the physio-
pathological progress of tuberculous inflammation and the clinical aspects
of its chief localizations, it is not possible to do the same with regard to
their statistical distribution.
The unity of tuberculosis was not hinted at until the eighteenth century,
nor was it demonstrated until the full flowering of the new medicine of
the nineteenth century. 85 The physicians of antiquity could not conceive
of this unity, since it is based on the anatomopathological traits of the
specific microbial lesion. In ancient times, tuberculosis was only known as
a host of subdivided, autonomous ailments. In any case, that was a legiti-
mate analysis of the nosological reality as long as a disease was defined by
its symptoms. For the physicians of yore there was in principle rio reason,
apart from their possible coexistence, to associate pulmonary consump-
tion, scrofula, white swelling (of the joints), lupus, and certain inflamma-
tions of the genito-urinary organs.
It is therefore all the more remarkable that the Hippocratic physicians
noted the coincidence and even established a causal link between the
pulmonary form of the disease and acquired angular gibbosity. According
to a Hippocratic aphorism, "Those who become hunchbacks before pu-
berty as a result of dyspnea (dsthma)-are lost. " 86 A chapter in the treatise
Joints devoted to the curvature of the spine from internal causes provides
more detail:
TUBERCULOSIS 195
When the spinal vertebrae are drawn into a hump by diseases, most cases are
incurable, especially --when the hump is formed above the attachment of the dia-
phragm ... When humpback occurs in children before the body has completed
its growth, the legs and arms attain full size, but the body will not grow corre-
spondingly at the spine; and those limbs are spindly. And where the hump is
above the diaphragm, the ribs do not enlarge in breadth, but forwards, and the
chest becomes pointed instead of broad; the patients also get short of breath and
hoarse, for the cavities which receive and send out the breath have smaller capacity
... They have also, as a rule, hard and unripened tubercles (phumata) in the lungs;
for the first appearance (pr6phasis)of the curvature and contraction is in most cases
due to such gatherings, in which the neighboring ligaments (t6noi)take part. Cases
where the curvature is below the diaphragm are sometimes complicated with affec-
tions of the kidneys and parts about the bladder, and besides there are purulent
abscessions in the lumbar region and about the groin, chronic and hard to cure;
and neither of these causes resolution of the curvatures. 87
There can be no doubt about this author's familiarity with Port's disease. 88
His text was to be retained and paraphrased by Celsus and Galen; the
disease in question seems to have been a common one. This allows us to
consider reasonable the diagnosis of tuberculosis of the spine in a certain
number of cases that make up the rich iconography of hunchbacks in the
Hellenistic and Roman eras. 89
Osteoarticular tuberculosis can attack all bones and joints. In modern
times it has a clear predilection for the spine, but tuberculous caries also
can reside relatively often in the articular regions of the lower limbs. In
ancient descriptions of coxalgia and knee arthritis, tuberculosis must surely
play a part, but it is hardly possible to make a differential diagnosis. Visceral
forms of tuberculosis readily accompany skeletal affections. Tuberculous
peritonitis certainly existed in ancient Greece, but we are once more faced
with the impossibility of distinguishing, just on the basis of a clinical
description, tuberculosis from other inflammatory diseases of the gut or
even diseases of an altogether different type.
The dropsy of antiquity, particularly ascites (effusion of liquids into the
peritoneal cavity), can originate in tuberculosis of the serous men1branes,
but that is only one among n1any pathogenetic possibilities. As for the
"eloquently laconic" Hippocratic aphorism to the effect that "the cough
supervening in dropsy is bad," Charles Coury thought it sensible in the
context of a tuberculous pathology. 90 But in its clinical context, this cough
may well be the symptom of pulmonary edema arising from a cardiac
disorder, which can also produce ascites as a result of circulatory
insufficiency.
I can cite a concrete instance in which the diagnosis of visceral and
osteoarticular tuberculosis is perfectly suitable but still not obligatory:
"Bion, after a prolonged dropsical state, had no appetite for several days
and suffered from strangury; a deposit formed on his left knee; suppura-
tion; death.' ' 91 Was this patient a victim of tuberculosis? Or did he suc-
cumb to the complications of a gonococcal infection? Or, again, did he
have Bright's disease and, independently of it, acute arthritis of the left
knee? We do not know Bion's age. Another example: osteoarticular tu-
berculosis often strikes children and kills them by miliary spread. There is
a strange funerary inscription in Greek that deserves mention in this con-
text, even though it belongs to a relatively late period (third century A.D.).
The medical-historical interest of this epitaph is considerable, since it re-
lates a true clinical history, namely, the sufferings of one Lucius Minicius
Anthimianus, who died at age 4 1/2 after a painful disease of the testicles,
the decay of his metatarsal bones, and a swelling of his viscera that was
accompanied by a wasting of his body. The retrospective diagnosis is un-
certain, since it cannot be ruled out that the three diseases (orchitis, osse-
ous caries, and an abdominal disorder) were autonomous. But if we
interpret all the pathological manifestations of this brief life as phases of a
single disease, by far the most likely diagnosis is tuberculosis. Other possi-
bilities are leukemia and lymphosarcoma. 92
The Hippocratic corpus contains a description of scrofula (khoirades)
that, though somewhat confused, agrees with the modern profile of tu-
berculous cervical adenitis. 93 The kinship between scrofula and pulmonary
tuberculosis was not even suspected, despite the fact that the Hippocratic
writers insisted on explaining the two diseases by the same pathogenetic
process, namely, local corruption of the plhegm flowing from the head.
In the case of consumption, the fluxion was by way of the lungs, and in
scrofula, it passed to the tonsils, where it brought on an inflammation of
the throat and of one or both ears and, finally swelling of the glands in the
neck, the armpits, or the groin. 94 The Greek physicians of the classical
period knew of the painless but insidious nature of scrofula, its sluggish
progress, and the difficulties of treating it. From the evidence in a Hippo-
cratic text of the fourth century B.C., it was very common in children. 95
Though it was rare from age 42 to age 63, 96 it became a common disease
once again among the old. 97
It is perfectly reasonable to suggest the tuberculous nature of certain
nephrites, cystites, endometrioses, meningites, otites, kerato-conjunctiv-
ites, iridocyclites, and serious dermatoses that are mentioned here and
there in the Hippocratic corpus and the works of Aretaeus, Celsus, Rufus,
Galen, and other physicians. However, the diagnosis in these cases cannot
with certainty pass the anatomoclinical level to reach that of a specific
etiology. In the genito-urinary region, the ravages of tuberculosis were
important, but they evade historical analysis: the symptoms of genito-
urinary tuberculosis are confused with those of nonspecific inflammations
of those organs.
The texts we have been examining prove that from the first medico-
historical documentation, there existed in the Mediterranean, alongside
the pulmonary form of tuberculosis, osteoarticular, visceral, and glandular
diseases that can be ascribed to Koch's bacillus or its ancestor. But what
TUBERCULOSIS 197
was the frequency , of these diseases? Was there a preponderance of extra-

pulmonary as against pulmonary forms of tuberculosis? Was the pulmo-
nary form more often serious or fibrinous? All questions of this type must
unfortunately remain unanswered. Our sources are silent about the kind
of quantitative information that is so important for the reconstruction of
the natural history of mycobacterioses.
Chapter Eight
LEPROSY AND TUBERCULOSIS

Their BiologicalRelationship
A study of the protohistory of leprosy and tuberculosis would be incom-

plete without a comparative examination of their pathogenic microbes, of
the biological relationship between them and their hosts, and of the im-
munological phenomena that arise when their infections cross. 1
The germs of leprosy and of human and bovine tuberculosis are so
closely related to each other that there can be no doubt as to their deriva-
tion from a common bacillus. They are all mycobacteria, species belonging
to the genus Mycobacterium)which was defined in 1896 by K. B. Lehmann
and R. 0. Neumann. Mycobacteria are clearly distinguishable from other
microbes by their slender, sticklike form and their immobility, as well as
by a lipoid envelope that is dyed red by the Ziehl-Neelsen reagent and
cannot be discolored by nitric acid: they are acid-fast bacilli. Though it is
relatively simple to recognize this genus, determining the differences among
the species that compose it can be difficult. Aside from the three "classic"
pathogenic species, namely Mycobacterium leprae)M. tuberculosis)and M.
bovis)we know today of a large number of so-called ·"atypical" species. To
confirm a specific diagnosis, certain bacteriological culture characteristics
must be taken into account (such as optimal growth temperature, growth
speed, pigmentation), as well as biochemical and enzymatic tests, immune
properties, pathogenic capability in various animals, and sensitivity to cer-
tain antibacillary substances.
M ycobacteria are spread throughout nature, either as free-living organ-
isms (for instance, M. ten-aeand M. aq_uae),or as commensals on various
animals and man without the least pathological consequence (for instance,
M. phlei and M. smegmatis), or as parasites producing diseases whose seri-
ousness varies from slight ulcerations to the fatal destruction of tissue and
LEPROSY AND TUBERCULOSIS 199
organs. Some spectes produce diseases in animals whose local lesions re-
semble human tuberculosis histologically: M. marinum (pathogenic agent
in saltwater fish), M. piscium (agent in freshwater fish), M. ranae (agent of
amphibian tuberculosis, to which reptiles are also susceptible), M. tham-
nopoecilus(agent of reptilian tuberculosis), M. chewnei (agent of pulmonary
tuberculosis in turtles), M. microti (agent of glandular tuberculosis in cer-
tain rodents), M. ainum (agent of aviary tuberculosis, especially affecting
chickens but also pigs and humans), and so on. For cattle, in addition to
M. bovis) I should mention two pathogenic species: M. fortuitum) which
produces purulent adenitis, and M. paratubercuwsis)which is the agent of
diarrheic hypertrophying enteritis. In contrast to this broad spectrum of
animal tuberculoses, there is only one known species of Mycobacterium that
specifically produces leprosy in animals: M. lepraemurium) agent of murine
leprosy. It is a parasite chiefly in rats, though it can occur in mice and
hamsters.
Aside from the three species that are strongly pathogenic in humans,
several "atypical" mycobacteria (especially M. kansasii) M. av-ium) .lvi.
intracellulare) M. gordonae) M. scrofulaceum) M. balnei) M. ulcerans) can
produce pulmonary, glandular, cutaneous, and more rarely osseous, gen-
ito-urinary, and septicemic diseases in humans. These "paratuberculous"
affections have been attracting increasing interest among physicians since
the middle of the century. Their frequency-at least in the pulmonary
form-seems to be mounting to a disquieting degree in countries with
highly developed health care systems. The rise in "atypical" mycobacter-
ioses (in some areas of the United States they account for up to 30 percent
of the clinical cases of pulmonary tuberculosis) may be linked to a disequi-
librium in the pathocoenosis of industrial societies that is caused by recent
successes in the chemotherapy of "classic" tuberculosis and of other dis-
eases with very virulent microbes. 2
The worldwide distribution of mycobacteria and the variety in their
habitats, ranging from free aquatic and terrestrial species to commensals
and parasites of all classes of vertebrates, support Aidan Cockburn's opin-
ion that the appearance of the first representatives of the Mycobacterium
genus must go back several hundred million years. He explains the evolu-
tionary birth of most mycobacterial species as a radiation parallel to the
phylogenesis of their hosts. There was a "vertical" transmission of parasitic
species, starting from a common ancestral form that infected salt- and
freshwater fish on one side and reptiles, birds, and mammals on the other.
This diachronic infection sequence produced the specialization or rather
the ramifying specification of pathogenic mycobacteria that in some ways
parallels the appearance of certain great animal phyla. That would account
for the appearance of, among others, Mycobacterium marinum) a patho-
genic bacterium for fish; M. avium) which infects birds; and the mycobac-
terium responsible for Johne's disease in cattle. The parasitism of most
mycobacteria is not strictly elective: for example, M. marinum does not

seem different from M. balnei) which produces cutaneous lesions in hu-
mans; M. avium plays a certain role in the etiology of tuberculous ailments
in pigs as well as humans; and so forth. However, the human species has
the privilege of harboring a highly specialized mycobacterium, M. leprae.
According to Cockburn, leprosy is vertically derived from our ancestors
and is therefore a very old disease in terms of human history, as old as
humanity itself. 3
But where does M. tuberculosis come from? Cockburn devised a fine
hypothesis in this regard: that human tuberculosis was acquired from bo-
vine tuberculosis by interspecific, "horizontal" transmission of the germ
during a relatively recent era, the Early Neolithic Age (about 7000 B.c.).
The appearance and spread of Koch's bacillus resulted from the domesti-
cation of cattle-apparently they did not suffer harm from it in the wild-
and from the development of agriculture, a sedentary lifestyle, and the
beginnings of urbanization. A mycobacterium weakly pathogenic to mam-
mals was the ancestor of M. tuberculosis. It profited from changes in the
ecological conditions of its hosts to increase its virulence and adapt itself
via a bovine form to human populations living in more and more crowded
towns. 4 According to Ronald Hare, "It would seem probable that pul-
monary tuberculosis, caused by the human strains of M. tuberculosis) did
not mal<.eits appearance anywhere until about the second millennium B.C.
If this be so, it is possible that the human strain is a mutant of the bovine
which had previously become established in the dairy cattle that were
domesticated during the Neolithic period. " 5 He also stresses that the germ
of bovine tuberculosis has never been isolated from wild animals that were
not in contact with their domesticated counterparts. 6 The bovine strain is
infectious for humans, in whom it is transmitted mainly by way of milk
and tends to produce abdominal, glandular, and osseous tuberculosis. For
the Mediterranean societies, historical proofs of the existence of tubercu-
losis in domesticated cattle go back to classical antiquity.7
These hypotheses of Cockburn and Hare open new vistas on the study
of the "natural history" of mycobacteria. I hope to contribute to it by
modifying and complementing their ideas, especially as regards the genetic
relationship between M. leprae and M. tuberculosis. But before formulating
my own opinion on this subject, I should mention, however briefly, some
immunological and epidemiological facts.
The Epidemiologyof Leprosy

From an epidemiological standpoint, there is continuity between hu-
man tuberculosis and animal tuberculoses, but there is almost none be-
tween leprosy and animal mycobacterioses. Mammals living in proximity
to men, particularly certain domestic animals and captive apes, easily fall
•
victim to human strains of Koch 's bacillus. The tuberculous diseases they
suffer from hardly differ from those attacking humans. The situation is
different in the case of the leprosy bacillus. Handbooks tell us that M.
lepraeis not pathogenic for any animal species, that it cannot be cultured,
and that it cannot survive normally in an environment other than the
human body. Research done in the last few years has shown that such
categorical statements are not totally justified. For one thing, it has been
discovered that Hansen 's bacillus can be cultured on the footpads of mice,
a finding that can have great practical value but sheds no new light on our
theoretical problem, since the disease does not become generalized in an
animal infected in this way. I mentioned a specific form of murine leprosy.
More interestingly, recent observations suggest that some other animals
may become leprous. Even though descriptions of leprosy in cats or frogs
are unconvincing-like the strange observation of a "leprous" buffalo pub-
lished in 1936, and unheard of since-apparently the existence of a leprous
disease in armadillos is to be taken seriously. Research on the subject is
ongoing, and it is not easy to get a clear picture at the moment. I merely
point out a recently published finding that is adequately documented and
upsets several accepted notions: two American researchers observed in 1977
that one of their laboratory animals, a chimpanzee, was suffering from a
cutaneous disease similar to leprosy. They isolated a bacillus in the creature
that possessed all the traits of M. lepraeapart from some minimal enzy-
matic differences. 8 This chimpanzee had been captured in Africa, in Sierra
Leone, which is a region with endemic leprosy. Probably the infection is
transspecific, having spread from human to ape, and not an independent,
simian type of leprosy.
None of this alters the fact that leprosy is above all a human disease, our
prerogative, and that it is, moreover, strictly limited to human populations
with a certain sociocultural profile. Its current geographic distribution
testifies loud and clear to the role of poverty in the maintenance of its
endemic zones. Leprosy, like tuberculosis, is a by-product of indigence.
However, there is a great difference that is both evident and unexplained
in the way poverty operates in regard to each disease. The causal chain in
the case of tuberculosis is fairly dear-easier contamination in close quarters
without aeration and sunlight, diminished resistance to infection as a result
of malnutrition and exhausting work-but in the case of leprosy, the way
poverty fosters the disease is largely a mystery to epidemiology. To be sure,
population density, migrations, and the absence of personal hygiene all
favor endemic leprosy, but these factors are insufficient to explain com-
pletely the vagaries of its geographic and sociocultural distribution. As for
climatic conditions, they seem to play a secondary role, in contrast to what
happens with treponematoses, malaria, and some tropical parasitic dis-
eases. Leprosy as well as tuberculosis has been rampant in all climates,
from the Far East and black Africa to Scandinavia. Moreover, sociocultural
and geographic factors cannot account for historical fluctuations in the

contagiousness of leprosy. They are probably linked to changes in the
bacillus itself and perhaps even more to complex immunological phenom-
ena, such as a biological balancing process between the populations of this
parasitic microbe and those of its sole host.
Nowadays leprosy is, relatively speaking, just slightly contagious. Only
a small number of individuals exposed to it become leprous, and the
majority of persons who contract the disease undergo no malignant clinical
evolution. That is true even in the absence of any protection or treatment
(we do have at our disposal fairly effective chemotherapy for leprosy). A
single figure is enough to convince us of the current low contagiousness of
the disease, even though its chronic nature should favor, as in the epide-
miology of tuberculosis, the wide dissemination of its bacilli: cases of
conjugal leprosy are less than 3 percent of the total. The difference, then,
between leprosy and tuberculosis is striking, for tuberculosis is clearly a
familial ailment that is often transmitted by cohabitation.
The situation of lepers in Greece in the nineteenth century is a good
illustration of how socially innocuous this disease can be, unless certain
and, to repeat, still unknown conditions are met that facilitate its spread. 9
During the whole century, the number of lepers in Greece remained more
or less constant-a few hundred cases, about 150 according to official statis-
tics.10 They were grouped into small colonies that were not strictly isolated
from the rest of the population. Moreover, numerous lepers were married
to healthy persons who themselves did not suffer from the disease. Clon
Stephanos relates that in the village of Keneri, where 150 families lived in
great poverty beside an open leper house, not a single healthy villager was
known to become infected with the disease. 11 It is almost certain that the
clinical form leprosy takes depends on the "terrain," that is, the host's
resistance, not the intrinsic properties of the germ that is responsible for
infection in a given case. Tuberculoid leprosy is the high resistance form
of reaction to Hansen's bacillus, while lepromatous leprosy is the low
resistance form. Resistance itself is partly the result of hereditary, hor-
monal, nutritional, and other factors whose nature and mechanism are still
mostly obscure, 12 and partly the result of previous contacts with Hansen's
bacillus or other mycobacteria that set in motion immunological defense
processes.
The presence in an individual of specific antibodies can be detected by
Mitsuda reaction (a reaction to the intradermic injection of lepromin, a
sterile extract from tissue rich in Hansen's bacillus). This reaction is nega-
tive in young children, the majority of adults not exposed to leprous
infection, and persons who have contracted lepromatous leprosy. Mitsuda
reaction is positive (that is, it proves the presence of a specific defense
against bacillary antigen) in persons with tuberculoid leprosy, healthy per-
sons who live around lepers, and a certain number of adults who have
never had contact ,with lepers. The positive result of this test in the last
group, at first very surprising, can be explained as a consequence of im-
munological affinities between various species of mycobacteria: it has been
shown that the first stages of infection with tuberculosis as well as BCG
(bacille Calmette-Guerin) vaccination can produce a positive Mitsuda re-
action . 13 This makes plausible the hypothesis that infection by tuberculous
mycobacteria provides relative immunity against leprosy. 14
Clearly, such a hypothesis has important consequences both for public
health professionals and for historians of diseases. For the latter, the hy-
pothesis provides an elegant and unforeseen solution to the problem of
the disappearance of the medieval leprosy endemic in western Europe. 15
From the standpoint of an antagonism between leprosy and tuberculosis
based on immunological competition, it seems possible or even probable
that the retreat of leprosy in the West was linked to a rise in tuberculosis,
which for its part coincided with the social, economic, and demographic
changes of the fourteenth century. 16 In the domain of public health, BCG
vaccination has been undertaken in leprous endemic zones to determine
whether it provides increased resistance to Hansen's bacillus, either by
inciting the production of paraspecific antibodies or by arousing some
nonspecific protective factor. The results now available are not statistically
significant. Though they do not confirm the initial hypothesis, they do
not refute it either. 17
Whatever the case may be, the relations between leprosy and tubercu-
losis are very complex and cannot be reduced to a simple case of antago-
nism. To begin with, there is no real cross-allergy. Koch's bacillus can
make Mitsuda reaction positive, but not always, and, more importantly,
the inverse is not true: infection with Hansen's bacillus does not produce
a positive Mantoux reaction (intradermic tuberculin test). So if, theoreti-
cally, tuberculosis may compete with leprosy, leprosy can have no effect
on tuberculosis. In criticizing Chaussinand's ideas about the antagonism
of leprosy and tuberculosis, several recent authorities have noted that
(r) the number of observed cases of tuberculosis-sufferers with secondary
leprosy is fairly high, (2) systematic radiological examinations have uncov-
ered the presence of tuberculosis in Africa hard by havens of leprosy, and
(3) the disappearance of tuberculosis in Scandinavia has not resulted in the
recrudescence of leprosy. 18 These argun1ents effectively destroy the notion
of a simple antagonism between the two diseases. But they do not, in my
opinion, destroy the historical explanation for the disappearance of leprosy
in medieval Europe, according to which it was the result of competition
between two related mycobacteria, with the issue depending on numerous
ecological factors and the dynamics of the European pathocoenosis as a
whole.
In any case, the third argument of the three given above is of no weight
at all, since the recrudescence of leprosy in Scandinavia is impossible be-
cause of a total change in sanitary conditions. The suppression of tuber-

culosis is only one component of the complex struggle against infectious
disease in general. The second argument is banal and irrelevant to our
problem. The coexistence of leprosy and tuberculosis in a given region is
an undeniable fact that historical studies repeatedly confirm. But such
coexistence does not exclude the possibility of a balanced relationship
between the two diseases, of a subtle competition that in certain circum-
stances could swerve sharply in favor of one and mal<.ethe other almost
disappear. At the moment when leprosy invaded Europe, tuberculosis was
well established, but we do not know its frequency or its preferences
among social and age classes. If tuberculosis really did become an effective
antagonist of leprosy across the population of Europe at a moment in
history near the end of the Middle Ages, it may have been the result of
specific epidemiological conditions, of a sociobiological disturbance in the
pathocoenosis whose underlying causes we do not yet understand.
Finally, the first argument calls for the kinds of experimental investiga-
tion and meticulous observation that perhaps will provide the basis for an
answer to my remarks above on the riddle of pathocoenotic equilibrium
among the mycobacteria. It is undeniable that tuberculosis and leprosy,
especially lepromatous leprosy, can be associated in the same patient. There
is even an osteoarchaeological instance of it. 19 However, this only proves
that the relation between Koch's and Hansen's bacillus is not a simple
antibiosis-no one said that it was-and that in some individuals resistance
to both germs can be lacking at one and the same time. It is enough that
a tuberculous infection can protect even a relatively limited number of
individuals against leprosy; that could produce, on the level of popula-
tions, a complex antagonism between the two diseases. An essential and
often neglected factor in this is the age of the patient at the time of
infection. We know the receptivity of children to leprosy. Does age at the
moment of first infection with tuberculosis play a role in the ultimate
reinforcement of an individual's resistance to the germ of leprosy? The
question is interesting because the average age of first contact with the
tubercle bacillus is closely dependent on sociocultural factors. In long-term
processes, a slight shift can ultimately reverse tendencies and make one
disease do1ninate or even almost eliminate another.
The Epidemiologyof Tuberculosis

In today's world, tuberculosis is a ubiquitous disease with an incalcula-
ble number of victims. According to cautious estimates published by the
World Health Organization, there are now on this earth more than
15 million cases of flourishing, infectious tuberculosis. At least I million to
2 million tuberculosis-sufferers die each year. New medicines have strongly
reduced tuberculous mortality. However, it seems that current reductions
in the frequency of this disease are less the result of specific chemotherapy
than of BCG vaccination, genetic increases in resistance, and, above all,
the improven1ent of hygiene. In developing countries, especially in Africa,
it is still one of the major causes of disease and death. In some underde-
veloped countries, more than 70 percent of the children 14 years of age are
infected with it, as against 2 percent in countries with more favorable
socioeconomic conditions. Tuberculosis beds well with debilitating para-
sitoses like malaria and schistosomiasis and especially well with malnutri-
tion. One can speak of a true pathocoenotic symbiosis of these diseases. In
Greece, that symbiosis has been plain to see since the Turkish occupation.
In some rural regions, it lasted into recent times. 20 We should not forget
that even in the 1930s respiratory tuberculosis still held a place in official
Greek statistics as one of the chief causes of death. 21
In the history of modern Europe, the spectacular rise in pulmonary
tuberculosis, which reached its climax between the latter half of the eight-
eenth and the first half of the nineteenth century, has often been ascribed
to large-scale urbanization. That can only be partially true. The medical
historian must take account of the fact that tuberculosis also raged in rural
environments, which were certainly its original haven. In ancient societies,
an important factor in its spread was horizontal, transspecific transmission
made possible from close contact with domestic animals, as when farmer
and cattle live in a single dwelling. In modern times the transmission of
tubercle bacilli is mainly from human to human, and in this regard no one
could deny the importance of demographic density and crowded living
spaces. But in themselves they cannot explain the historical vagaries in the
distribution of malignant forms of tuberculosis. R. and J. Dubos devised
a felicitous expression for the set of conditions mentioned above that are
capable of igniting a tuberculosis endemic: physiological misery. 22
Unlike leprosy, tuberculosis is an extremely contagious disease. Its bacilli
enter the human body chiefly by inhalation and ingestion, but infection
can also take place via the skin or placenta. The clinical form that the
disease assumes depends largely on the resistance of the individual infected.
The immediate result of infection is often subclinical or even completely
lacking. From my point of view here, it is not insignificant that resistance
to tuberculosis is genetically based. The immunity acquired after contact
with Koch 's bacillus is a particular form of "infection immunity" called
premunition; its persistence is ·determined by the continuous presence in
the human body of the infecting agent. It follows that in the epidemiology
of this disease, the notion of "virgin ground" that is helpful, say, in the
study of some viral diseases must be used with extreme caution. In the
case of tuberculosis, acquired immunity is not preserved after the destruc-
tion of the germ in the host organism. The eventual growth of resistance
in healthy individuals cannot result from acquired immunity but is the
expression of genetic increases in natural resistance by the process of selec-
tion. This is a historical process that, if it takes place at all, must be very
slow.
Different ethnic groups do not all offer the same natural resistance to
tuberculosis. For instance, blacks and American Indians are less resistant
than whites. Family differences can also be very pronounced. K. Pearson
has shown that, in the incidence of overt forms of pulmonary tuberculosis,
statistical correlations are much greater between parents and children than
between spouses. Experimental research by Lewis and Lurie on laboratory
animals has shown that any acquired resistance to tuberculosis is just a
specific increase from the mechanisms of natural resistance. The physiolog-
ical n1isery mentioned above gets in the way of these natural immunologi-
cal mechanisms and so becomes a dominant factor. 23 Finally, I point out a
recent discovery by French researchers, the isolation of M ycobacterium af
ricanum) an agent of human, bovine, and perhaps simian tuberculosis with
traits intermediate between those of M. bovis and M. tubercu/.osis.24
Diagram of the Evolution of Mycobacteria

Does Koch's bacillus derive from Hansen's, or vice versa? The question
is a poor one, and it is not surprising that all those who have sought to
answer it, however learned, have gone astray. No living species can derive
from another existing species. These bacilli can only have a common an-
cestor whose form and other properties can and even should be closer to
one of the two species (conservative or plesiomorphic branch) than the
other (anagenetic or apomorphic branch). That is why the genetic relations
within a monophyletic group must be represented not by a genealogical
tree but by a cladogram. 25
The genus Mycobacterium derives from a common ancestor (I have called
it M. archaicum) whose descendants diversified and specialized themselves
for subsistence in very different environments. The degree of specialization
in the way of life of a given species is a valuable indicator of its apomorphy.
The founding species lived in a free state. From it derive, on the conser-
vative line, current free-living species and probably some saprophytic ones.
The anagenetic branch includes species that live in association with multi-
cellular organisms as pathogenic parasites. At the origin of this branch were
species that became associated with sea animals (probably placoderms) in
the very distant past. On the conservative line, their descendants are the
current parasites of fish, and on the anagenetic side, the parasites of all
other vertebrates. The initial association of mycobacteria with animals must
have taken place before the passage of vertebrates from aquatic to terrestrial
life. I would date it around the Upper Devonian period, a geological era
that goes back more than 300 million years. As for the rest of the evolution
of parasitic mycobacteria, until humans appeared it was parallel to the
evolution of the host organisms. Basically it seems to be the result of the
evolution of the vertebrates and the vertical transn1ission of their parasites.

To be sure, that is only a hypothesis offered up for criticism and skepticism.
In an article that is relatively recent yet largely outdated in its microbio-
logical and phylogenetic documentation, J. Grober asserts that the hy-
pothesis of a parallelism between the evolution of mycobacteria and their
hosts is not more likely than an alternative explanation, that of ''furtive
multiple introduction" by repeated mutation of free saprophytes into a
variety of host species. 26 In my opinion, he is wrong. His alternative
hypothesis would perhaps be defensible if the number of mycobacterial
parasitic species and that of their host species were as small as he seems to
believe. It is a theory that demands too many coincidences. Moreover, the
microbes living as saprophytes on Metazoa (for instance, M. smegmatis)
are more likely to be descendants of an originally pathogenic germ than
the ancestor of such an agent.
The cladogram I have established (Figure 1) on the assumption of evo-
lutionary parallelism corresponds to the real morphological and biochemi-
cal "distances," insofar as they are known to current research, between
the various mycobacteria. As for the genetic relation between Koch's ba-
cillus and Hansen 's, it is my opinion that (1) a direct mutation of the
ancestor of the tuberculous branch into an ancestor of the leprous branch
or vice versa is ruled out, and (2) the separation of the tuberculous branch
is prior to that of the leprous branch. Against Cockburn, I think it is not
likely that leprosy is a mycobacterial disease transmitted vertically from one
generation to the next, from the fossil hominids to Homo sapiens.One can
suppose that M. lepraeresults from a horizontal infection not older than
the Neolithic Age that resulted in the mutation of a parasite of rodents,
either M. lepraemurium or a species that has disappeared or is as yet un-
known. If humans have mycobacteria that are proper to them as the result
of a parallel evolution, they must be found among the microbes that now
inhabit their bodies in silence or are responsible for minor complaints, not
among the germs of leprosy and tuberculosis. Tuberculosis was transmit-
ted as far as cattle, either vertically or by a horizontal infection originating
in birds. From there it passed into humans as an unfortunate consequence
of the domestication of animals and a settled way of life.
The biological "distance," that is, the difference in specific properties,
between M. lepraeand most of the other mycobacteria (except for the
tuberculous group) is greater than that separating the tuberculous branch
from the rest. I have already listed the traits of M. lepraethat attest to its
maximal specialization, its perfect and almost exclusive adaptation to hu-
mans. That adaptation in itself suggests that leprosy is an "ancient" disease
in terms of human history (the realization of an immunological equilib-
rium between the germ and its host means that leprosy is older than
plague, for example), but on another level, that of the history of mycobac-
teria, extreme specialization is a sign that a species is ''young'' in relation
Figure 1. Cladogram of the Genus M ycobacterium
1. archaicum
......
-----------------------------------M. aquae, M. terrae
•
•
•
M. balnei
M. piscium
M. thamnopoecilus
"---------------------M. chelonei
--------------------Various para-
tuberculous species
'------------------M. avmm
M. bovis
•
•
•
•
• M. africanum
•
•
.
•
.
•
.
•
•
•
•
M. tuberculosis
• •
•
• .
•
•
M. m1crot1
•.
.
•
..
• •
•
•
• .•• •
.•
• .• •
•
•
• • M. lepraemurium
• • •
•
.
•
• . •
•
.•.
• •
•
•
•
• • M. leprae
• •
• •
• •
.
•
•
•
•
•
~
•
•
• .
•
300 million years ago 25000 B.C. 2000 B.C.
208
to other species of the same genus. The absence of leprosy fro1n the pre-
Colun1bian New World is no mean argument in favor of its "youthful-
ness." It suggests a terminus post quem: the appearance of leprosy should
have happened after the interruption of prehistoric human migrations
between Asia and America, that is, after the disappearance of the land-
bridge across the Bering Strait. 27 The date of this interruption is not pre-
cisely known: more timid hypotheses place it around 10,000 B.C., and the
most rash at around 25,000 B.c. What seems to me most reasonable is to
place the date of the origin of leprosy, of the birth of M. leprac) between
this date and, as I showed above in the analysis of historical documenta-
tion, the beginning of the second millennium B.C.
Some historical and biological considerations lead one to believe that
the origin of M. tubercu/.osisdoes not go further back than the Neolithic
Age. At first tuberculosis spread from cattle to humans. With the devel-
opment of herding in sedentary populations, 28 this infectious disease be-
came more common, and a strain of ancient tuberculous microbe adapted
itself for survival in richly oxygenated human pulmonary tissue. It seems
likely that M. tubercu/.osisappeared even later than M. leprac. That would
contradict neither my cladogram nor what I was just saying about the
differentiation of mycobacteria. To be sure, in that case it must be asserted
that the pre-Columbian osteoarchaeological cases in the New World are all
due to humans being infected with M. bovis) or more precisely to a myco-
bacterium of which M. bovis is the plesiomorphic descendant. There is no
need to confuse the origin of M. tubercu/.osiswith the older origins of the
tuberculous branch, that is, of the common ancestor of the existing species
M. bovis) M. africanum) and M. tubercu/.osis. The birth of M. tubercu/.osis
probably took place in Africa from a "bovine" form (that is, by horizontal
transmission) and through the intermediary of the "African" form. In
classical Greece, M. tubercu/.osiswas already present alongside M. bovis.
Chapter Nine
THE HARM IN BROAD BEANS

Legendand Reality
The physicians of classical Greece understood clearly how important envi-

ronment can be for keeping the components of an organism in equilib-
rium. 1 They were aware, that is, of the ecology of health. Treatises
attributed to Hippocrates-actually, the very ones it has become custom-
ary to praise to the skies-have a valid appreciation of the role that air, soil,
and water play in the etiology of disease. Other texts in the Hippocratic
corpus treat with remarkable philosophical insight and impressive technical
mastery the medical issues of diet and exercise. These writers had observed
the hereditary nature of epilepsy, strabismus, and certain deformities. In
special cases, such as gout, a disease that corresponds better than any other
to the explanatory principles of the pathology of humors, or pulmonary
consumption, ancient practitioners began to understand some rules of the
complex interaction between the innate and the acquired. They caught
glimpses of the necessary nexus of internal and external factors, of disposi-
tion and triggering cause. But there are pathological states in which this
two-sided causality is so complex that reason is n0 longer able to track
down all the interconnections. In such cases, physicians can just refuse to
see parts of reality in order not to compromise the rationality of the overall
system. It is possible that in order to satisfy such needs for coherence
certain intuitions from the archaic period had to be suppressed. A good
example is the notion of infection, which, since it was cloaked in the
magico-religious concept of pollution, found no quarter among the cham-
pions of rational medicine. Out it went, and as a result certain events in
the day-to-day practice of medicine became literally invisible. A core of
truth too well hidden in mystical trappings ended up on the trash heap of
superstition. The event ultimately hindered subsequent attempts at a the-
210
THE HARM IN BROAD BEANS 2II
ory of infection that better answered to the demands of scientific rigor. It

JI'
is my purpose now to examine closely the history of a certain medical

theme in which, at one point, magical thought was probably nearer reality
than strict reason.
A Pythagorean ltohibition and Its Justifications in Antiquity

A surprising rule of conduct, kuamon apekhesthai 'abstain from broad
beans,' is included among the Pythagorean sumbola) or pithy teachings,
whose esoteric meaning escaped the common run of mortal men. 2 Since
classical antiquity philosophers and historians have been fascinated by this
prohibition. It was attributed to Pythagoras himself, the famed sage of the
sixth century B.C., but Aristotle was aware that by his time the personal
lucubrations of the master could not be clearly distinguished from the
mass of material from his school. The person and teachings of Pythagoras
are veiled in legend, transformed in the image of late accounts that divinize
a charismatic leader and drown his original thought by systematizing it
into a Platonizing doctrine that is more Hellenistic than archaic. 3
At the outset I wish to stress two special aspects of the historical docu-
mentation on which our knowledge of the Pythagorean teaching about
broad beans rests. First, the existing texts that attest to it directly are not
earlier than the first century B.C. (they are Latin authors, like Cicero and,
in a fairly obscure allusion, Horace), or if we consider only Greek sources,
not earlier than the second century A.D. (Plutarch, Lucian of Samosata,
Artemidorus, Clement of Alexandria). No information has come down to
us on the exact words that Pythagoras used to formulate the tabu. The
only way we happen to know of Empedocles' and Callimachus's verses
about broad beans as well as the opinions of some fourth-century B.C.
authors is through readers' citations. 4 On the other hand, the multiplicity
of Greek and Roman witnesses on this subject, their recourse to earlier
authors, and the already incomprehensible (to their eyes) nature of the
prohibition lead me to believe that it is indeed older than the classical
period in Greece. In a word, it seems to me that one can reasonably
attribute to Pythagoras or the Pythagorean community at Croton a partic-
ular attitude toward broad beans; but one cannot determine that it origi-
nated there or what exactly its original form was. The Samian master's
words, the famous ipsedixit) have not fled the barrier of time. Scarcely a
few generations after the death of Pythagoras, the express motivation and
original justification for the tabu were totally lost.
What, precisely, does the Pythagorean prohibition consist of? To begin
with, it concerns a legume called kuamos in Greek and securely identified
with the broad bean, Viciafaba L. (Fabavu!garisMonch; also called horse
bean or fava bean). The broad bean comes either from the region south of
the Caspian Sea or from North Africa. It flourished in the wild in Persia
and on the northwest coast of Africa. Its use as food, its cultivation, and
its spread around the Mediterranean go back to prehistoric times. 5 Its seeds
have been found in very ancient archaeological sites in Italy, in Swiss lake-
dwellings, and in Egyptian tombs from the Pharaonic period. They have
also been reported in Mycenaean tombs and in the ruins of Troy, begin-
ning with the layers that date from the Early Bronze Age. Homer compares
an arrow glancing off Menelaus's breastplate to the way "blackskinned
broad beans" fly at winnowing time, which presupposes that his audience
was familiar with this peasant activity. 6
Ancient broad beans were smaller than modern ones, but they were
doubtless ancestral varieties of Viciafaba) not a similar species that has by
now disappeared or is not appreciated by modern farmers. From the bo-
tanical description in Theophrastus, it is certain that the Greek term ku-
amos (consistently translated by the Latin faba) denotes in the first place
the broad bean, the whole plant as well as the seeds meant for eating. 7
However, a distinction should be drawn between the substantive kuamos
by itself or the expression kuamos Hellenik6s 'Greek broad bean' and the
expression kuamos Aiguptios 'Egyptian broad bean'; they are set apart im-
plicitly in the Hippocratic corpus 8 and explicitly by Dioscorides. 9 The
former designates the broad bean as such, while the latter is the pink lotus,
an exotic plant known in Greece for its medicinal uses, not as a food.
The existence of this kuamos Aiguptios and of the chapter Dioscorides
devotes to it has led some historians and botanists since the sixteenth
century to consider the Pythagorean kuamos to be the edible seeds of the
nelumbo (Nelumbium speciosumWilld. [Nelumbo nucifera Goert. and Nym-
phaea nelumbo L.]). 10 No author from antiquity suffers from such a confu-
sion. Herodotus and Theophrastus discuss the two separately. It was
known in antiquity that the latter, a kind of lotus, was considered sacred
by certain Oriental peoples, though they did not forbid their priests to
partake of it. In the modern literature on this subject some incorrect
statements are to be found that result from a vicious circle: the identifica-
tion of kuamos in Herodotus and Pythagoras with the nelumbo is made
and then used to prove that it is correct. Pliny, who speaks of both "broad
beans'' (he call the first faba and the second faba. Aegyptia or colocasia),
mentions the Pythagorean prohibition only apropos of the common broad
bean. 11 Plainly, his opinion in that regard was shared by other ancient
authors, since it is hard to see why they would have discussed the subject
at such length if in their opinion the Pythagorean rule concerned only a
plant that was not usually eaten, was little known in Europe, and was
considered sacred by foreigners to boot. What fascinated these thinkers
was precisely the fact that the Pythagoreans forbade the consumption of a
very popular food.
Several other botanical identifications have been proposed by modern
scholars. I think they are to be rejected. Out of sheer ignorance some
THE HARM IN BROAD BEANS 213
physicians translate kuamos as "haricot bean" (Phaseolussp.), a plant not

introduced into th'e Old W odd until the discovery of America. 12 Still more
anachronistic is the desire to identify Pythagoras's bean with a toxic species
similar to the Calabar bean. L. Livet believes that "the bean of which
Pythagoras speaks is none other than henbane. " 13 He stresses that the
word huoskuamos (which designates hen bane, Hyoscyamusniger L.) is com-
posed of hus and kuamos and means "wild boar bean." Consequently,
Pythagoras and his pupils could have meant this plant and used only the
generic portion of its usual name. Actually, henbane, which is a poisonous
plant inspiring holy terror, was well known to ancient Greek and Latin
authors, who were careful not to confuse it with a harmless legume. 14
Liver's hypothesis, which is all conjecture, tries to explain the Pythagorean
prohibition by reducing it to something so obvious that the interest it
generated among ancient philosophers becomes incomprehensible: with
or without a tabu, no one would be tempted to think of a notoriously
toxic plant as food.
But it is worth considering whether or not the Pythagorean rule origi-
nally envisaged the use of broad beans as food. The wording most likely
to be closest to the original formulation occurs in two citations that besides
being early are also in verse. As a result, they were less exposed to corrup-
tion in the process of being transmitted, whether orally or in writing.
Empedocles of Agrigentum, a philosopher and physician of the fifth cen-
tury B.C. who in some ways was a perpetuator of Pythagorean teachings in
Magna Graecia, says, ''Unhappy, very unhappy one, do not touch broad
beans. " 15 Callimachus, a poet of the third century B.c., authored this
couplet: ''Do not touch broad beans, a food some devour, I tell you this
just as Pythagoras did. " 16 It is striking that both these writers use the same
warning formula: kuamon apo cheiras ekhesthai (or ekhein), which literally
means to keep broad beans far from one's hands. Certainly the significance
of this recommendation is far from clear. Its ambiguity has allowed diver-
gent interpretations, as I will illustrate in the course of this historical
analysis. However, the appositive "a food some devour" in the citation
from Callimachus suggests strongly that the context is nutritional. Other
testimonia, numerous but late, are more explicit in the same direction. So
Cicero brings up ''the Pythagorean prohibition against eating broad
beans"; 17 Iamblichus speaks of it in his chapter on the food of the Pytha-
goreans, in which he places it alongside the prohibition against eating· an
animal's heart or its brain, two types of saltwater fish, and mallow. 18
Diogenes Laertius insists that Pythagoras "above all forbade the eating of
mullet and sea perch as well as animal hearts or broad beans. " 19
So the broad bean was not the only food that Pythagoreans were forbid-
den to eat. But it is readily apparent that the context of the prohibition,
though it is, broadly speal(ing, nutritional, is not really concerned with
health in the medical sense. It is a section of the sacred diet, not a chapter
on profane nutritional hygiene. The prescriptions cited above derive from

the principle of ritual purity, not health. The distinction emerges clearly
in a text that Alexander Polyhistor, a writer of the first century B. c., says
he found in the Memories of Pythagoras: "Purity is obtained by purifica-
tions, ablutions, sprinklings, from not having had contact with a corpse, a
woman, or any other defilement, and by abstaining from the meat of dead
animals, mullets, blacktails, eggs, birds born from eggs, broad beans, and
all the things forbidden by those who are in charge of celebrating the rites
in sacred ceremonies. " 20 According to Artemidorus, the broad bean is
forbidden in all religious ceremonies. 21 Porphyry says that "initiates into
the Eleusinian mysteries must abstain from domestic birds, fish, broad
beans, peaches, and apples. " 22 Pausanias must be alluding to the same
custom in the following passage from his description of the Sacred Way at
Eleusis: "On the road a small construction of Cyamites has been built; I
cannot say whether this hero was the first to have cultivated broad beans
or if he has his name from the fact that it was impossible to attribute the
creation of broad beans to Demeter; those who have witnessed the cele-
bration of the mysteries at Eleusis or who have read the Orphic poems
know what I am referring to. " 23 The Orphics' horror of broad beans is
likewise attested in several other Greek texts. 24 The abstinence from broad
beans in the myth of Amphiaraus, the divine dream-interpreter, belongs
to the same current of mysticism. 25
This particular attitude toward broad beans is not the only link connect-
ing Orphic teachings, Pythagoras, and Empedocles. In all three there is a
philosophical and religious system with cosmological elements and ethical
or ascetic consequences whose common roots are undeniable and at whose
core are the magical, pantheistic concept of nature and the notion of the
immortality and transmigration of the soul. 26 Analogous concepts and
practices mark ancient Hindu philosophy. It has even been said that the
tradition of the Ayurveda imposes abstinence from broad beans on the
sacrificer and that the superstition is therefore an Inda-European one that
the Achaeans brought with them when they migrated into the Mediterra-
nean basin. 27 That is highly uncertain. Greeks could have received philo-
sophical ideas from India in the archaic period once ~hey were well settled
along the perimeter of the Mediterranean basin. It is generally admitted
that by the time of Pythagoras Hindu philosophical thought had already
reached Persia. In recognizing this cultural link, one should not overlook
the possibility of intellectual exchange in both directions. If the prohibi-
tion against eating broad beans actually occurs in the religious or medical
practice of India, I see in that a borrowing from the West instead of the
other way around. Natural historians maintain that the broad bean was
not cultivated in prehistoric India or the Chinese world: The plant itself
has no native name in Sanskrit or in any modern Indian language. Those
who assert the contrary are mistaken in the botanic identification of the
ancient terms. 28 '
If abstention from broad beans is not a practice invented by the Greeks
themselves, it could only have come to them from western Asia or North
Africa. That is the inescapable conclusion from geographic data concerning
the spread of the bean's culture. The same conclusion also rests on histor-
ical documentation. To quote Herodotus, "as for [broad] beans, [the
Egyptian priests] cannot even bear to look at them, because they imagine
they are unclean (in point of fact the Egyptians never sow broad beans,
and even if any happen to grow wild, they will not eat them, either raw or
boiled).' ' 29 That was surely the situation around the middle of the fifth
century B.c., when Herodotus was writing, and perhaps also during a
more or less lengthy period before then, but it does not seem to have been
the case in Pharaonic times. Broad beans have been found among the
mortuary offerings in ancient burials, and Egyptian documents nowhere
condemn the cultivation or use of the plant. 30 In any case, what Herodo-
tus says proves that the Egyptians of his day knew of broad beans. The
plant grew there, either wild or cultivated by persons whose hunger got
the better of their religious scruples-and as de Candolle remarks, it was
probably grown on farmland, since the soil suited to it was generally
farmed. In the Egyptian scheme of things, the prohibition against the
broad bean was relatively recent, but it still can be older in Egypt than in
Greece.
Pythagoras is said by all three Vitae to have been a disciple of the Egyp-
tian priests. Modern opinion is reluctant to credit the late stories about
Pythagoras's visit to the temples of Memphis and his knowledge of Egyp-
tian language, but even so one cannot reject as nonsensical the opinion of
Isocrates about the Egyptian background of Pythagorean philosophy or of
Herodotus on the theory of metempsychosis. 31 Essentially, Herodotus
says that the Egyptians were the first to formulate the idea that the human
soul is immortal and that it migrates from a dying body to another, living
one. The chapter on this subject concludes with a transparent allusion:
''This theory has been adopted by certain Greek writers, some earlier,
some later, who have put it forward as their own. Their names are known
to me, but I refrain from mentioning them. " 32 Most likely Herodotus had
his contemporary Empedocles in mind, along with the ancient Orphic
poets and, to be sure, Pythagoras. ·
Is it a matter of chance that all those who espoused the transmigration
of souls forbade the eating of broad beans? Is it possible to understand the
prohibition as a practical consequence of the general idea of 1netempsycho-
sis? From antiquity to the present, there have been those who thought so.
Celsus-not the medical writer but the anti-Christian polemicist of the
second century-was clearly trying to be systematic when he said that
"Pythagoras and his disciples ate no broad beans or anything else that had
life. " 33 But matters are not that simple. Although it is relatively easy to
deduce the prohibition against killing animals and eating their flesh from
the overall doctrine, it is hard to see why the same should be true for a
plant. And exactly why should it be the broad bean? Besides, it is not even
certain that the Pythagoreans advocated strict vegetarianism. Aristotle
maintained that they "abstained from the womb and the heart of animals,
from the sea-anemone and other animals like it, but used everything
else. " 34
But before examining the various attempts to get at the underlying sense
of Pythagoras's rule, it is crucial to highlight one aspect of it that is often
neglected. The Pythagorean dread of broad beans did not stop at the
prospect of simply eating them; it implied that one should not even set
foot in a field of them. It was forbidden to trample them. 35 It is not known
whether this prohibition is a corollary of the original one or a secondary
extension of it. Whatever the case, legend ascribes it to the master himself
and makes his obedience to it in exemplary fashion part of the story of his
death. It relates how Pythagoras at a venerable age surrendered himself to
death out of weariness with life, or, alternatively, was killed after an at-
tempted flight was cut off by a field of broad beans that he chose not to
cross. Diogenes Laertius was inspired by the latter account to compose the
fallowing verses:
Alas! Why did Pythagoras so dread the broad bean?

Behold him dead with his disciples.
There lay a field of broad beans, and so as not to trample them,
He died at the fork, slain by the Agrigentans. 36
Diogenes also tells us how Pythagoras and his followers were attacked by
the Agrigen tans. The house where they were lodging was set on fire by a
man who was disappointed at not having been chosen as one of the
philosopher's disciples. Pythagoras fled: "He was met near a field of broad
beans; he refused to cross it, saying that he preferred to be killed than to
trample the beans with his feet, and adding that he preferred to die rather
than to speak. His attackers put him to death and, most of his followers
along with him-there were about forty. " 37
This story does not hold together and is totally fabricated, but it con-
tains a moral that proves the importance and the extent of the tabu about
the broad bean in the Pythagorean tradition. The statement by Pythagoras
that it is better to die than speak does not follow from what precedes it.
Armand Delatte was right to see it as a sign of the contamination of the
story of the master's death with another legend from the hagiographic
cycle of the Pythagoreans. Here is that other story as told by Iamblichus,
in a version he says he had from Hippobotus and Neanthes, philosophers
and historians of the end of the third and the beginning of the second
century B.C.: Dionysus, the tyrant of Syracuse, wanted to learn the secrets
of the Pythagoreafis. So he ordered his henchmen to bring him, even
against their will, some members of the sect. An ambush was staged near
Tarentum and a group of ten Pythagoreans was taken by surprise. They
disengaged themselves and began to flee; making headway on their attack-
ers, who were heavily armed, they had the misfortune to arrive at the edge
of a broad bean field in full flower. Stopping there and "not wanting to
violate the prohibition against touching broad beans," Pythagoras's disci-
ples were massacred in obedience to the teaching that it is better to die
than be captured. On their way home, Dionysus's men met Myllias of
Croton and his wife Timycha of Sparta, Pythagoreans who were lagging
behind their group because the woman, who was pregnant, had a hard
time walking. Her state also permitted Dionysus's men to capture the
couple and bring them before him. The tyrant then demanded that they
reveal the mysteries of their sect. Despite his threats and enticements,
Myllias and Timycha refused to betray their secrets. Dionysus insisted and
ordered them on pain of death at least to explain to him the reason why
their fellows had been unwilling to walk on the broad beans. M yllias
answered, ''They chose death so as not to trample on beans, and I myself
would rather trample on beans than reveal the reason for the prohibition."
His wife, a Spartan, proved the extreme loyalty of her character by biting
off her tongue and spitting it out in front of the stupefied tyrant. 38
Voltaire said that martyrs make converts. It matters little that the story
of Myllias and Timycha is a tissue of improbabilities. At the time its main
purpose was moral edification and propaganda for a sect. For us, it en-
dorses a broad interpretation of the bean tabu in the Pythagorean com-
munities and confirms the fact that the official justification of it was utterly
esoteric. Chances are that, being reserved for a closed circle of initiates,
said justification was mythical rather than physiological, mystical rather
than rational. But nothing in all this is certain. At least the Pythagorean
role of silence explains why the persons in antiquity who dared write on
this subject were already in the dark. Aristotle-the "master of those who
know,'' according to Dante-says that Pythagoras "proscribed broad beans
either because· they have the shape of testicles; or because they resemble
the gates of hell, for they alone have no hinges; or again because they
spoil, or because they resemble the nature of the universe, or because of
oligarchy, for they are used for drawing lots. " 39 The multiplicity of reasons
advanced proves that the justification Pythagoras and his immediate disci-
ples gave was unknown. And the wealth of explanations is no embarrass-
ment to the eclectic Iamblichus, who states that "one should abstain from
broad beans for several reasons, some religious and natural and some hav-
ing to do with the soul. " 40
Did people really believe that broad beans could be the home of souls
in the process of transmigrating? It is tempting to understand an obscure
Orphic line in those terms: "Eating broad beans and gnawing on the head
of one's parents are one and the same. " 41 According to Pliny, some writers
were convinced that Pythagoras condemned the use of this plant "because
the souls of the dead are in the bean." And he adds that "in any case, that
is why they are used in offerings for the dead" and that "according to
Varro, that is why the Jlamines do not eat them. " 42 In the Dream) a comic
play by Lucian, a rooster with the gift of speech comes on stage and
introduces himself to a shoemaker as the reincarnation of Pythagoras. The
shoemaker is confused and says the rooster must be lying-he is too talka-
tive, and he ate some broad beans. Here is a portion of the argument that
the shoemaker addresses to the rooster: ''If you are really Pythagoras, you
have broken your own laws, and when you swallowed the beans you
committed a sacrilege as great as eating the head of your own parents.''
The rooster, more of a sophist than a Pythagorean, answers that different
rules of behavior are appropriate for each form of life, and that he abstained
from broad beans when he was Pythagoras but as a rooster he is no longer
obliged to heed the tabu. 43
In another work by Lucian, Sectsfor Sale) there are several philosophers
auctioning off their ideas and praising their own moral and dietetic pres-
criptions. This is the way Pythagoras presents his way of life: "I feed myself
nothing that has been alive; I eat everything else except broad beans.''
When a buyer is surprised by this exception, Pythagoras explains: "I con-
sider broad beans sacred. Their nature is in some ways to be admired, for
they encompass in themselves all kinds of procreation: if you shell them
when green, you will see that they closely resemble a man's testicles; and
if, after cooking them, you expose them for several nights to the moon-
light, they bleed.' ' 44 So it seems that in Pythagorean thought, the broad
bean is actually related to humanity not so much by way of metempsycho-
sis but in terms of mystical analogies and cosmic speculation. Aristotle
adduced the resemblance of the seeds to testicles. The flower also has its
symbolic aspect: "Pythagoras says that one should not eat broad beans,
because there are letters of mourning in their flowers.' ' 45 The Orphic line
that likens broad beans to the head of one's parents is cited by Plutarch
during a discussion of the egg, its cosmological sigDificance, and the pro-
hibitions concerning it. According to Porphyry, the Pythagoreans taught
that humans and beans sprang from the same original matter. Broad beans
in flower, he declares, if they are closed up for some time in a terracotta
vase and kept moist, can be transformed into the head of a child or female
genitals. 46
Critical observers of nature must have found such tales amusing. It is
not surprising if some found the sacred dread and magical mystique of
beans just so much superstition. For a rationalist, there were two ways to
preserve Pythagoras's prestige: to deny the tabu of broad beans (that is, to
give it a sense different from the accepted one), or to provide it with a
"physiological" justification. Both ways of keeping up appearances were

tried, and not inftequently. The oldest testimonia on the tabu itself are
not free from the rehabilitation of magic through reason. At the end of
Aristotle's list of widely divergent explanations, he mentions the use of
beans for drawing lots in political life. Here is an interpretation that ne-
gates the nutritional one: actually, Pythagoras was prohibiting his disciples
from touching broad beans not as food but so that they would not take
part in the public responsibilities of a democratic city-state. Plutarch takes
up and develops this idea, considering the Pythagorean adage a rule for
social order. 47 But Iamblichus denounces this explanation, saying that it
comes from a political pamphlet by Ninon whose purpose was to pervert
the ideas of Pythagoras in order to stir up the people of Croton in oppo-
sition to the teachings of his immediate successors. 48
Aristoxenus of Tarentum, a disciple of Aristotle and the Pythagorean
Xenophilos of Chalcis, took another tack. As a musical theoretician, he
appreciated the work of Pythagoras, but in his loyalty to peripatetic ration-
ality, he had no love for the mystic side of Pythagoreanism. If we are to
believe Gellius, Aristoxenus said in a book on Pythagoras that Pythagoras
"never enjoyed a single vegetable more often than broad beans, since it
was food that gently calmed and soothed the stomach." This is the op-
posite of what is known from other sources. Aristoxenus's words were,
"Broad beans were the vegetable Pythagoras preferred, calling them sooth-
ing and laxative; he enjoyed them often. " 49 Gellius, who reports and
appropriates this correction "of an old error that has carried away many,"
also sets about ridding the famous verse of Empedodes of any dietary
content:
People thought that kuamiJn 'broad beans' designates the vegetable, as in ordinary
usage. But those who have interpreted the poems ofEmpedocles with greater care
and sophistication say that kuamon refers here to testicles, and that he called them
beans in a secret, symbolic way, after the fashion of Pythagoras, because they are
a[tioi tou kuein 'to blame for pregnancy,' and because human procreation gets its
impetus from them; so Empedocles in this verse was not trying to keep men from
eating broad beans but to abstain from sexual pleasure. 50
Another way to rehabilitate the Pythagorean precept was, as I said above,

to find natural, "physiological" justifications for abstinence from broad
beans. It is a vegetable that provokes flatulence, a bloated abdomen, and
rumbling in the stomach, which are improper during sacred rites or in a
sanctuary, as Artemidorus tells us. 51 But the release of flatulence was less
fearful in antiquity than its retention within the organism. Otherwise it
rose to the head, disturbing the reason of those awake and giving false
dreams to those asleep. That is Cicero's opinion: "Plato prescribes caring
for one's body on going to bed in such a way that nothing leads the soul
into error or disturbance. It is thought that the Pythagorean prohibition
against broad beans derives from that, since they cause substantial flatu-
lence, which is against the search for the tranquillity of the soul. " 52 Here
are some other passages in the same vein:
The broad bean is said to blunt the senses and stimulate dreams; that is why the
teaching of Pythagoras condemns it [Pliny the Elder]. 53
One should abstain from broad beans, since ~hey are full of wind and take part in
the soul, and if one abstains from them one's stomach will be less noisy and one's
dreams will be less oppressive and calmer [Diogenes Laertius communicates this
version of Pythagoras's teaching; he thinks it superior to others]. 54
The naturalists say that broad beans dull the stomach of those who eat them. That
is why they prevent a person from having true dreams, since they cause flatulence
[an anonymous Byzantine author who was probably dipping into Didymus of
Alexandria]5 5
A rationalizing explanation that is peculiar and unknown to others occurs

in Clement of Alexandria (second century A.D.): the lawgivers, he tells us,
forbade the eating of the broad bean because it made women sterile. 56
Before concluding this review of the ancient evidence, we should con-
sider the enigmatic expression that bedecks a Horation verse: faba Pytha-
gorae cognata.57 What does "broad bean, sister of Pythagoras" mean? At
first sight, the poet seems to want to establish a close kinship between the
bean and the human race. The expression belongs in the mystical tradition
on the common origin of the two species and the transmigration of souls.
But that is not the only interpretation: it is possible that the broad bean is
called Pythagoras's sister because they are both equally untouchable for
him. So a relationship is established not so much between humans and
beans as between the incest tabu and eating broad beans.
BlinkeredPhysiciansand the Discoveryof Lathyrism

Tradition says that Empedocles was a physician, and Pythagoras himself
is considered an expert in the healing art. But in the eyes of posterity, both
are personages endowed with charisma and exceptional power; they are
sages, thaumaturges, or educators rather than scholars, clinicians, and wri-
ters on technical subjects. Their magical and religious attitudes toward
nature and the healthy or sick person found no favpr among the techni-
cian-physicians of the fifth and fourth centuries B.C. whose works have
come down to us under the name of Hippocrates. Greek medicine in the
classical period was an art (tekhne), and as such it purged itself of all notions
of the sacred and the supernatural, of transcendance of the determinism in
the existence of living beings and their relations with the environment. So,
interestingly enough, subsequent to the two putative fathers of the pro-
hibition against broad beans, not a single other physician in antiquity
wished to take the trouble to speak of it. All the evidence cited above
derives from writers whose occupation was not medicine.
The Hippocratic writers could not have been ignorant of the Pythago-
rean diet. They knew the writings of Empedocles. 58 But the tabu against
broad beans is never mentioned in their treatises. Plainly, they considered
it outside the competence of medical art. The texts do confirm that the
kuamos was a common food in classical Greece. As for its dietetic proper-
ties, the treatise Regimen of Acute Diseaseslimits itself to the following:
"Beans afford an astringent and flatulent nourishment; flatulent because
the passages do not admit the abundant nourishment which is brought,
astringent because it has only a small residue from its nourishment. " 59
Elsewhere, information of this kind is still more meager. Cooked broad
beans are prescribed as treatment for the dysenteric syndrome (dusenterfa)
or, more generally, for upset stomach (tarakh~gastros).60 Dioscorides (first
century A.D.) provides a good overview of the place held by broad beans
in the Greek materia medica. The chapter he devotes to this plant begins
with a summary of the common vie"r that we have already seen in several
nonmedical writers: ''The broad bean swells the stomach and makes one
break wind; it is hard to digest and produces bad dreams.'' According to
the sequel, these beans are nourishing and put flesh on one's bones. It is
suggested that they be eaten as treatment for coughing and vomiting.
Once cooked, seasoned with vinegar, and swallowed pods and all, broad
beans are said to put a stop to diarrhea, especially dysentery. The preferred
way to eat them is boiled in two waters, of which the first is to be thrown
out, since "green broad beans discomfit the stomach and cause more
flatulence.'' External application of broad bean flour (in the form of a
powder, or a plaster, or an ingredient in salves) and even of fresh seeds cut
in two is suggested as treatment for various diseases and wounds. 61
Modern medicine essentially confirms these dietetic observations and
therapeutic indications. Broad beans contain some indigestible oligosac-
charides that cause flatulence, but they can also be beneficial in the treat-
ment of some forms of diarrhea of infectious origin. However, though the
beans do tighten up the stomach when the digestive tract is in certain
states, that is not due to the slightness of their residue, as the author of
the Regimen believes, but, to the contrary, the large amount of it. It is
doubtful that the undeniable influence that the ingestion of broad beans
has on psychic activity, specifically on dreams, is solely due to the flatu-
lence they produce. Broad beans contain a fairly high concentration of
levodopa (L-dopa), a substance used nowadays in the treatment of Parkin-
son's disease. The administration of this substance increases the quantity
of dopamine, a precursor of norepinephrine, in the central nervous system,
which can produce insomnia, nightmares, or even hallucinations, as well
as increase nervous tension and stimulate sexual activity.
I have surveyed the use of words in the Hippocratic corpus that are
close in meaning to kuamos) that is, those denoting various leguminous
plants and their edible products. In book 7 of the Hippocratic Epidemics)
there is a very long list of foods that are supposed to be possible causes of
"bilious incidents" (kholerika).The items are so varied-overcooked pork,

crabs, leeks, onions, lettuces, pastry with honey, fruit, milk, wine, and so
forth-that from the standpoint of modern medicine their only common
pathogenic feature is that they are capable of feeding infection. So the
"incidents" in question are acute kinds of gastroenteritis caused by in-
fected food. However, the list also includes the chickpea (erebinthos)and
the bitter vetch (6robos))·for them, their own toxic effects cannot be ruled
out. 62 In general, the Hippocratic writers thought that "all leguminous
plants are flatulent" and that "one should not partake of them without
bread," since "each has its own disadvantages." Thus the chickpea "causes
some pain" and the lentil is "astringent and causes distress if eaten with
its pod. " 63 A passage in the Epidemics that is included in book 2 and
repeated as such in book 6 is especially instructive in this regard: '' At
Ainos, in the course of a famine, those who ate pulse continually became
powerless in their lower limbs; the condition lasted; and moreover, those
who ate bitter vetch had pain in their knees. " 64 Since it belongs to the
second group of books of the Epidemics (2-4-6), this account goes back
either to the last decade of the fifth century or the first half of the fourth
century B.c. It refers to Ainos, a town in Thrace that was suffering from
war and severe economic collapse. 65
Two key terms arouse attention: ospnophageontes and orobophageontes. Al-
though these words only occur here in Greek, their meaning is transpar-
ent. The first denotes those who eat 6spria'pulse.' It does not specify the
nature of the plants eaten, for how long they were eaten, or even whether
or not other foods were eaten along ,vith them. However, in context the
term conveys the impression that the people of Ainos fed themselves
almost exclusively for a certain period of time on vegetables with low
nutritional value. In his commentary on this Hippocratic text, Galen says
that 6spriaare the seeds of leguminous plants not usually used for making
bread. The inhabitants of Ainos ate them for a long while, without eating
bread made from wheat flour even occasionally. 66 Littre says nothing of
the famine. Indeed, the V manuscript and most of the recentioresderived
from M (which has a lacuna here) contain, in the first phrase of the passage
at Epid.) II, 4, 3, the incomprehensible words en, aimo or enafo. Littre
suppressed these words, considering them a faulty repetition of en afnoi 'in
Ainos.' There is indeed a transcriptional error, and a very old one, but of
another kind: the correct reading is en limoi ('during a famine,' which is
preserved twice in Galen's commentary, once in his note on Aph.) III,
r6, and again in his note, known only in the Arabic version, on Epid.)
VI, 4, II.
As for the orobophageontes) they are eaters of bitter vetch, since 6robosin
classical texts is indeed Vicia ervilia Willd. (Ervum ervilia L.), a plant used
for forage and known in Greece since prehistoric times. 67 Bitter vetch was
a substandard human food with medicinal uses. The Greeks knew that if
eaten in certain proportions it could have toxic effects. According to Dios-
corides, bitter vetch causes headache, disturbance of the bowels, and lower
cavity hemorrhages. 68 Pliny considers it harmful to humans, since ingesting
it is followed by headache, vomiting, and diarrhea. 69 According to Galen,
it is eaten only when food is scarce. 70 The Hippocratic writer attributes
pains in the knee to the consumption of this leguminous plant by the
inhabitants of Ainos, but he does not elaborate on the clinical features.
The observation was made amid nutritional difficulties of such a kind that
the modern physician must suspend judgment. I suspect that the patho-
genic factor responsible for the joint disease described was deficiency rather
than poisoning.
But though pain in the knee is too vague a symptom to allow valid
inferences about its etiology, the same is not true of permanent paraplegia,
the major symptom mentioned in the first part of the Hippocratic ac-
count. A loss of mobility in both legs after prolonged ingestion of pulse
on an epidemic scale among the inhabitants of a Thracian town-that
suffices for a retrospective diagnosis: lathyrism. Such is the name that the
Neapolitan physician Arnoldo Cantani gave to poisoning by certain species
of pea or vetch.7 1 It is by far the likeliest diagnosis of the chronic disease
afllicting the poor "ospriophagic" inhabitants of Ainos. Their paralyses
would then have been caused by excessive consumption of a bread pre-
pared with flour made from a species of Lathyrus) either L. sativus (chickling
pea) or L. ciceraL. (chickling vetch). 72
The first modern information in the West regarding this neurotoxic
syndrome dates from the middle of the seventeenth century. During the
Thirty Years' War, terrible famines forced people to find a substitute for
bread. In the Duchy of Wurtemberg, the population fed itself on such
legumes, not without major mishaps. In 1691, Bernardino Ramazzini, the
founder of occupational medicine, described pea intoxication in subjects
living in the Duchy of Modena. All during the eighteenth and nineteenth
centuries, isolated cases or small epidemics of the aflliction were noticed in
Europe. 73 The cultivation of chickling was particularly common in India,
Syria, North Africa, and Greece. Most of the neurotoxic accidents due to
the ingestion of seeds or flour from this plant are situated in these geo-
graphical areas. In India, lathyrism has long been known. 74 There were
often very serious poisonings from it, and at times they took on the
proportions of epidemics. In the Greco-Roman world, chickling pea was
cultivated not just as fodder but also for human consumption. 75 People
made barley bread and added vetch or chickling pea flour to it. 76 Use of
the latter as food is attested in the comic poets. 77 Such uses are explicable
since only excessive or exclusive consun1ption of the plant produces the
neurotoxic syndrome. Lathyrism became a medical problem only when
food was scarce, when a type of pea meant to feed the livestock became
the main ingredient in everyday bread, and when its consumption was
associated with important quantitative and qualitative deficiencies.
The pathogenic factor resides only in the seeds, but its nature is not yet
wholly clear. The pure chemical substances that produce lathyrogenic ef-
fects in experiments on animals have been isolated and even synthesized.
They are chiefly derivatives of aminopropionitril. 78 The absorption of the
seeds in considerable quantities leads to selective attack (degenerating le-
sions) on nerve cells in the lumbosacral marrow. The clinical picture is
dominated by muscular asthenia in the legs and vasomotor disturbances.
It evolves into spasmodic paraplegia. 79 This irreversible paralysis of the
lower limbs is the symptom that the Hippocratic writer describes with
typical, inimitable concision.
So Greek physicians of the classical age saw lathyrism. But they noticed
no accidents following the ingestion of broad beans. Couldn't we interpret
this silence as an indication of the absence of favic idiosyncrasy in Greece?
That is a hasty conclusion to come to. Favism certainly existed in Greece
in the eighteenth century, and yet no physician saw, wished to see, or
rather was able to see attacks caused by broad beans. To be sure, physicians
noticed serious malaise and abruptly appearing jaundice, but they set aside
any causal link with broad beans. Lathyrism is a kind of poisoning. That is
why it poses no conceptual problems. But the situation with broad beans
is altogether different, since in the everyday experience of most people
they are a totally harmless food.
The Reali-tyof Favism

Since the end of the last century, Western medicine has officially taken
cognizance of certain individuals whose habitus and way of life seem en-
tirely normal but who suffer from a strange idiosyncrasy: give them a few
broad beans to eat, or let them just cross a field of these papilionaceous
plants in flower, and they fall into a hemolytic crisis. The first modern
mention of this intolerance is in the letters-to-the-editor section of a Lis-
bon magazine. In a letter dated May 3, 1843, one Manuel Pereira de Mira
Franco reported, "as a curiosity," the case of an inhabitant of the town of
Cuba (Portugal) who reacted to each ingestion of green broad beans with
an attack of jaundice. 80 Though validated today by the historian's back-
ward glance, this observation by a Portuguese practitioner escaped the
attention of the learned men of the day and had no influence on the
development of medical knowledge. Under the prevailing mentality of
physicians in the middle of the nineteenth century, at the moment of the
full flowering of their new rationality, it was just a tale. The publication of
the letter itself in a periodical of general interest rather than a specialists'
journal seems to me to suggest that for official n1edical circles of the day
an idiosyncrasy of ~his kind was still something marginal, inexplicable, even

inadmissible. It defied the postulate of the interchangeable biological na-
ture of individuals of the same species and, still worse, appeared to threaten
the postulate of strict determinism in physiological reactions.
Sicilian physicians of the second half of the nineteenth century were the
first to admit the reality of such a personalized hypersensitivity. First in
1856, Antonio Mina La Grua drew attention to the peculiar "endemic
jaundice" of peasants in Castelbuono (a village in the hinterland of Cefalu)
"caused by the fragrance of the flowers of broad beans. " 81 Other practi-
tioners, particularly Di Pietra-Leone from Piazza Armerina, Mule Bertolo
from Villalba, and Rizzo-Matera from Syracuse, confirmed La Grua's ob-
servations, noted the presence of this ailment in several Sicilian communes
and in Calabria, discovered its hereditary nature, and, in polemics with
colleagues who thought that the jaundice was ascribable to malaria, in-
sisted on the etiological role of the broad bean in triggering it. Mule
Bertolo reported that the clinical picture regularly includes hematuria. 82
Taking the region of the Nebrodi Hills to be the cradle of this disease, La
Grua called it malattia vegetalenebrodese.In 1894, at the Eleventh Congress
of International Medicine in Rome, Dr. Montano proposed the name
favismo to designate the set of troubles brought on in subjects who were
particularly sensitive to the ingestion of broad beans or to the inhalation
of the pollen of Viciafaba. 83 Short, expressive, and easily convertible into
other tongues, this term quickly entered the nosological vocabulary of all
lands.
Apparently, among the people of Sicily, there was a dialectal "Word,
zafara) that denoted malarial jaundice and also, particularly, the jaundice
that resulted from crises of favism-this was long before the latter was
recognized by physicians. 84 Doubtless the first impetus for the scientific
study of favism is to be found in Sicilian folklore. Pierro Messina alludes
to popular belief, in his Mediche osservazionisopranon ovvierilevanti infermita
(published in Catania in 1851, and so prior to the monograph of La Grua),
when he says, "Not unknown to us is the miraculous influence on the
formation of bile in some subjects brought about by the fragrance of the
flowers of the broad bean.'' The story of Salvatore Greco, a young peasant
from the province of Syracuse who suffered, like his maternal grandfather,
from favism, attests to the endurance or at least the prescientific revival of
ancient tradition. Without any medical intervention and solely on the basis
of his own experience and that of his grandfather, Greco "long since and
scrupulously abided by the Pythagorean teaching that forbids the con-
sumption of broad beans" (Rizzo-Matera, 1878). 85 This Syracusan peasant's
name is probably a reminder of the Greek origin of his family.
On the threshold of the twentieth century, physicians understood that
the central physiopathological event in favism is hemolysis, the decompo-
sition of the red cells in blood. In his communication to the Congress at
Rome in 1894, Montano states that the hematuria of favism is actually

hemoglobinuria. The miscroscopic and chemical examination of peripheral
blood during an acute reaction to the ingestion of raw broad beans enabled
him to report diminution in the number of red blood cells, discrete leu-
kopenia, and the presence of hemoglobin in solution in the serum. There-
after, all the other symptomatology of favism could be explained as a
secondary consequence or complication of hemolytic crises. Montano also
expresses very clearly the distinction between "the determining cause
(broad bean) and the predisposing cause (particular individual idiosyn-
crasy)," a distinction that saves biological determinism. He stresses the
important role of "determining cause" played by a chemical factor that is
essentially toxic, that is, not microbial; as for the "predisposing cause,"
he stresses the familial factor. In this conceptual context, Montano denies
the influence of climate on the appearance and geographical distribution
of favism. 86 His clinical experience rested on the meticulous examination
of several dozen patients observed in Lavello in the province of Potenza
(Basilicata). Other typical cases of favism were discovered in central Italy
(Girotti, 1899; De Camillis, 1901; and so forth), in Sardinia (Piga, 1899;
Stevani, 1904), and in Greece (Ducas in Lamia, 1895; Skavenzos on the
island of Skopelos, 1895; Cawadias in Leukas, 1897; Vellopoulos in Ata-
lante, 1898; Kontogouris in Leukas, 1900; Tselios in Thebes, 1904; and so
forth). Despite this extension of the cases diagnosed outside the area (Sic-
ily, Calabria) in which the disease was first recognized, it still seemed that
favism was limited to the inhabitants of a relatively narrow zone of the
Mediterranean basin.
Thanks to careful epidemiological studies undertaken before W odd War
I, 87 the clinical picture of favism has been refined, the patients have been
classified according to their specific degree of sensitivity, the familial nature
of their predisposition has been confirmed, and a tendency for the disease
to appear in young persons and preponderantly in males has been re-
ported. These studies have revealed the seriousness of the defect in ques-
tion: according to Fermi's statistical survey, for 72 Sardinian communes,
the morbidity, amazingly, was greater than 5 per 1,000 in 1905, with a
lethality at 8 percent (that is, the number of deaths ·as against the number
of sick persons). From then on, favism stopped being a rarity of interest
only to collectors of medical oddities. It has its due as one of those en-
demic diseases that have real consequences for the public health of certain
regions.
Although they defined favism impeccably as a clinical entity, the physi-
cians of the first half of the twentieth century still did not know the true
nature of its underlying pathogenic processes. Montano, Fern1i, and other
pioneers in the clinical and epidemiological exploration of the disease
thought of it as poisoning by a chemical substance with a molecular struc-
ture and a kind of action that were as yet unknown. However, among the
various proteinic and glucosidic substances isolated from the seed of the
broad bean and tested in appropriate dosages on laboratory animals, none
seemed to present toxic properties. In particular, none displayed hemolytic
powers great enough to account for the crises of favism in terms of direct
action on red cells. And between the second and fourth decades of this
century, several researchers uselessly persisted in wanting to reduce favism
to ordinary poisoning by cyanhydric acid. 88 But the concentration of this
poison in the beans does not reach the critical threshold, its action mech-
anism (enzymatic blocking of tissue respiration) suits poorly the symptoms
of favism, and, last but not least, the "toxicological" hypothesis is of no
help in accounting for the enormous differences in individual sensitivity.
In the description of favism, the triggering cause was well accounted for,
but the intrinsic cause remained totally obscure. By what biochemical
mechanism does the hemolysis start, given that the substances in the beans
themselves are not toxic in the usual sense of the word? What is the actual
role of heredity? How can one explain the possibility that an individual
who is "predisposed" genetically can eat the fatal bean with impunity,
after having reacted violently to a previous ingestion of it? So many ques-
tions, long without answers.
From the time of the definition of the concept of allergy by Clemens
von Pirquet (1904), the existence of a pathological reactivity as individual
and specific as that of favism recalled the mechanism of allergic sensitiza-
tion. In most medical handbooks of the 1930s and 1940s there is a vague
suggestion of the "allergic" nature of the clinical manifestations brought
on by broad beans. However, this analogy met an impasse: instead of
immunity sensitization by previous contact with the substance responsible
for favism, one was faced with an innate idiosyncrasy that was largely
independent of prior chemical conditioning.
The actual point of departure for the discovery of the complex etiology
of favism was the observation by American physicians that absorption of
synthetic antimalarials can produce a hemolytic anemia in some individuals
that can be very serious. Actually, in 1896 Pucci had already noted at
Catanzaro in Calabria (a place not far from the ancient location of the
Pythagorean community at Craton) that certain persons who could not
tolerate broad beans had a similar intolerance for quinine. Once pama-
quine (plasmoquin) was introduced in 1926, disturbances like those of
favism were noticed as side effects of the medication. To be sure, these
were just occasional observations whose pathogenic mechanism no one
knew how to explain. Finally, during World War II, the struggle against
malaria in the American army, in particular the preventive oral administra-
tion of primaquine (like pamaquine, a derivative of 8-aminoquinoline),
brought to light significant individual intolerances in many blacks and in
some whites with family origins in southern Italy and Greece. In 1948, A.
Turchetti made the connection between drug-induced anemia in Ameri-
can blacks and Mediterranean favism. 89 Subsequent research showed that

in both cases there was an innate enzymatic deficiency, a lack in the activity
of glucose-6-phosphate dehydrogenase (G6PD). This is a red blood cell
enzyme that intervenes in the breakdown of glucose, permits the mainte-
nance of reduced glutathione, and fights against the tendency toward
oxidation in red blood cells, thus protecting the hemoglobin in its func-
tion as a gas carrier. A true epidemic of intolerance to primaquine was
observed during the Korean War among blacks in the American expedi-
tionary force; this facilitated systematic research. Decisive hematological
and biochemical research was performed in a prison laboratory in Joliet,
Illinois, with selected volunteers from among the prison's inmates. In 1956,
Paul E. Carson and an excellent multidisciplinary team at the prison suc-
ceeded in proving that deficiency in G6PD was actually responsible for the
hemolytic distress occurring in individuals sensitive to the absorption of
primaquine. 90 It quickly became clear that the same deficiency was at the
origin of other drug-induced, bacteriotoxic, and nutritional idiosyncrasies.
Among the latter, the place of honor belongs to favism, as William A.
Crosby remarked during a trip to Sardinia in 1956, shortly after Carson 's
ground-breaking publication. 91
Nevertheless, there must be some difference between primaquine intol-
erance and favism, since individuals who are sensitive to antimalarials or
other artificially synthesized drugs do not necessarily have a hemolytic
reaction to the ingestion of broad beans. The two diseases are not to be
confused and are not necessarily associated. It seems that the molecular
lesion is not identical in both cases, and in addition, favism demands a
specific seric disorder besides the exogenous cause and the enzymatic defi-
ciency. K. L. Roth and A. M. Frumin have reported that the serum of
persons vulnerable to the crises of favism lacks a protective factor that is
present both in normal individuals and in most carriers of the G6PD
genetic defect. 92 So favism is the result of the intersection of three factors:
(1) introduction in the organism of one or of several specific substances of
vegetal origin; (2) biochemical fragility of erythrocytes due to a hereditary
deficiency of G6PD; and (3) seric deficiency (lack of a protective substance?).
The noxious vegetal principle in favism is still poorly understood. It is
probably a whole family of chemical substances whose basic representative
is vicine, a substance isolated in 1962 by J. Y. Lin and K. M. Ling. 93 This
agent is present in the broad bean, of which there are two rnain varieties,
Viciafaba major and Viciafaba minor. The two varieties are equally harmful.
Beans (Phaseolussp.), although they belong to the same family as broad
beans, are not dangerous for favic individuals. Hemolytic anemias occur-
ring in persons with favism have been described upon ingestion of peas
(Pisum sativum) or stinking bean-trefoil (Anagyrisfoetida). Do these plants
contain the same noxious molecules as the broad bean but in smaller
quantities, or do they act through the presence of a similar chemical sub-
stance? It is possible, even likely, that such an agent occurs in several plants
that do not serve as food. For instance, it is likely that the pollen of
verbena (Verbenahybrida) when inhaled is the cause of a hemolytic anemia
like that of favism. Called Baghdad Spring anemia, this ailment is known
chiefly from the observations of R. Lederer among Iraqi Jews. 94
The most spectacular incidents of jaundice and hematuria are provoked
by the ingestion of raw fresh broad beans. Dried broad beans or flour made
from them can also bring on hemolytic crises, but boiled beans are not at
all dangerous. The harmful substance is destroyed by heat. However, it is
transmitted in the milk of nursing women or goats, which explains the
frequency of hemolytic anemias among newborns in regions like Greece,
where, from antiquity up to the present, a high degree of hereditary en-
zyme deficiency has coexisted with extensive cultivation of broad beans.
The pods, leaves, and pollen of the beans also contain the harmful agent,
but in lesser quantities than the seeds. Simple contact with the pods or the
inhalation of particles emitted by the broad bean's flowers can provoke
relatively mild hemolytic episodes or, more often, migraine headaches.
Sometimes minute quantities are enough to trigger a pathological reaction.
Serious distress has been reported in persons who simply lingered at a
distance of several hundred feet from a field of broad beans in flower. 95
As was said above, since the work of Carson and his collaborators, we
have known that the biochemical basis for individual hypersensitivity is a
deficiency of glucose-6-phosphate dehydrogenase (G6PD). But research
performed in the past twenty years in the domain of molecular pathology
has shown that behind the phenomenon thought of as a simple deficiency
of G6PD, that is, as a purely quantitative lack of a single molecule, there
lies hidden a complex and diverse reality consisting in qualititative anoma-
lies of the enzyme in question. Instead of being an enzymopeny in the
strict sense of the word, the G6PD deficiency actually constitutes a family
of erythrocytic enzymopathies. The first step toward the discovery of this
pathogenic heterogeneity was the observation in 1962 of a normal electro-
phoretic polymorphism in G6PD. In a healthy population there exist two
fully functional variants of this enzyme (types A+ and B +). 96 The second
and decisive step was bringing to light abnormal enzymatic molecules. 97 It
became necessary to admit the biochemical and genetic diversity in the
G6PD deficiency. So this "inborn error of metabolism" consists in differ-
ent mutations at the level of the gene that specifies G6PD. The abnormal
variants that result are probably to be distinguished by different substitu-
tions located on one amino acid. Their phenotypic expression consists in
qualitative and quantitative modifications of the enzyme that vary accord-
ing to the nature of the particular mutation. The intolerance in blacks for
primaquine is very likely due to a mutation that hits the allele of type A+.
Accordingly this variant is designated by the symbol A-. In whites with
favism, especially in Italy and Greece, the common normal allele of type
B + is replaced by the so-called Mediterranean variant (B-). These two

abnormal variants are frequent, as is a third that was discovered in some
Chinese suffering from drug-induced hemolysis and sometimes also favism
(the so-called Canton variant). 98 However, these three main pathological
forms of the G6PD deficiency hardly ~xhaust the polymorphism of the
disease. Worldwide, about 100 abnormal forms of G6PD have been iden-
tified, but it appears that only the 3 main variants have a major statistical
incidence of morbidity. 99
The study of the hereditary transmission of favism entered a new phase
when consideration of its biochemical basis (latent favism recognizable
through tests of enzymatic deficiency) replaced that of its clinical pheno-
type (overt favism). Research carried out mainly in Sardinia by U. E.
Carcassi, P. Larizza, and other Mediterranean experts confirmed for favism
what B. Childs and his associates had already found for intolerance to
primaquine: the locus determining the synthesis of G6PD is part of the X
chromosome (called the Gd locus). It is situated near the loci responsible
for hemophilia and col or blindness. Favism is a defect whose heredity is
sex-linked. It can be fully expressed in males and is partially expressed in
females with the defect, who play the role of carriers in its genetic
transmission. 100
G6PD deficiency is a very widespread genetic trait that most often re-
mains latent in clinical terms. At the moment there are more than 100
million individuals with this defect, and yet true cases of favism are rela-
tively rare. The enzymatic anomaly, especially the B- variant, is an indis-
pensable but not a sufficient condition to make a person hypersensitive to
broad beans. As was said above, a second anomaly must intersect with the
first. It is probably expressed in the serum, but its true nature remains
unclear. In any case, we must assume that it, too, is determined by a
specific, probably autosomal genetic mechanism. From this standpoint,
favism is still an enigmatic idiosyncrasy. 101
The typical clinical profile of favism is acute hemoglobinuric jaundice.
In the hours following alimentary contact with raw broad beans, gastro-
enteritis with acute abdominal pains is followed immediately by hemoglo-
binuria-the urine can become black-severe anemia, bouts of fever, and
moderate jaundice. These symptoms disappear after a week, but the crisis
leaves behind an anemia that can last for several months. The milder
reaction is expressed as general discomfort, with headache and nausea. In
about one case in twelve, the hemoglobinuria can be so violent that it
results in death from anoxemia or acute renal failure. 102 Modern Greek
pediatricians have reported that G6PD deficiency is an important cause
of severe forms of neonatal jaundice, which is relatively frequent in Athens
and various regions in Greece. At times the children can be saved only
by exchange transfusion. 103 To be sure, this jaundice is not triggered by
broad beans. In children the anemia that follows the crises of favism can
produce cranial lesions that resemble porous hyperostosis from thalas-

semia. 104 "
Historically, the main haven of favism is the territory of Magna Graecia

(southern Italy, including Sicily), Sardinia, certain parts of what is now
Greece, especially several islands, Anatolia, Corsica, and the Mediterranean
coast of Africa (Egypt, Algeria, etc.) . 105 If we consider only the B- variant
of the G6PD deficiency, the genetic defect is practically restricted to the
Mediterranean basin. For my purposes, it is not important to note the
geographical distribution of the A- variant (widespread in Africa), the
Canton variant (Far East), or the countless other mutations of G6PD
scattered throughout the world. There is no longer any doubt that their
origin is polyphyletic, without links to the birth and spread of Mediterra-
nean favism.
Before the modern mixing of populations and the massive migrations of
southern Europeans, the frequency of favism in northern Italy was mini-
mal, as it was in other countries of northern Europe. The ancient home-
lands of the Etruscans and Latins have only very recently been tainted by
the disease. By contrast, the B- variant of G6PD deficiency is extremely
common in Sicily, the fatherland of Empedocles, in Sardinia, and in the
parts of Calabria where the Greek colonies once stood and Pythagoras held
forth. There, up to one-third of the autochthonous males have the favic
defect. 106 In modern Greece, hereditary G6PD deficiency is present in all
the oldest settlements, but its distribution is not uniform. Surveys made
in Greece during the 1960s provide a sense of the magnitude of its presence.
The frequency of the defect ranges from 2 percent to 12 percent depending
upon the region. The areas in which it surpasses 10 percent are the plains
of the Peloponnesus, Epirus, and Thessaly. By contrast, favism is relatively
rare in Asia Minor and in mountainous areas. 107 S. A. Doxiadis has drawn
attention to the prominence of the deficiency in newborns hospitalized in
Athens. 108 There is a fine study by C. A. Kattamis based on 506 cases of
favism in children who were treated from 1956 to 1966 at the Pediatric
Clinic in Athens. It shows the extent of the disease's pathological manifes-
tations and the prevalence of the Mediterranean variant (B-). 109 A study
undertaken by this same investigator among the people of Rhodes con-
firms reports by Italian physicians who had noted the prevalence of favism
in the Dodecanese in 1930.110 The frequency of G6PD deficiency among
Rhodian men is higher than 30 ·percent. It is also very high on the island
of Cyprus. Such concentrations of favism on islands, likewise observed in
Sardinia, Sicily, and Corsica, could be the result of a certain genetic segre-
gation among these populations, of a greater degree of consanguinity in
island peoples than in those of the mainland.
How can one explain the maintenance, not to say the progress, within
a given population of hereditary traits like this, which at first sight offer
grave biological disadvantages? Why are they not eliminated by natural
232 DISEASES IN THE ANCIENT GREEI( WORLD
selection? The geographic distribution of the genetic trait expressed by

G6PD deficiency largely corresponds to that of the ancient malarial en-
demic, exactly like the distribution of sickle cell anemia or the various
thalassemias. This is sometimes true even for distribution within a district.
So, for instance, among boys in the region of Arta (Epirus), the frequency
of the favic enzymopathy varies according to the exposure of the region to
malaria in the past: much more common among the inhabitants of the
plain than those of the mountains. 111Accordingly, it seems wise to apply
to the genetics of this defect the hypothesis of balanced polymorphism
invented by J.B.S. Haldane and developed by A. C. Allison in connection
with sickle cell anemia. According to this hypothesis, the maintenance at
a high rate of this erythrocytic enzymatic deficiency would be ascribable
to the selective advantage it provides in zones of malarial endemic. 112 The
research of L. Luzzatto, U. Bienzle, and others shows that in men (of
necessity hemizygous for traits linked to the X chromosome), the G6PD
deficiency does not at all increase resistance to malaria, but that heterozy-
gous women (especially those of type B +A - ) do benefit from an evident
protection when exposed to malarial infestation. 113A mathematical model
can be constructed showing how such an advantage is enough to balance
in demographic terms the deleterious effects of favism. Nevertheless, the
correlation between the prevalence of this enzymopathy and the intensity
of the malarial endemic in antiquity is not always satisfactory. Among
African Jews, for instance, it does not correspond to the requirements of
the theory. 114The hypothesis that introduces malaria into the history of
favism as yet offers only a partial solution to the problem it is intended to
resolve.
Nor should we forget that the heterogeneity of the G6PD deficiency
and its interaction with hereditary hemoglobinopathies infinitely compli-
cates the history of favism. 115Even in Greece, where the B- variant is
largely dominant and where the "Mediterranean phenotype" of the idio-
syncrasy appears homogeneous, current research reveals unexpected com-
plexities.116 Numerous minor variants of the G6PD deficiency have just
been described. They are named after the place in which they were found:
Athens, Attica, Corinth, Orchomenos, Levadia, Thessaly, Karditsa, and
so forth. The genesis of these variants should not be confused with that of
the "Mediterranean" mutation in the strict sense of the term. The latter
especially interests the medical historian of the Greek world: the frequency
of the B - type (as against the other alleles responsible for the erythrocytic
enzymopathy in whites) and also its particular geographical distribution
force us to acknowledge at once its antiquity and its place of origin. Med-
iterranean favism goes back at least to the archaic period of Greek
civilization.
Has modern science, then, unexpectedly discovered an empirical justi-
fication for the ancient prohibition against eating broad beans? At the
moment I am posjng this question rhetorically I will pose it again in a

more appropriate form after examining the work of modern exegetes on
this strange piece of ancient history.
Modern Opinions on the Pythagorean ~ohibition

From the rebirth of classical studies in the fifteenth century until the
nineteenth century, countless humanists glossed the tabu against eating
broad beans. Taking up the explanations of ancient authors, they defended
those that suited their predilections without really strengthening or em-
bellishing them with new arguments. 117 Since physicians firmly maintained
that eating these beans was harmless, the prevailing opinion was that the
underlying reason for abstinence from them in the Pythagorean tradition
was mystical or symbolic. For N eoplatonists, gnostics, and followers of
similar philosophical systems, nothing was more natural than reference to
cosmogonic n1yths, to the passage of souls into the beans, and to the
mystical correspondence between this plant and the human body. On the
other hand, for authors of an Aristotelian persuasion and for partisans of
modern rationalism, the most comfortable approach was that of Aristox-
enus. One simply denied the dietetic nature of the prohibition, preferring
to assert its sexual or political significance. Some commentators gladly
thought it originated in the resemblance of the broad bean to a testicle
and claimed that it was a symbol of the moral law limiting sexual relations.
The political explanation, given prestige by the cautious approval of Plu-
tarch, stressed the use of broad beans in the election of magistrates in
Greek city-states. For instance, after citing the expression '' Abstain from
broad beans!" Kurt Sprengel, the famous physician, botanist, and founder
of pragmatic medical historiography, says, ''Probably Pythagoras wished
by these words to divert his disciples from seeking political office, so that
they would thereby remain more attached to his own order. " 118 A similar
view on the political significance of the Pythagorean rule was given cre-
dence among German philologists thanks to a monograph by A. B.
Krische. 119
However, belief in the real harmfulness of broad beans does not seem
to have been lacking in Greek folklore. During a study of popular represen-
tations of the causes of infantile mortality, Richard and Eva Blum came
across several Greek peasants who were convinced that broad beans are
dangerous to infants and children, causing sickness or death. According to
the Blums, ''This belief is widely held and its historical importance is
attested to by the doctrine of Pythagoras that proscribed the eating of
beans. " 120 Is this really a survival of the Pythagorean teaching in the unlet-
tered world of the Greek countryside? I do not believe so. Given the
genetic conditions in some Greek communities, the danger was only too
real. The harm in broad beans was probably discovered many times over
by mothers attentive to the food and health of their children. Knowledge

acquired this way gets transmitted orally from one generation to the next
without help from any philosophical tradition. Balkan folk medicine, which
has only recently begun to disappear, rejoiced in a mixture of empiricism
and magic that in many ways resembled the Presocratic mentality; as such
it was rich in observations that did not conform to the imperatives of
scientific medicine. So nothing prevented Greek peasants from noticing
that a food that was usually well tolerated could also be a poison. But
physicians were unwilling to see in that anything other than superstition,
since they were blinded by a concept of determinism that was at once very
effective and overly simple.
In an encyclopedia article written at the end of the nineteenth century,
F. Lenormant summarized current historical understanding of the prohi-
bition, which he attributed to the tradition of the Orphic mysteries. For
him, the key to the problem was provided by Aulus Gellius: the word
kuamos originally refers to the genitals of animals and not to broad beans.
Abstinence from such beans can only have a symbolic meaning. Lenor-
mant insists on the analogy of this ritual and moral precept to the prohi-
bition against eating pomegranates for Athenian women participating in
the Thesmophoria and the one against eating celery for priests during the
Corybantic mysteries. The analogy is instructive, since these customs have
symbolic links with bloody mythological events. 121 But Charles Darem-
berg, the physician who was chief editor of the Dictionnaire des antiquites
grecqueset romaines)in which Lenormant's article appeared, could not keep
himself from adding to it the following remark, which is imbued with the
positivist's faith: "Surely at the heart of all these legends there is a simple
rule of hygiene, like the rule about the abstinence from the meat of the
pig among the Hebrews, on which ancient superstition embroidered a
wide variety of themes.'' A rational interpretation of another kind was
proposed in 1904 by E. Bourquelot of the Society of Biology in Paris, then
a shrine to the experimental method and metaphysical agnosticism of men
. like Claude Bernard and Charles Robin. First Bourquelot states that the
Pythagoreans abstained from eating animal flesh. According to him, that
was because of the doctrine of the transmigration of souls. But they also
abstained from eating broad beans, which were a food much enjoyed by
Mediterranean peoples. Clement of Alexandria's explanation is inade-
quate: he attributed to broad beans the property of making women sterile,
but then Pythagoras had no grounds for denying them to men. So Bour-
quelot cites the passage in the play of Lucian in which Pythagoras relates
that cooked broad beans, when exposed to the light of the moon for a
certain number of nights, give off blood:
This explanation [writes Bourquelot] does not appear as ridiculous today as it did
to Lucian. Apparently people must have observed the spontaneous development
on cooked broad beans of one of those chromogenic microbes whose cultures so
resemble bloodstains that one can understand how people in antiquity n1ight have
made this mistake. Th.e fact struck Pythagoras, who saw it as proof of the animal
nature of the broad bean.
Faithful to his experimental method, Bourquelot studied these phenom-

ena in the laboratory:
For that purpose, I did not think it necessary to expose the cooked beans to
moonlight; but I did innoculate them with a well-known chromogenic bacillus,
Kiel's bacillus. So I can report that the conditions which permit one to obtain
abundant production of red spots are exactly those Pythagoras describes. The raw
broad beans must be shelled and cooked; otherwise, it is not possible to succeed,
or one can only be partially successful. The best technique is, after having mois-
tened and then sterilized the beans at no O , to put them in the oven at 33° for 24
hours, and then leave the1n at room temperature (18-20 °). At the end of eight
days, the beans will be covered with bloody spots.
Bourquelot concludes that "by an altogether natural association of ideas,

the bloody broad beans of Pythagoras recall the bloody hosts [of the
Catholic Eucharist] that were so disturbing to look at 1,800 or 2,000 years
later. " 122
The positivism of Daremberg and Bourquelot had no audience, at least
during the first half of the twentieth century. All research in that period is
marked by the work of Sir James George Frazer on tabus and their links
with totemism, 123 and by the work of Erwin Rohde on the concept of the
soul and the cults related thereto. 124 In the light of comparative ethnolog-
ical studies carried out by British anthropologists (E. B. Tylor, J. G. Fra-
zer, and others), the prohibition against touching broad beans seemed like
a classic food tabu based on primitive totemic representations. The code
of superstitious maxims attributed to the sage Pythagoras goes back, says
Frazer, to a much more distant past and reflects in symbolic form the
assumptions of a lost and forgotten sociocultural context. 125 This set of
notions from cross-cultural anthropology soon received the support of
psychoanalysis. First one digs up the roots of rites and myths, and then
one understands their special unconscious logic. Masterfully exploited by
Salomon Reinach, 126 elaborated in a series of monographs, 127 and cited in
handbooks on the history of medicine or of science in general, 128 such
psychosocial accounts reach their height in penetrating studies by the phi-
lologist Armand Delatte 129 and the philosopher Marcel Detienne. 130 Rein-
ach recalls Daremberg's remark and even takes the trouble to cite it
completely, the better to denounce his purpose and insist on the futility
of utilitarian explanations of tabus: "This error is instructive, since it is
typical of many others like it that are not only anachronistic but also put
the cart before the horse; they see superstition as a corrupt form of science,
when science is its distant descendant or even its posthumous child, if
superstition can ever be said to die. " 131 It is an arrogant, unacceptable
point of view. A goodly number of modern "superstitions" are indeed
corruptions of the science of an older age. Still, it is true that magical

thought, which is the subject under discussion here, is not the corruption
of a prior science. In that, Reinach is correct, but he imputes to Daremberg
an idea that is not his, and in doing so he does not even advance his own
point of view: the chronological priority of magical thought does not mean
that it cannot contain empirical, utilitarian knowledge. In a word, super-
stition is not necessarily all "false," any more than science is absolutely
and completely "true."
Pythagoras's fellow citizens planted broad beans, and Reinach concludes
that "this plant was not considered impure or unhealthy, but sacred in-
stead." The tabu necessitated that it not be killed by being eaten or trod
upon, since it was considered one of man's totemic ancestors. That is the
underlying sense of the Orphic verse that likens the act of eating broad
beans to gnawing one's parents' heads: "The ancients did not understand
this prohibition and devised extravagant ways to account for it, and the
more extravagant they were the more they were hygienic and utilitar-
ian. " 132 Jean Larguier de Bancels also believes that Pythagoras and his
followers ''adopted or revived by means of a symbolic prohibition an old
ritual whose original sense had long since been lost." It was a tabu of
Inda-European origin. Even flatulence, which the ancient authors men-
tion in their discussions of eating broad beans, can have its symbolic as-
pect. Larguier de Bancels translates the view of Diogenes Laertius a little
differently from the way I have above: it was necessary to abstain from
broad beans ''because, being windy, they participate in the nature of the
soul." Broad beans are animate, and so likened to animals, since they
possess breath. He also draws attention to the Inda-European theory of
the psychic pneuma, which implies close conceptual relations between the
soul, the air breathed, and flatulence. 133 Besides, according to the psycho-
analyst E. Jones, breath is only a symbolic displacement of flatulence, and,
more important, the notion of the soul in the last analysis derives from
the internal sensations that an infant has of gas trapped in the belly . 134
Armand Delatte notes that ancient observations about broad beans al-
ternate between awe and dread, which are the two poles of a single sphere
of confused impressions that the Greek language represents by the term
hier6s'sacred.' This word was applied to "anything loaded with a power-
ful, supernatural fluid, alternately religious and magical, whose touch can
be either beneficial or harmful. " 135 So the attitude toward the broad bean
is "ambivalent," as is to be expected with a tabu. Delatte classifies the
ancient texts on this subject and arranges them in such a way that "once
they are examined in the light of the customs and beliefs called totemic"
they constitute a unique system of explanation for the superstition:
If it were possible to express the latent and confused reasoning that guided the
primitive mentality at God knows how distant an era, here is how it could be
formulated: leguminous plants and especially the broad bean, which is the most
widespread and often used of them in human food, produce, during digestion,
flatulence, which disrt.Irbs certain mental functions (the clear perception of dream
visions). So broad beans contain winds that have an unfavorable effect; however,
the souls and, in particular, the souls of the dead, which are usually considered
harmful, are also winds, so broad beans contain the souls of the dead. By what
mysterious route? That is what they struggled to explain, apparently in a less
distant era, first by establishing a relation between the theories about the return of
souls to the earth for reincarnation and the observation of the special shape of the
plant's stem; and then-and this is the contribution of those who enjoy finding
original causes, which is already a habit of mind in "primitive" peoples-by in-
venting the cosmogonic myth about the common origin of broad beans and men.
A single practical conclusion flows naturally from this: eating broad beans is crim-
inal cannibalism; and just touching them is a defilement and a danger, like touch-
ing a cadaver. 136
Here is another long citation from a work by Marcel Detienne that is so

lively I prefer not to paraphrase:
For strict vegetarians, every blood sacrifice is a murder and ultimately an act of
cannibalism for which their horror is expressed by way of the broad bean. In fact,
this legume is at the antipodes of spices, the marvellous food of gods and the
Golden Age. Thanks to a stalk without nodes and by virtue of affinities with the
rotten, the broad bean establishes the same direct communication with the world
of the dead as spices establish with the world of the gods to which they belong
through their solar quality and dessicated nature. But in the Pythagorean system
of thought, the broad bean is still more. It is a being of flesh and blood, the double
of the man at whose side it grows and from whose rotten compost they both feed.
As a result, say the Pythagoreans, it is the same crime to eat a broad bean as to
gnaw one's parents' heads. Proof of it is produced by a set of experiments known
to Pythagorean tradition. A broad bean is placed, for its mysterious cooking, in a
pot or closed container that is then hidden in manure or buried in the ground.
After a more or less lengthy period of gestation, the bean transforms itself either
into female genitals with a child's scarcely formed head attached to them or into a
human head whose features are already recognizable. In these experiments, the
pot is a womb entrusted with the task of revealing the broad bean's true nature.
But it can already be discovered by taking a half-eaten or slightly squashed broad
bean and placing it in the sun for a moment or two. Immediately a smell arises
which is said to be either the smell of sperm or that of blood shed in a murder.
The Pythagoreans are explicit: eating broad beans is feeding on human flesh, is
devouring the most marked type of meat . . . The broad bean is indeed the most
marked source of generation in the plant world, to the extent that it appears as a
mixture of blood and genitals in the fantasies of the Pythagoreans. But the prohi-
bition against tasting it only restates in more urgent terms the hackneyed tabu
against eating meat or spilling the blood of a living being. 137
Detienne also notes that the egg, the heart, and the broad bean occur
together, as for the disciples of Orphism or Pythagoreanism,
in one and the same list of prohibitions that a cultic regulation from Smyrna dated
to the second century A.D. enumerates for the initiates ofDionysos Bromios: don't
approach the altars wearing black clothing, don't strike the victims that cannot be
sacrificed, don't serve eggs in the banquets honoring Dionysos or burn the heart
(of the victim) on the altars, and abstain from mint, which accompanies the
DISEASES IN THE ANCIENT GREEI{ WORLD
accursed race of broad beans. But the horror provoked by these leguminous plants
in this Dionysiac context receives a novel justification that the regulation suggests
be told to the mysteries' initiates: broad beans were born from Titans, the mur-
derers of Dionysos.
The originality of this text, according to Detienne,

does not lie in the way it combines Orphic precepts with Pythagorean tabus while
giving them a certain Dionysiac flavor. It lies instead in the way it overdetermines
parallel motifs of different origin, for example, by inventing a Titanic origin for
broad beans. These vegetables are not only the nocturnal, bloody doubles of the
human plant, whose consumption is tantamount to cannibalism; broad beans are
themselves born from the primordial beings who were promoted to the rank of
ancestors of carnivorous humanity by their anthropophagic behavior. 138
Clearly, Detienne has learned his lesson well from Georges Dumezil and
Claude Levi-Strauss. It would also have been interesting to look at the role
of the raw and the cooked in the mythemes about broad beans from a
structuralist point of view. I should also point out a central notion in his
discussion: '' overdetermining parallel motifs.'' Characteristic of the logic
of dreams, overdetermination slips in everywhere in magical thought; I
will return to it later.
Depth psychology, the sociological approach, and structuralism all in
turn offer keys for the constantly renewed interpretation of ancient texts.
Their insights are dazzling; the conclusions made by specialists in anthro-
pological analysis of myths dominate the field. Alongside them are only a
few rare and timid attempts to complement, never replace, the magico-
cultural and structuralist interpretations of the Pythagorean tabu by a
dietetic justification of it in medical terms. To Joseph Schumacher, for
instance, the prohibition against eating broad beans has its "natural expla-
nation" in the application of the general Pythagorean rule whereby one
should eat in conformity with the nature of the human body, and avoid
foods that swell the belly and cause concern. 139
Once favism became visible to medical eyes as a real disease, the link
between it and Pythagorean teaching was made, at first in a marginal and
superficial way in specialized medical studies that were totally unknown to
historians of the ancient world. Had the doctors finally found the true
sense of this "simple rule of hygiene" that Daremberg suspected lay be-
hind the legend? To my knowledge, the first to take up the question from
this point of view, with appropriate knowledge both of historical sources
and the medical discoveries, was the Greek physician Solon Veras. In an
account given in May 1939 to the Hellenic Society of Pediatrics in Athens,
he reviewed the principal evidence on the prohibition against broad beans
in antiquity, stressed the magical or simply dietetic nature of the ancient
justifications of it, and came to the conclusion ''that no proof existed in
accounts from the past of a real knowledge of favism by Greek writers. " 140
There is no doubt that Veras is correct: nothing in the historical documen-
ration at our disposal allows us to infer any clinical knowledge of incidents

of favism. But Veras does not take account of the fact that all that docu-
mentation is second-hand and informs us only indirectly and very poorly
of the actual lore of the first Pythagoreans.
The debate began anew once the intrinsic cause of favism had been
unearthed. The tabu against broad beans could, it was suggested, be ex-
plained on the simple supposition that Pythagoras himself or one of his
associates had G6PD deficiency and was aware from direct experience of
the crises of favism. 141 This hypothesis was constructed on the basis of very
little evidence, without an in-depth study of the problems it poses. In two
recent publications, the medical explanation of the Pythagorean tabu is
vigorously criticized with arguments that go beyond those of Solon Veras.
Moreover, the two are unaware of each other and do not know or cite the
work of Veras. The first of them, M. Enrique Laval, a philologist from
Chile, claims to have heard from a doctor friend that Pythagoras knew of
favism. Wishing to be sure about it, he examined both modern medical
publications as well as Greek and Roman texts and came to the conclusion
that '' neither Pythagoras nor any other philosopher or physician believed
that eating broad beans or inhaling the fragrance of their flowers could
damage one's health; the prohibition against eating them is religious, and
probably linked to the doctrine of the transmigration of souls. " 142 The
second authority, Constantin Ballas, a physician from Athens, discusses
the subject with barely concealed contempt for his overly rational col-
leagues. He dismisses them in a footnote: "In an attempt to find 'hidden
medical knowledge' almost everywhere, some contemporary historians of
medicine have claimed that most of the Pythagorean akousmata are plain
hygienic rules; this quite naive interpretation doesn't deserve any serious
consideration. '' 143
This is an easy way out of the problem, too easy in fact. We should
avoid the false reasoning that has been mechanically repeated since Salo-
mon Reinach and that tacitly presupposes that a genuinely mythological
motivation for a custom will not consist in something hygienic or generally
utilitarian. The mutual exclusivity of two kinds of determinism, one myth-
ological and the other utilitarian, is not at all proven; it is even unlikely to
be true. The overdetermination in myth is not restricted to the sociocul-
tural and psychological register. It is indeed possible that apparently con-
tradictory explanations are actually complementary and work on different
levels. This was known in antiquity, and people did not deprive themselves
of combined justifications.
In concluding this review of modern opinions, I should stress the suc-
cess of the anthropological analysis of the broad bean prohibition. It does
indeed resonate with totemism. The proof of its magical/religious nature
is provided by an examination of the overall role of the sumbola and akous-
mata of the Pythagoreans. However, analytic procedures that have been
DISEASES IN THE ANCIENT GREE!( WORLD
useful in decoding the tabu on the mythological level have also taught us
that in this domain the discovery of a single reason for something does not
necessarily exclude others.
ConcludingQuestions
The reality of favism on the one hand and the legendary Pythagorean
teachings about broad beans on the other pose pressing questions for a
methodology that seeks to integrate medical history with epistemology
and the history of daily life. In the hope that the clear formulation of such
questions is a step toward answering them, I would pose them as follows:
1. Did the favic genetic defect exist in the ancient population of the
Mediterranean?
2. If the answer to question 1 is yes, how did it come about that the
disease was not observed by physicians prior to the nineteenth century?
3. Was the prohibition against eating broad beans dictated or at least
prompted by the actual observation of hemolytic incidents in Egypt, the
Greek colonies in Italy, or in Greece itself? In other words, did the Egyp-
tian priests, Pythagoras, and Empedocles know about the favic idiosyn-
crasy even in a vague way?
4. If the answer to question 3 is yes, why is the tradition silent with
regard to such knowledge?
I do not hesitate to answer yes to the first of these questions. The
current geographical distribution of the favic defect cannot be well ex-
plained according to the rules of population genetics unless we admit the
ancient existence of this mutation in the Mediterranean. The rates of favic
defect seem to evolve relatively slowly, at least in the absence of significant
migrations. For obvious reasons the eating of broad beans should lead to
the selective elimination of favism. However, field observation shows that
the persistence of a high frequency of the enzymopathies linked to G6PD
deficiency is compatible with the presence of broad beans in the standard
diet. In Iran, there is almost no G6PD deficiency among the Zoroastrians,
who represent the oldest layer of the current population, while the defi-
ciency rate among Moslems is greater than 7 percent. True favism is rare
there. It is worth noting that the broad bean probably originated in that
part of the world or at least that its consumption there goes back to the
remote past.
The maintenance of a high rate of the favic defect appears to be linked
to the falciparum type of malarial endemic; it is then a consequence of
balanced polymorphism, as was explained above. Pure favism does not
come from Asia. Its cradle is North Africa or Magna Graecia, and the zone
of malignant tertian fever passes through them. The malarial endemic in
Sicilr is attested from the start of the historic period. Even a legendary tale
about Empedocles speaks of it. 144 The city of Croton, center of the first
Pythagorean community, was known to be particularly salubrious, 145 but
that does not mean that malaria was not rampant in the vicinity . 146
In sum, these are my reasons for believing that the favic defect was
present at the dawn of classical civilization precisely where the prohibition
against eating broad beans was first articulated. Favic incidents must have
taken place in antiquity in that region, since their triggering cause was also
present there. For the ancient Greeks, as for the Romans after them, broad
beans were a basic food cultivated everywhere as far back as anyone knew: 147
"Phainias in his treatise On Plants says, 'When they are fresh, we put bitter
vetch, broad beans, and chickpeas in appetizers; when they are dry, we
serve them boiled or almost always roasted.' " 148 Eaten fresh and raw as
snacks, broad beans were also a main dish when served boiled. 149 Some
storing and cooking techniques significantly modify their toxicity for per-
sons who suffer from favism. There is a large number of recipes for prepar-
ing the beans that reduce the danger. Such recipes first appeared in the
eastern Mediterranean and in the course of time became widespread. 150
As for the second question, I have outlined an answer to it in the first
part of this chapter. There were epistemological obstacles that hindered
the free growth of knowledge. Science sharpens perceptions in one direc-
tion only by blinding it in others. It is not an accident that the discovery
of favism in the nineteenth century was effected not by the medical elite
but by provincial practitioners who were open to folklore and relatively
free from the dominant medical ideology. It is also no accident that current
research on this enzymopathy is being carried out by veteran scientists in
highly specialized institutions. Science sets aside certain facts only to take
them up later on at another level of development. They are the facts that
at a given point in time trouble the accepted conceptual framework and
run counter to what since Thomas Kuhn has been called the scientific
paradigm. They are conjured away, made invisible by an intellectual selec-
tion process that takes place below the level of the proper processes of
critical elimination and conscious, logical scrutiny.
In the past physicians saw acute jaundice, the appearance of blood in
the urine, and other symptoms of favism, but they refused to see them as
causally linked with broad beans. In this instance they were blinded by an
absolute demand for deterministic coherence. In the eyes of a Greek phy-
sician or naturalist of the fifth and fourth centuries B.C., the harmlessness
of broad beans was proven by daily, irrefutable experience: a person who
eats broad beans does not fall ill from them; at worst he or she gets bloated
and sleeps poorly. If someone does fall ill after a meal of broad beans, that
proves nothing. Science had unmasked the fallacy in the rule post hoe ergo
propterhoe. In this way it could always preserve the coherence of its causal
chains. So in terms of Hippocratic nutritional science, it was established
that everyone can eat broad beans with impunity, at least from the point
of view of physical health. To be sure, the notion of an individual dispo-

sition, a particular phusis)was already well developed, but with a significant
limitation. In the doctrinal perspective of humoral pathology, a disposi-
tion is the result of a special mixture of the components of the body, an
apparent individual state that is visible to the clinical eye. In a disease like
consumption, for example, the disposition is revealed by external tokens,
by a habitus phthisicus. There is nothing of the kind for favism: the
individual intrinsic state is perfectly hidden, given the limits of ancient
investigative technique.
According to the recent opinion of Salomon Katz, "Deeper knowledge
about susceptibility to favism should have developed, and in such a way as
to reduce its deleterious effects to a minimum. And yet it turns out that
favism is still a widespread disease throughout the [Mediterranean] re-
gion. " 151 If the broad bean, says Katz, had only negative effects, humans
would have ceased to use it. Since "evolution has provided no implicit or
explicit knowledge of its toxicity," the blindness should be counterbal-
anced by a biological advantage. To explain what he represents as an
"evolutionary paradox," Katz formulates the hypothesis that in those
individuals not sensitive to its dangerous hemolytic effects, the broad bean
had antimalarial properties similar to but less potent than those of syn-
thetic antimalarials:
The broad bean has been eaten since Neolithic times and it is clear that neither
tabus nor treatment techniques have been enough to diminish the number of cases
of favism. Although the broad bean is an important agricultural product in the
Mediterranean basin, it remains difficult to explain its continuous use given the
rates of morbidity and the high mortality that it produces in individuals with G6PD
deficiency. Accordingly it seems likely ... that the fact that its use confers a
cumulative resistance to malaria constitutes a supplementary selective advantage.
More specifically, several kinds of active components in broad beans that have
been shown to be responsible for crises of favism in subjects deficient in G6PD
actually increase the sensitivity of red blood cells to oxydants in normal individuals,
and they do so without toxic side effects. 152
Malaria, then, promotes the maintenance of the favic defect, while eating
broad beans combats malaria. At least, that is the. hypothesis, which is
indeed attractive but as yet lacking in experimental proof.
As for the hereditary aspect of favism, E. Giles has shown that in Indo-
Europeans transmission of any trait linked to the X chromosome was made
invisible by exogamy and the patrilineal, patrilocal system of kinship. It
was very hard to perceive the relation between a disease transn1itted by
mothers and a line of descent based on fathers. 153
The third question is the heart of the matter. A secure answer seems to
me impossible, but I tend toward a nuanced yes. In the absence of proof,
we have to rely on clues. The Pythagoreans were extren1ely attentive to
,,
the effects of food on the state of the organism. The tradition is unanimous
on the subject, anait attributes to Pythagoras and his immediate followers
extensive research in this domain. 154 The master from Samos is said to have
been a vegetarian himself and the inventor of a meat diet for athletes: "Of
all forms of medicine, [the Pythagoreans] understood dietetics above all.
In that they sho,ved themselves to be very fastidious. They were the first
to try to understand the signs of the relation between exercise, food, and
rest. They were among the first to provide explanations and advice about
the preparation of food. " 155 We have seen the importance of mytholog-
ical links between broad beans, blood, and death. The most distinctive
symptom of favism is hemoglobinuria, which at times can cause sudden,
spectacular death. Moreover, Pythagoras and Empedocles did not yet
have an epistemological obstacle, a fundamental prejudice, to make fav-
ism "invisible." Here I touch upon the fourth and final question. The
essential element of an answer to it occurs in a statement by Plutarch:
"Philosophy was rife with visions, fables, superstitions when he [So-
crates] received it from Pythagoras and Empedocles; it was possessed,
and he trained it to get in step with reality and pursue truth by means of
sober reason. " 156
The magical thought of the Presocratic thinkers is known to us only
through the filter of their rationalizing successors. We know nothing of
the real rootedness of Pythagoras's teaching in his personal experience of
life. We know only a few bits of his teachings in any case, and they are
only what writers of the fourth century B.C. chose to transmit. As exe-
getes, those writers were poorly informed (considering the esoteric nature
of the philosophical sects) and also, they were firmly committed to retriev-
ing the great figures from the past and giving them a new, corrected image,
in tune with a new scientific morality. When I cited the evidence of
Alexander Polyhistor above, I observed that the prohibition against eating
broad beans had to do with a sacred dietetics that was linked to the idea
of ritual purity, not health. Although nowadays these conceptual domains
are clearly distinguished, was such a distinction really valid in the archaic
period of Greek civilization? After taking a close look at the Pythagorean
prohibitions reported by Alexander, I would say that although all are tabus
and as such religious, a good number of them can also be justified on
hygienic grounds.
In 1975, in my seminar at the Ecole Pratique des Hautes Etudes, I offered
an explanatory schema that divided this historical process into three phases:
r. Observation of crises of favism and their imprecise interpretation by
a largely "magical" mentality (Pythagoras, Empedocles).
2. Abandonment of this belief by the rationalizing mentality of thinkers
like Hippocrates or Aristotle. The harmfulness of broad beans is such a
complex phenomenon that medical experience on an elementary level does
not confirm it; it was therefore relegated to the domain of superstition

that people tried to demythologize (Aristoxenus, Cicero, Plutarch, Aulus
Gellius); for millennia, then, it was "invisible" to the learned.
3. Modern discovery of the particular modalities of favism and partial
rehabilitation of the ancient tabu. 157
Chapter Ten
POROTIC HYPEROSTOSIS,
HEREDITARY ANEMIAS,
AND MALARIA
It may be surprising to find a single chapter embracing such disparate

pathological states as a peculiar morphological modification of ancient
skulls, a group of hereditary blood abnormalities, and a fever caused by
parasites. 1 But the combination is not a random one; any single part of
this medico-historical triptych illuminates the other two. I also wish to
demonstrate herein the usefulness of interdisciplinary research on prob-
lems of biological history: in writing this chapter I have had recourse to
discoveries in paleopathology and modern clinical medicine, recent
achievements in genetics and molecular biology, the historical exegesis of
ancient texts, the examination of art objects, and historical geography.
From the time of the first systematic paleopathological investigations,
the presence of a very strange sort of lesion has been observed on ancient
bones from various parts of the world. It was mysterious for the simple
reason that no one was able to identify it with any current disease entity
in a truly persuasive way. The lesion consists of porous zones disposed
bilaterally and almost symmetrically on the cranial vault. The surface. of
the affected areas, which resembles pumice stone, is rough, pitted with
small cavities, and sometimes covered by a thin latticework of bone. His-
tological examination reveals that the diploe is enlarged, has rarefied zones
and restructured trabeculae; the outer table is thinned down, hollowed
out, and often disappears altogether in a newly formed bone mass; the
inner table is unchanged. The preferred locale for such lesions is the pari-
etal bones and the frontal bone. They can extend to the temporal, occipi-
tal, and sphenoid bones or even to the long bones of the axial (ribs) and
245
appendicular skeleton. Localization on the upper walls of the eye sockets

1snot rare.
The first description of this curious osteopathy was made by Hermann
W elcker, who studied it in 1885 on skulls from Peru, Java, Tuscany, and
Africa. He was struck by the cribrate or riddled appearance of the eye
sockets and named the lesion "cribra orbitalia." Influenced by R. Vir-
chow's ideas about "pathological races" and "degenerate" peoples,
Welcker thought this morphological state was a racial characteristic, an
inherited abnormality due to excessive development of the blood vessels. 2
Symmetrical porosity of the cranial vault was also observed on a Dayak
skull, on a very ancient Egyptian one, and then on bones excavated in
Nubia. 3 J. Saint-Perier thought the lesions syphilitic on a child's skeleton
he found in 1913 amid the debris of a Gallo-Roman building at Souzy-la-
Briche near Etampes. In the light of current knowledge, the "circular
exostoses projecting outward from a surface riddled with channels perpen-
dicular to a surface of newly formed bone,'' localized on the parietal bones
and the eye sockets, accompanied by thickening of the diploe but with the
inner table intact-all this cannot be the consequence of syphilitic osteitis,
but belongs instead to the same nosological group as the porotic hyperos-
tosis on ancient American and Egyptian skulls. The child whose skeleton
was found at Souzy-la-Briche cannot have been older than 8 years of age at
the time of death, while the Roman ruins in which it was buried date
approximately from the fourth century A.D. 4 Unfortunately, old descrip-
tions like this one suffer from a low level of general knowledge in osseous
pathology. They can be used today only with the greatest caution. At the
time, fine interpretation of X-rays of bone lesions was still in its infancy.
In 1914, Ales Hrdlicka gave a fairly detailed description of cranial porosity
in the pre-Columbian Indians of Peru. He attributed a special paleopatho-
logical significance to this bone state and called it "osteoporosis symmet-
rica." By his lights, symmetrical osteoporosis was a disease sui generis that
affected American Indians living along the coast. Everything seemed to
suggest that it was a childhood disease whose issue was often fatal. He did
observe lesions of the same sort on the skulls of adults, but they were less
serious and even seemed "scarred" or marked by a repair process. Probably
they were victims of the disease who had escaped death during an acute
infantile phase and even considered themselves cured, though they kept
the anatomopathological consequences for the rest of their lives. Hrdlicka
was correct in observing that these cranial lesions were not in fact a local-
ized disease but the only remaining signs of a "systematic disorder" whose
other consequences were lost on tissue less resistant to the passage of time. 5
In 1929, Herbert U. Williams confirmed the majority of Hrdlicka's con-
clusions. He also studied the histological and radiological aspects of this
osteopathy on the skeletons of Anasazi children in Utah and Arizona,
explaining their lesions as a functional hyperplasia of the bone marrow.
POROTIC HYPEROSTOSIS, ANEMIAS, MALARIA 247
He drew attention to the morphological affinity of these bone lesions with

others that had just been described in some anemic children. 6 At the same
time, S. Moore was stressing the similarity between the cranial lesions in a
kind of hereditary anemia and those of a cranial osteopathy among the
ancient Maya.7 E. A. Hooton (1930) demonstrated the extraordinary fre-
quency of this disease in Indians of the Yucatan peninsula: two-thirds of
the children in Chichen Itza were seriously affected by it. 8 In France, Leon
Pales diagnosed this kind of osteoporosis in 1930 on some fragmentary
children's parietal bones from the Neolithic period (the dolmen of Boujas-
sac in Lozere). They had previously been considered syphilitic, on the
authority of J. Parrot. 9 I have examined these specimens (Musee de
l'Homme, Prunieres Collection, inv. nos. 17229 and 17230). Syphilis
can be ruled out. A diagnosis of symmetrical osteoporosis is plausible
but uncertain, since the bone fragments are too small to reconstruct the
exact topography of the cranial lesions. The work of R. L. Moodie
has filled out our knowledge of osteoporosis in Egypt. 10 A few cases
have been described among black Africans and the Chinese. Finally,
in 1946 J. L. Angel reported the frequency of osteoporosis on prehis-
toric skulls from Greece, but at first he thought it just a sign of malnu-
trition.11
The expression "symmetrical osteoporosis," which has been used by
most experts since Hrdlicka, is inappropriate and should be eschewed. It
does not correspond to the anatomopathological reality and, moreover, it
generates a regrettable confusion with senile, endocrine, and nutritional
osteoporosis. 12Osteoporosis in the strict sense of the word is marked by
thinning down and rarefaction of the osseous trabeculae without any ac-
companying osteoplastic activity. However, the paleopathological lesions
under discussion here consist above all of a thickening of the bone struc-
ture. Osteoporosis of the medullary space is just a secondary aspect of
hyperactivity in that organ. Accordingly the term "hyperostosis spongiosa
cranii" (H. Mueller, 1934) has been suggested, 13 but I prefer "porotic
hyperostosis,'' a term invented by the American anthropologist J. L. Angel
during his paleopathological investigations of Mediterranean archaeologi-
cal remains. 14
Although I think it judicious to drop the term and the notion of sym-
metrical osteoporosis, I do no_t accept the current habit of totally con-
founding under one name the porous zones of the cranial vault and those
in the orbital region of the frontal bone. Without a solid proof of the
etiological identity of these two manifestations, I prefer to keep Welcker's
old expression, "cribra orbitalia," exclusively to designate the orbital lo-
calization. My position agrees with that of D.S. Carlson, G. J. Armelagos,
and D. Van Gerven, who stress the difference between '' osteoporotic
pitting" (a mild form of osteoporotic affection of the vault), "spongy
hyperostosis," and "cribra orbitalia. " 15
In any case, all these names aim merely at a condensed anatomopatho-

logical definition. They do not contribute to the problem of the idiopathic
disease with or without secondary diseases that these bony lesions betoken
or, more precisely, whose aftereffects they are. In the past, paleopatholo-
gists first thought of artificial deformations of the skull, then of rickets and
syphilis. Nowadays, the latter diagnosis can be confidently ruled out, since
X-ray analysis reveals morphological peculiarities that differentiate porotic
hyperostosis from the osseous lesions that are produced by the infectious
activity of the treponeme. The same is true for primitive osteoporoses and
osseous lesions of toxic and endocrine origins. They have radiological char-
acteristics that are indeed distinct from those found on ancient skulls
affected by porotic hyperostosis. 16 Hrdlicka imagined some toxic factor
but was unable to specify what kind. Wood Jones suggested that the
"cranial ulcerations" of the ancient Nubians were caused by their carrying
heavy water jars on their heads. 17 Under the influence of E. A. Hooton,
anthropologists at times confuse this disease with osteoporosis of nutri-
tional origins. 18 That is certainly a mistake, but the possibility remains that
porotic hyperostosis is linked to malnutrition by another physiopatholog-
ical process. Vitamin D deficiency may be responsible for it, at least in
part. 19 Experienced pathologists have said as recently as 1955that anatomical
examination of affected skulls does not in itself allow us to decide whether
the disease is a kind of rickets, or a hemolytic anemia, or a combination of
the two. 20
H. U. Williams was the first to hint at the right solution by comparing
X-rays of porotic hyperostosis with those of the lesions that had just been
discovered in children suffering from hereditary anemias. For S. Moore,
the analogy between this paleopathological affection of the cranium and
sickle cell anemia merited detailed investigation. Despite these insights,
Williams went astray and ended up with a hypothesis that is both ingen-
ious and fallacious: that the pressure produced on the occiput of a baby
resting on a head board could produce passive venous hyperemia and that
it, along with disturbances due to vitamin deficiency, results in symmetri-
cal osteoporosis of the skull. 21 Subsequently, Leon Pales returned to the
initial hypothesis of Toldt and Adachi and combined it with Williams's.
He explained this disease as the result of pressure exerted on the skull
of newborns in an effort to deform it artificially for ritual or aesthetic
reasons. Any such action on the skull of a newborn, says Pales, "produces
venous stasis and, as a result of the changes in blood pressure, a hyper-
vascularization of certain cranial regions, in particular the diploe and the
outer table. " 22 No concrete clinical observation was evoked in support
of this etiology. In 1951, H. E. Sigerist still cited the opinions of Williams
as the best explanation of porotic hyperostosis. 23 However, a much
more plausible solution was already within the grasp of medical his-
tonans.
Clinical Picture t}nd Physiopathologyof the Thalassemias

In 1925, Thomas B. Cooley and Pearl Lee, physicians in Detroit, discov-
ered a strange kind of anemia through clinical observations in five children.
They defined a distinctive syndrome consisting of a drop in circulating red
cells, pronounced enlargement of the spleen along with slight liver enlarge-
ment, skin discoloration, and osseous lesions primarily on the skull. From
the start, Cooley was struck by the mongoloid facial appearance and sig-
nificant bone changes in his patients. He thought their anemia was con-
genital but not hereditary. 24 He and his colleagues subsequently refined
the morphology of the red cells in these anemics: of unequal size, they are
poor in hemoglobin, which is often concentrated in their middles (so-
called target cells). It soon became clear that this type of anemia is familial.
The Americans suffering from it were almost all of Mediterranean stock,
that is, they were the sons and daughters of Greek and Italian immi-
grants. 25 The Italian physicians F. Rietti (1925), E. Greppi (1928), and
F. Micheli (1929) described some relatively benign hemolytic syndromes
that subsequent researchers associated with Cooley's anemia to constitute
a single disease entity called thalassemia. 26
The term "thalassemia," which was proposed in 1936 by G. H. Whipple
and W. L. Bradford, is an inept contraction of the word "thalassanemia,"
devised in 1934 by J. Com by on the pretext that the anemia in question
was proper to inhabitants of the Mediterranean basin. As G. W. Corner,
one of George Whipple's co-workers in Rochester, remembers it, the word
thalassa was chosen to associate Cooley's anemia with the Mediterranean.
The inspiration came from lingering schoolboy memories of the exclama-
tion Xenophon imputed to his soldiers when they found themselves finally
before the sea. Apparently, it had been forgotten that the thalatta (sic) in
question was the Black Sea and not the Mediterranean. Because of a certain
euphony, no doubt, the term "thalassemia" will live on, although its
etymology and semantics can hardly be justified. The sea, thalassa) has
absolutely no role in the pathogenesis of this anemia, which, moreover, is
not even limited to the Mediterranean region. According to the formation
rules in modern medical terminology, thalassemia should be a pathological
state marked by the presence of sea water in the circulatory system. The
reality is somewhat different. Once more, usage has triumphed over mor-
phological principles.
The basic disturbance in thalassemia is an inborn error of hemoglobin
metabolism. The error is embedded in a nonsexual chromosome and is
transmitted from generation to generation like a dominant Mendelian trait
whose strongest expression is in homozygotes. 27 Some of those who suffer
from the disease have hemoglobin F in their blood (a fetal type that is not
produced in normal adults) alongside the usual hemoglobins A and A 2 • In
others, the quantity of hemoglobin A 2 is increased relative to hemoglobin
A. In still other, rarer cases, the appearance of special hemoglobins has

been reported. In short, there are several thalassemias, not a single disease
entity defined by a single molecular lesion. For the historian of diseases,
that is an important fact, for it justifies an explanation of its origin in
several foci. According to current hypotheses, the mutations that originate
the thalassemias affect regulatory, not st.ructural genes. Some thalassemias
slow down the synthesis of the alpha hemoglobin chain, as distinct from
others that act on the delta or the beta chain. In {3-A2 thalassemia, the
synthesis of hemoglobin A is reduced, so the amount of hemoglobin A 2
in circulating blood is relatively increased. The {3-F type is expressed by the
persistence of hemoglobin F after an individual's fetal stage. In alpha thal-
assemias, there are abnormal hemoglobins (H and Bart's) in the blood. 28
The severe clinical form, thalassemiamajor or Cooley's anemia, occurs in
hon1ozygous children, in whom the genetic trait is inherited from both
parents. It makes its presence known early, first insidiously, through ex-
treme exhaustion, irregular fever, sallow or yellowish skin, growth retar-
dation, and, at times, pain in the left hypochondrium. The fully developed
clinical picture is dominated by symptoms resulting from hemolysis (ane-
mia, jaundice, brown pigmentation), enlarged spleen, bouts of fever, and
deformed facial features (mongoloid facies). Thalassemic children have an
enlarged skull with lumps in the frontal and parietal regions, a broadened
forehead, prominent cheekbones, slant eyes set far apart, a large, subsident
nose, and a projecting upper jaw. 29 Death in childhood or adolescence is
the usual lot of homozygous victims. By contrast, those who inherit the
trait from only one parent reach maturity without particular difficulty.
The heterozygous forms are expressed clinically as mild disturbances (thal-
assemia minor or the Rietti-Greppi-Micheli syndrome) that can regress
and even become latent (thalassemia minima or Silvestroni-Bianco
microcytosis).
In all persons who suffer from thalassemia, there is hyperplasia of the
bone marrow, which compensates for the diminished longevity of red cells
by increased hemopoietic activity. The radiological signs of thalassemia
were outlined by Cooley but not precisely defined until about thirty years
ago. Modifications of the cranial vault are very pronounced in homozy-
gous patients and more restrained, though still evident, in heterozygous
ones. The diploe is thickened and contains clear zones; it forces back the
outer table, which is often thinned down and riddled with holes. The
marrow's hyperactivity demands a supplementary vascularization that is
provided by hyperplasia of the periostal zone and produces a porous ap-
pearance on the bone surface. The trabeculae take on a radiating form
perpendicular to the inner table; sometimes they end in a "hair-on-end"
pattern. 30 In homo zygotes, the hair-on-end pattern appears less frequently
than one might imagine after reading the earliest observations. In fact, it
is frequently lacking and only shows up late. In heterozygotes, by contrast,

who do not present severe clinical symptoms, it is relatively common. 31
The X-ray just described coincides perfectly with that of porotic hyper-
ostosis. 32 However, it is not a pathognomonic sign of thalassemia. It is
only the nonspecific expression of bone marrow hyperactivity. The locali-
zation of porous zones in patients with thalassemia corresponds to the
localization characteristic of some ancient skulls with hyperostosis. In its
minor forms, the lesions are restricted to the cranial vault (sometimes with
thinning down of the cortex of the long bones), while thalassemia major
does not spare the facial bones or the long bones of the appendicular
skeleton, in which the cortex is reduced, the medullar space enlarged, and
there is osteoporosis as well as a reticulated disposition of the trabeculae in
the metaphyses. Some authors attribute an almost pathognomonic value
to the thickening of the temporal bones and the upper jawbones with
inhibition of the pneumatization of the mastoid cellules and the maxillary
sinuses. 33
Other Hereditary Anemias

Another hemoglobinosis similar to thalassemia but characterized by the
presence of sickle-shaped red cells (drepanocytes or sickle cells) in periph-
eral blood also produces bone modifications linked to bone marrow hy-
peractivity. It could, at least in some cases, be the cause of lesions on
ancient skulls. Drepanocytosis or, to use its more common name, sickle
cell anemia-the all too common barbarism "sicklemia" should be
avoided-was described for the first time by James B. Herrick of Chicago
as the result of a microscopic examination of the blood of an anemic
black. 34 That first observation dates back to 1905, but the publication of
the case, extraordinary for its time, dates from 1910.
First detected among blacks in the United States, sickle cell anemia was
later discovered in black Africa as well as the Mediterranean along with an
important focus in southern India. Because of a mutation that is transmit-
ted from one generation to the next, the individuals afflicted with sickle
cell anemia synthesize an abnormal hemoglobin, called hemoglobin S. So
the disease is a "molecular" one. 35 Hemoglobin S is distinguishable from
hemoglobin A only through a structural detail: in position 6 of the b~ta
chain, glutamic acid is replaced by another amino acid, valine. That suffices
to endow the hemoglobin molecule with a particular instability. Under
certain environmental conditions, it crystallizes in the erythrocytes and
makes them take on an elongated shape resembling a sickle or a holly leaf.
These sickle-shaped red cells stick to one another and can inhibit local
circulation. Moreover, their longevity is reduced relative to that of normal
red cells. 36
Hemoglobin S occurs in 50 to roo percent of the red cells in homozy-

gotes and in 20 to 40 percent of them in heterozygotes. Persons born from
the union of two parents with the defect suffer from childhood on (severe
anemia, jaundice, progressive weight loss, bouts of fever, splenomegaly,
growth anomalies, thrombotic accidents, painful bones, abdominal crises,
ulcers on the legs). Only exceptionally do they survive beyond puberty.
By contrast, heterozygous carriers of the gene of this disease have only a
handful of minor complaints. 37
The osseous modifications that accompany this disease are due to bone
marrow hyperplasia, exactly as with thalassemia. So it is not surprising that
the morphology of the lesions is practically the same in both anemic affec-
tions being discussed here. Nevertheless, among those stricken with sickle
cell anemia, radiological examination of the skeleton reveals several signif-
icant peculiarities, in particular, areas of infarction with necrosis in the
small bones of the hand and, for adolescents or adults, osteoporosis of the
vertebrae. 38
Some other hereditary anemias that are much rarer (at least nowadays)
may be responsible for osseous reactions similar to porotic hyperostosis:
scherocytosis, elliptocytosis, and hemoglobinopathy C. In these ailments
the lesions are moderate and do not attain the severity of a thalassemia.
Favism, a genetic flaw that is widespread in Greece and southern Italy, can
also be the cause of a chronic anemia that sometimes produces cranial
lesions. 39 We now know that even a microcytic hypochromic anemia, if
prolonged, can induce porotic hyperostosis in children. This hematologi-
cal syndrome is not inherited; it is an iron deficiency anemia caused by
environmental factors. A drop in serous iron results either from a lack of
the mineral in food or from its loss through chronic intestinal hemor-
rhages. In the latter event, iron deficiency anemia is usually the result of a
parasitic ailment (ankylostomiasis, schistosomiasis, tapeworm infestation,
amebic dysentery, and so forth). Since the histological and radiological
picture of bone damage due to iron deficiency anemia can mimic exactly
that of a thalassemic or drepanocytosic osteopathy, it is that much more
important to single out physiopathological and epidemiological features
that distinguish among them. Lack of iron only stigmatizes children's
bones, and its results are practically restricted to the skull. Although this
type of anemia occurs often in children nowadays, its osseous complica-
tions are seen only rarely. This means that the appearance of porotic
hyperostosis in a case of iron deficiency anemia must depend on one or
more additional factors. According to recent clinical observations, the con-
comitant factors may be protein deficiency or rickets. Maternal malnutri-
tion during pregnancy and premature birth seem to favor the osseous
lesion. In some historical settings, all the factors just mentioned can be
interconnected to form a complex of nutritional distress. 40
So there is an astonishing multiplicity of possible causes for porotic
hyperostosis. The list could be made still longer by adding to it several
pathological states that produce the same kind of X-ray: polycythemia
vera, some congenital cardiopathies, and so forth. The inevitable conclu-
sion is that all pre- and protohistoric skulls having such lesions were not
necessarily afilicted with one basic disease and do not necessarily share a
common etiology. To make possible a differential diagnosis in specific
cases, one can imagine a variety of criteria: morphological differences that
take account of the state of all the bones in a skeleton, biochemical pecu-
liarities, the frequency of affected skulls in a given site and their geograph-
ical distribution, data on the sociobiological conditions of the population
in question, and so forth.
In children who have died as a result of Cooley's anemia (homozygous
type), one should find thickening of the facial bones, osseous invasion of
the paranasal sinuses including the mastoid cellules, exaggerated develop-
ment of the zygomas, and, in particular, osteoporotic foci in the long
bones sometimes accompanied by precocious fusion of the epiphyses.
Contrary to Cooley's first observations, in homozygous thalassemics there
are more often lesions of the appendicular skeleton without effects on the
skull than the reverse. The most consistent signs are thinning of the cortex
of the long bones and a reticulated appearance at the level of the elbow.
Poynton and Davey have pointed out certain aspects of the teeth that are
characteristic of thalassemia, but they have not yet been used by paleo-
pathologists for a differential diagnosis of porotic hyperostosis. 41
As for sickle cell anemia, it becomes probable when there is osteoporosis
of the vertebrae along with necrotic foci in the appendicular skeleton. The
sure absence of noncranial osseous lesions reduces the possibility of thal-
assemia major and of sickle cell anemia and increases the possibility of
thalassemia minor or iron deficiency anemia. The importance of comple-
menting the paleopathological examination of a skull with a methodical
inspection of the whole skeleton cannot be overstated. Such an inspection
is conclusive only after X-rays are made of all available long bones. In some
ancient populations, for instance, in central Europe or in Nubia, cribra
orbitalia occur fairly frequently without any other manifestations or po-
rotic hyperostosis. According to recent research, it seems more and more
likely that in such cases the main etiological factor is iron deficiency. That
hypothesis seems to coincide best with current understanding of the living
conditions in such populations and, moreover, it agrees well with obser-
vations on the etiology of this osseous lesion in apes. 42
But it is also apparent that a certain uniformity in the reaction of bone
marrow to very different physiopathological situations requires that in
paleopathology the etiological diagnosis of such states not be founded
exclusively on the examination of isolated skeletons. We must take into
account archaeological data on socioeconomic well-being, nutrition, rites,
and customs, nor should we forget to consider carefully all available dem-
ographic data and direct or indirect proof of the presence or absence of

various parasitic diseases. 43 Nor is the frequency of affected skeletons in a
specific necropolis a minor diagnostic factor. When it surpasses a certain
threshold, sporadic diseases like congenital cardiopathy can no longer be
contemplated. Differential diagnosis must then be oriented either toward
inherited endemic diseases like thalassemia or toward acquired ones, such
as iron deficiency anemia with a specific, common cause in a given popu-
lation, and ailments of infectious origin. In the case of thalassemia, there
must be severe cases in which long bones are affected as well as milder ones
with moderate and isolated lesions on the skull. Moreover, the relation
between the frequency of these two forms of the disease, one homozygous
and the other heterozygous, should not be far from that predicted by
population genetics (Hardy and Weinberg's distribution).
In a diachronic study of populations over a specified area, one would
expect the incidence of porotic hyperostosis, insofar as it reflects chronic
anemic states, to be inversely proportional to parameters that reflect the
overall well-being of the general population, such as longevity, fertility, or
abundance of food. The reversal of this relationship, that is, a positive
correlation between these parameters and the frequency of porotic hy-
perostosis, could be understood as a sign favoring the etiological diagnosis
of thalassemia or of some other nondeficient, hereditary anemia. I will
show below how John Lawrence Angel's research points to exactly this
positive correlation during certain periods of the ancient history of the
Mediterranean.
In the future, further refinements in differential diagnosis will be made
possible by the quantitative chemical microanalysis of the bones. 44 A still
more decisive method would be the biochemical analysis of hemoglobin
molecules preserved in mummified tissue or even, perhaps, along the by-
ways of the spongy portions of skeletons. For instance, a paleopathological
diagnosis of thalassemia could be confirmed by the presence of isoleucine
(a component of hemoglobin F) in red blood cells once they are isolated
by microchemical means, or, alternatively, by their form when imagined
by a scanning electron microscope.
Distribution ofThalassemias and

Hypotheseson Their Origin
Today the thalassemic trait is probably spread throughout all of main-

land Greece as well as the islands. 45 It poses a serious public health prob-
lem. A single statistic can make its dimensions clear: during the seven years
from 1948 to 1955, two pediatric hospitals in Athens admitted 421 children
(representing about 2 percent of all hospitalizations in these institutions)
with Cooley's anemia. They were homozygous cases with severe clinical
symptoms and very high mortality. 46 Relatively high frequencies of thalas-
semia have been reported among the native inhabitants of several villages
in Attica, in the Peloponnesus (Sparta, Corinth, and so forth), in the
southern portions of mainland Greece (Thebes, Missolonghi), in Thessaly
and Epirus as well as among inhabitants of the islands (Cephalonia, Zante,
the Euboea, Mytilene, Samos, Crete, Rhodes, and so on). The rate of
healthy carriers of the trait varies between 5 and 10 percent, roughly speak-
ing, except for the Ionian islands and Rhodes, where the frequency is
as high as 14-16 percent. 47 Systematic examination of a group of 1,500
soldiers from various parts of Greece has made it possible to establish an
approximate national average of heterozygotes without overt problems:
7.7 percent. 48
These data relate to beta thalassemias and especially to the so-called
Mediterranean type, {3-A2 , which is the predominant one. Nevertheless, it
is true that the other forms exist alongside these: in Greece, the thalassemic
gene is not uniform. 49 Type F, which is well known in Asia and reported
in Africa, is rare in Greece. The alpha thalassemias that are common in the
Far East are exceptional in the eastern Mediterranean. It is significant that
the {3-A2 thalassemia is fairly evenly distributed in mainland Greece, while
the other types occur chiefly in restricted foci.
There is one place where several hereditary anemias meet: the region of
Arta in Epirus. For a very long time malaria was rampant in the flatlands
there. According to G. R. Fraser and his co-workers, 50 the inhabitants are
carriers of beta thalassemia (10 percent), alpha thalassemia (5.4 percent),
delta thalassemia (9.7 percent), 51 favism (10.8 percent), and sickle cell ane-
mia (9.3 percent). By virtue of the examination of 3,650 soldiers from all
over Greece, Klonakis and his co-workers reported the presence of sickle
cell anemia in a heterozygous state among 0.45 percent of their sample.
This average does not speak to the real situation obtaining in Greece, since
the sickle cell trait is concentrated in certain areas that are ancient zones of
the malarial hyperendemic. For instance, I can cite two foci: the peninsula
of Chalcidice, where 23.5 percent of the inhabitants are carriers of sickle cell
anemia; 52 and the region of Lake Copais in Boeotia, with 20 percent. 53 For
the opposite state of affairs, compare Rhodes, which has a high rate of
thalassemia but is practically free of sickle cell anemia.
Thalassemic hereditary anemias are not just the prerogative of Mediter-
ranean peoples. 54 They were discovered by an American physician among
children living in North America; according to Neel, the frequency of
carriers of the defect in New York in 1945 was slightly higher than
4 percent. However, the defective gene does not occur in American Indi-
ans of unmixed breed or in descendants of the Anglo-Saxon and Spanish
colonizers of America. Australia and Ja pan appear to be free of it. In most
DISEASES IN THE ANCIENT GREEK. WORLD
European countries, for example in mainland France, in Germany, or in

the Scandinavian countries, there are sporadic cases whose origins are ob-
scure. They could be a matter of immigration or recent mutations.
In Asia the situation is different indeed. Just as happened before with
the discovery of the Mediterranean endemic, thalassemia among the Chi-
nese was first reported among immigrants in the United States (L. P.
Foster, 1940), then confirmed by Chinese physicians working amid native
populations. Today, there can be no doubt about the magnitude and
antiquity of its Asian foci. Moving from east to west, one can see that the
Mediterranean focus extends at least to the edge of the Caspian Sea. Thal-
assemic defects are common in Iraq, Iran, Georgia, and Azerbaidzhan.
Then there is a gap that may as easily be due to the inadequacy of hema-
tological data about that part of the world as to the actual absence of
hereditary anemias there. The situation in Afghanistan is poorly known,
but the research of A. G. Maratchev, unknown in the West, reports the
presence of thalassemia in Tadzhikistan and Uzbekistan. 55 From Pakistan
and India on eastward, the rates become significant again.
Outside the modern Greek state but within the domain of the ancient
Hellenic world, thalassemia is very common on Cyprus (16-20 percent,
including both Greek and Turkish Cypriots), and widespread in Turkey
(especially along the coast of Asia Minor), Macedonia, and around the
Greek settlements in Dalmatia and Italy. The distribution of the thalasse-
mic defect in Italy is especially well documented thanks to the research of
Ezio Silvestroni and Ida Bianco. They examined close to 50,000 individuals
between 1945 and 1959. Thalassemia is common in four regions: Sardinia
(10-20 percent and even, in Carbonia, as high as 27 percent), Sicily (espe-
cially its southern coast, where the rate reaches II percent), the coastal
regions of the southern part of the peninsula, especially Calabria and Puglia
(5-12 percent), and finally, though at first sight paradoxically, in the Po
delta, between Ferrara and the Adriatic coast (8-22 percent). In the rest of
Italy, the distribution of thalassemia is fairly uniform and moderately com-
mon (0.4-5.0 percent). Most of the cases reported outside of these foci
originally stem from the four regions. 56 Since the current havens of thalas-
semia correspond exactly to areas in which Greek colonization of Italy was
particularly intense, Silvestroni and his co-workers have suggested that
Italian thalassemia was a disease with Greek origins. Thalassemia is also
found in Malta (4-7 percent) and in North Africa (particularly Tunisia and
Egypt, though the figures are still uncertain). It occurs in the western
Mediterranean, for instance in Corsica, Portugal, and in some Spanish
provinces (Valencia, Galicia, the Balearic Islands). All these cases consist in
small foci to which immigrants from the eastern Mediterranean may well
have brought their thalassemic genes.
Thalassemic regions have been reported in the Punjab, in the region of
Madras, in Bengal, and in Sri Lanka. But the defect seems relatively un-
common in Burma and in the parts of India not mentioned above. China
is assuredly much affected, but the exact distribution of thalassemia within
that huge land is unknown. It is believed to be more common in the
southern provinces than in the north.
Soviet publications like the handbook of geographic pathology by A. P.
Avtchin (1972) say nothing about thalassemia in Mongolia. I know of no
study confirming that thalassemia is common or even present in modern
Mongolia. That is a fact that should not be neglected in discussions of the
origin and historical migrations of thalassemia. On the other hand, thalas-
semia is frequently encountered in Thailand, Cambodia, Vietnam, Laos,
and Indonesia. According to Lucien C. Brumpt, it was imported from
China: the genetic defect is absent from Khmers not interbred with Chi-
nese. 57 Sporadic cases have been published from the Philippines and Bor-
neo. Finally, I mention the still unsolved problem of the presence of
thalassemia (and of pseudo-thalassemias) in black Africa, especially in Zaire.
Here I share the reservations of Brum pt, Lancaster, and Lehmann. The
types in question do not correspond genetically to the Eurasian beta
thalassemias.
In my presentation on the distribution of this disease outside of the
Mediterranean, I have not cited frequency figures. If we set aside some
doubtful claims, they vary between I and 10 percent (heterozygous carriers
as a percentage of the whole population). But the figures in question have
been obtained on such differing samples and by such a variety of proce-
dures that they do not all merit equal credence. The notion of statistical
comparability among them is but an illusion.
I have not taken the variety of forms of thalassemia into account because
of uncertainties in this regard in the documentation that I have been able
to consult. From existing surveys it is tempting to conclude that in the
thalassemic zone that stretches from the Mediterranean to Southeast Asia,
all the genetic forms of thalassemia and especially thalassemia (3-A2 are
represented everywhere, though with incidences varying from place to
place. Alpha thalassemias are more widespread in the Far East than in
Europe and Asia Minor.
Since the nature of the thalassemias was first understood, people have
suggested ways to account for their distribution. Numerous authorities
have sought to prove that thalassemia is a racial trait that appeared once in
the history of humankind. Its current distribution should then be expli-
cable in terms of interbreeding and migration. According to Ignazio Gatto,
Cooley's anemia is a pathological hereditary trait that arose by mutation
in a human group that lived in Europe during the Upper Pleistocene. 58
From this so-called paleo-insular group that peopled southern Italy and
Greece before the Mediterranean race, the genetic defect spread through-
out the Mediterranean. Gatto's main argument is the discovery in the cave
of San Teodoro in Sicily of Paleolithic skulls that, according to him, bear
thalassemic lesions. That paleopathological diagnosis is, as I will show

below, far from convincing. The localization of the initial mutation in
time and space proposed by Gatto seems to me utterly arbitrary. More-
over, the Asian foci are inexplicable on his hypothesis.
According to Edward C. Zaino, thalassemia was born more than 50,000
years ago in a Mediterranean valley to the· south of Italy and Greece that is
now submerged. 59 He justifies his choice by the fact that the place in
question is near the zones where the degree of thalassemia is now highest.
Zaino says that the original valley must have lain to the south of these
countries, since thalassemia is rare in Yugoslavia, which lies between them.
He can be reproached for a methodological error and an error of fact. First,
it is not inevitable that the locale of a mutation so distant in time be in the
neighborhood of those areas that nowadays have the highest frequency of
it. Second, surveys by G. R. Fraser have proved that thalassemia is not rare
in Dalmatia and that it has a continuous distribution from Greece along
the Adriatic shoreline and the Po delta probably extending northward into
Hungary. 60
But Zaino's theory has one advantage: his imaginary valley bridges Eu-
rope and Africa and thus makes more likely certain hypothetical migrations
of the genetic defect. Since he interprets as thalassemic all the osteo-
archaeological cases of porotic hyperostosis that I have spoken of in earlier
parts of this chapter, in particular the skulls from Nubia, South America,
and the Far East, Zaino has to face up to the awesome task of explaining
the routes for such a dissemination of thalassemia. In his opinion, the
mutation took place during the third glaciation of Wurm. Once the gla-
ciers melted, the inhabitants of the flooded valley emigrated to Greece,
Italy, and North Africa. From there, the thalassemic gene migrated toward
the east, into Turkey, Mesopotamia, India, and China. The possibility of
a second mutation taking place somewhere in Indochina is entertained but
considered unlikely. From China, the defect took a northward route along
the Pacific coasts of Asia and finally across the Bering Strait onto the
American continent. In passing Zaino asserts that thalassemia was perhaps
the mysterious scourge that caused the disappearance of the Maya. It is
surprising that this person, who bases all the first part, of his hypothesis on
the persistence of thalassemia among current populations, is in no way
embarrassed in the second half of his study by the total absence of the
pathological gene an1ong modern American Indians.
A more serious hypothesis has been put forth by two physicians from
Marseille, Andre Orsini and Louis Badetti. They use theories on the origin
of races to situate the thalassemic mutation in time and space. A voiding
risky business like small-scale n1igrations and interbreeding, they do not
take a sn1all ethnic group like the paleo-insular group or a single Greek
tribe as their point of departure. Instead, they suggest that thalassemia
"affected selectively a very broad racial grouping, which explains its diffu-
sion. " 61 The mutation took place in "short, dark dolichocephalics" before
this Mediterranean' racial complex, as defined by H. Vallois, split into
three races: Mediterranean, Inda-Afghani, and South Oriental. At first
sight, the superposition of current thalassemic zones over the area of ex-
pansion of these dark dolichocephalics is striking. However, the hypothe-
sis does not stand up well to testing anthropogeographic criticisn1.62 The
main stun1bling blocks are Chinese thalassemia (that of the north is unex-
plained while that of the south calls for the interbreeding of a Mongolian
racial group with Inda-Afghani elements) and the rarity of the genetic
defect in the western Mediterranean.
Thalassemia: Greekor Mongolian?

Before the discovery of Asian foci, the simplest and probably also the
most satisfying solution was to attribute the thalassemic mutation to the
Greek "race." This hypothesis was first put forth by authorities in the
r93os who were not yet worried about the Far East (J. M. Baty and
V. Chini, among others). It was revived and elaborated by the Italian
investigators Ezio Silvestroni, Ida Bianco, and Nereo Alfieri. 63 In order to
avoid the confusions that mark current discussions of this subject, I resolve
the so-called Greek hypothesis into three elements, which are the answers
to three distinct problems: (r) the historical period and geographic locali-
zation of the first mutation; (2) the spread of the genetic defect over the
Mediterranean; and (3) its transmission into the Far East.
As to the first, proponents of the Greek hypothesis do not take a clear
position. Was the mutation produced in Greece among some pre-Hellenic
people, or among the Dorians after their invasion, or on Greek soil after
the great migrations of the twelfth century B.c.? The latter seems the
preferred view, but it is couched in vague terms. It seems to me improper
to speak of a Greek "race" when referring to the Greek people of the
protohistoric period, which in my view was the result of complex inter-
breeding. If we accept the conclusions of J. L. Angel, which are based on
paleopathological research to be discussed below, thalassemia in Greece
dates from the Mesolithic period or even from the end of the Paleolithic
period. It is therefore prior to the formation of the Greek people.
The power of the Greek hypothesis resides in the second point: Hellenic
colonization accounts well for the spread of the thalassemic gene in the
Mediterranean basin. It is especially convincing for the distribution of
thalassemia in Italy and on the Dalmatian coast. The two usual objections
to it can easily be dismissed. The presence of the defect in places that were
not actually colonized by Greeks can be explained either by undoubtedly
real individual migrations or by the interbreeding of Greeks with Roman
settlers who became secondary carriers of thalassemia. The main argument
of those opposed to this theory, namely, the absence or rather the rarity
of the defect in certain regions where Greeks were very active over a long
period of time (the south of France, Catalonia), appears serious but is not.
Since the thalassemic gene is a pathological trait subject to natural selec-
tion, it could have been eliminated by then in specific zones under partic-
ular environmental conditions. Unfortunately, we still have very little solid
information about the environmental factors that exert negative pressure
or those that favor the thalassemic defect. Malaria, which will be discussed
shortly, cannot be the only exogenous cause.
The Greek hypothesis is vulnerable when it comes to the third issue.
According to Silvestroni and his co-workers, the soldiers of Alexander the
Great brought thalassemia to India, and from there it passed into China
on the silk routes. That thalassemia reached western Asia poses no prob-
lems to the historian, but it is more difficult to account for its diffusion in
the Far East. It is certain that commercial links existed between China and
the Mediterranean well before the period of the Mongolian conquest. In
population genetics, the time factor is as important as the space factor.
One only need introduce a single gene into the pool for it to proliferate, if
the environmental conditions are favorable and if there is enough time.
Instead of seeking the solution to the problem in great historical migra-
tions, it is possible to adduce exchanges on the individual level-but then
we need to allow more time. Clearly, the arrival of a single thalassemic
gene has the same effect as a mutation. Nothing prevents us from asserting
that one did arrive, for instance, in the Bronze Age (a period during which
there was a certain unity of style and technical execution in various objects
from China and the Mediterranean, according to the studies of 0. Janse).
To be sure, such an assertion is purely speculative, arbitrary, and without
scientific value. Nevertheless, I am playing this guessing game in order to
show that, biologically speaking, the genetic identity of the European and
Chinese foci is not impossible, and that the transmission of genes from
one region to the next could have happened in a historically invisible way.
Some researchers have tried reversing the general direction of the hypo-
thetical migration of thalassemia. The point of departure for this was a
clinical observation, namely, the mongoloid facies of thalassemic children,
which was noted in the very first studies by Cooley. This pathological
aspect of the face was studied mainly by the pediatrician J. Caminopetros,
who observed it in his Greek patients. His first thought was the possibility
of an Asiatic ancestry to the disease. In his article on erythroblastic anemia
in the eastern Mediterranean, Caminopetros honestly but reluctantly rec-
ognized that clinical and radiological investigations "tend to rule out a
racial etiology for the mongoloid facies in patients afllicted with erythro-
blastic anemia, in other words, its hereditary transmission going back to
the Mongol race. On the other hand, they justify considering the mon-
goloid facies and in general the deformation of the bones of the head in
erythroblastic anemia as the result of the effect of the disease on bones. " 64
Subsequent research fully confirmed Caminopetros's prudent conclusion.

The n1ongoloid facies is not linked to specific genetic factors but results
from osseous alterations caused by a hemolytic disease.
But Caminopetros did not turn from the road once taken. According
to him, thalassemic children had other Asiatic racial attributes, chiefly
Mongolian eyes and a blue spot in the sacral region. In some of these
patients, the family history brought to light Chinese or Mongolian ances-
try that was not so remote. So Caminopetros believed he was entitled to
conclude that "interbreeding with the Mongol race plays a role in the
etiology of the disease." In my opinion, his argument has no probative
value. It is true that, as against the facies, slanted eyelids, an epicanthus,
and the sacral spot are traits independent of hereditary anemia, and as a
result their frequency in thalassemics could be an important sign of racial
mixture. However, no study until now has confirmed the existence of a
significant statistical correlation between these morphological peculiarities
and the thalassemic defect. Their presence among Caminopetros's patients
is indeed due to the recent interbreeding that his family histories confirm;
but as we now know, that is just an exceptional circumstance that does
not occur in the majority of carriers of the thalassemic gene in the eastern
Mediterranean.
But the Mongolian hypothesis got a second wind from the research of
Lucien Brumpt in Southeast Asia. In Cambodia and Vietnam, thalassemia
"seems to exist in populations that have undergone crossbreeding with
the Chinese. " 65 Then why not also attribute to the Chinese responsibility
for the dissemination of the trait in Europe? However, anthropogeo-
graphic considerations made it necessary to relegate the Chinese to an
accessory role and to thrust the Mongols into the limelight as primary
carriers. According to Brumpt, thalassemia is a "sinemia," or better still,
a "mongolemia." "I consider thalassemia a Eurasian hemopathy brought
by the Mongols, who first spread it into South China five centuries before
Christ. Then the Chinese disseminated it throughout Southeast Asia. In
Europe, after Atilla, the Huns entered the service of the Byzantines as
mercenaries and scattered it in all their Mediterranean garrisons. The geo-
graphic distribution of thalassemia corresponds to the map of the Byzan-
tine Empire at its apogee. " 66 Jean-Fran~ois Pays espoused this view and
tried to buttress it with historical and genetic speculations. 67 The discovery
of thalassemia in China and Iran, Brumpt and Pays aver, "did not upset
those who cling to the theory that the Mediterranean basin was the cradle
of the thalassemic defect; none of them was amazed that subjects with
thalassemia major persisted in presenting a Chinese facies, though they had
every right in the light of their hypothesis to discover in such individuals a
Hellenic profile or the Roman gaze immortalized in so many master-
pieces. " 68 This is eloquent, but it does not rest on a single valid scientific
argument. Without a statistical analysis of precisely defined racial traits, it
is rash to assert, as they do, that other than being a pathological mani-
festation, the thalassemic facies represents ''the resurgence of certain
anthropological traits proper to the Mongol race." The contention seems
to me most unlikely. Why would racial traits that are not recessive be mani-
fested only in patients with thalassemia major? To my knowledge, it has
not been maintained that they occur in heterozygous carriers of the genetic
defect.
This is not the place to discuss the historical details with which Brumpt
and Pays try to orchestrate the Eurasian diffusion of thalassemia; they only
serve to conceal a faulty substructure. First of all, we should recall that the
thalassemic defect, as common as it is among modern Greeks, has not been
proven to exist among the Mongols of Gobi or Altay. The notion of such
a recent introduction of the abnormal gene into the Mediterranean basin
''to the gallop of Mongolian cavalry and under the canvas of the lumbering
wagons of the Hun" 69 does not accord well with current rates in the gene
pool, especially for mainland Greece, which cannot be considered a
backwater.
The difficulties become still greater if we suppose, along with P. Bugard,
that "the current distribution of thalassemia in Greece is a vestige of
Turkish domination.' ' 70 Selective action favoring the thalassemic gene,
whether due to malaria or to other, still unknown factors, could not then
have had enough time to produce frequencies higher than 7-10 percent. I
add, finally, that some paleopathological and archaeological indices favor
the presence of the thalassemic gene in the eastern Mediterranean at least
prior to the Hellenistic period (statuettes from Smyrna) and probably
before the Neolithic period. In this regard, it is surprising that Brumpt
and Pays were ignorant of the research of Angel that will be discussed
below.
Why search so obstinately for a single origin to thalassemia? "Unitarian"
hypotheses do not account for the diversity of the thalassemias. It may
well be that each of several genetically different forms had its own geo-
graphic distribution. If so, their superposition, which could produce a
picture like the one described above (' 'Current Frequency of Hereditary
Anemias in the Eastern Mediterranean"), makes it impossible to sort them
out at this point in time. Theoretically, it is inadmissible that thalassemias
{3-A2 , F, and alpha all derive from a single historical mutation. The hy-
potheses that feature a single origin, Greek or Mongol, should nowadays
be restricted to the most common beta variant. But even then, is it not
possible that the current distribution results from several mutations in
different zones? On that reading, the Mediterranean and Asiatic foci would
be independent of one another. For Brumpt and Pays, that is "a facile
solution" that must be rejected because of a theoretical obstacle: "Since a
mutation is actually almost always the result of chance, the chance that it
POROTIC HYPEROSTOSIS, ANEMIAS, MALARIA
took place at any particular place and at any particular time are practically
the same. It woukf then become difficult to explain why the thalassemic
defect is absent from Australia and America ... despite identical climatic
and ecological conditions obtaining both in Europe and Asia. " 71 This is
faulty reasoning, since the identity of the ecological conditions in thalas-
semic areas and the rest of the world needs to be proven, not assumed.
Moreover, the "randomness" of mutations is governed by the laws of
thermodynamic stability acting on the genome, and their appearance can
follow a certain order. In light of the most recent genetic studies, the
hypothesis of an origin for beta thalassemia in several foci seems to me the
only satisfactory one. Biochemical variants were individualized even with
the "Mediterranean" type.
Some isolated cases of thalassemia minor, reported when the disease was
being investigated in regions where it is extremely rare, lead one to suspect
the existence of mutations that are independent of the historical main-
stream. In fact, 0. Tonz and his co-workers had the extraordinary good
luck to discover a new mutation: a young Swiss girl with all the biochem-
ical traits appropriate to a heterozygous carrier of thalassemia {3-A2 , but
whose parents (their parenthood was confirmed by careful genetic analy-
sis), sister, and brothers are completely normal hematologically. 72 Accord-
ing to Tonz's calculations, at least five centuries must pass for such a
mutation to occur in a population the size of modern Switzerland. To be
sure, this makes it an exceptional event, but from the historical perspective
that we are taking here it is a sufficiently common event to exempt us from
believing in the more or less arbitrary hypotheses about genetic links be-
tween the Mediterranean and the Far East.
Another important fact has been reported only during the past few
years: a particular form of glucose-6-phosphate dehydrogenase deficiency,
the B( - ) variant of favism, has a global distribution that corresponds to
that of beta thalassemia (eastern Mediterranean, western Asia, India,
Southeast Asia). This seems to raise once again the question of genetic
migrations between Europe and the Far East. But that would be to mis-
state the problem once more. The essential question is, How has it come
about that neither the thalassemic defect nor the favic one was carried
northward? Clearly, the opportunities for such a migration were not lack-
ing. But it is not enough that an abnormal gene simply be introduced into
part of the world for it to attain statistical significance. Its maintenance· is
guaranteed only by environmental factors that favor it-an issue to which
we will return in the discussion of malaria.
The restriction of the main variant of favism and of all forms of thalas-
semia (whose origin, of necessity, is due to several independent mutations)
to the well-defined "Eurasian ribbon" must be the result of some specific
ecological factors. The selecting action of environment is as decisive as that
of the chromosomic substratum, and historical reality was forged by their

. .
1nteract1on.
Distribution and Origin of Sickle Cell Anemia
The zone of maximum frequency of sickle cell anemia, in which heter-

ozygous and homozygous carriers number 25 percent of the population,
tesembles a large belt that stretches across western Africa, Equatorial Africa
(from the Sahara and from the Sudan to Zambezi), and Madagascar to Sri
Lanka and southern India. Among some African tribes the frequency rises
to 30 or 40 percent (for instance, in Togo). The disease also occurs in lower
concentrations in Egypt, Italy, and Greece, and among American blacks. 73
Unlike thalassemia, which from the standpoint of molecular pathology
lacks nosological unity, sickle cell anemia is due to a simple mutation of
the genetic program for cell structure, not for regulation. The hypothesis
of a unique mutation, or at least of a very restricted number of mutations,
seems better founded here than in the case of thalassemic defects.
Everything points to the fact that this ancient mutation took place
within a black community and that the presence of hemoglobin S in
persons of other races is due to interbreeding at some early date. The
mountain people of the south of India who carry this gene do have some
Negroid traits. However, there are substantial difficulties in explaining the
transmission of the sickle cell gene as far as India. Hermann Lehmann and
Marie Cutbush thought that the trait derived from a Veddit population
that had lived in the Arabian peninsula during the Mesolithic period.
From this source, the gene traveled in two directions, toward the heart of
Africa and also toward India. For Lehmann and Cutbush, the transport of
the sickle cell gene from Africa is impossible because of the absence of
certain hematological traits-particularly the He and R (cDe) antigens proper
to blacks-in the present inhabitants of southern India. 74 According to
Peter Brain, the spread of the trait into Africa took place at a relatively late
date, around the beginning of the Christian era, and followed the same
path as the spread of zebu husbandry. 75
Some researchers, particularly E. C. Bucchi, believe that extant sickle
cell anemia results from two independent mutations, one African, the
other Asiatic. 76 Whatever the case may be, one mutation or two, African
or Arabic, the introduction of the trait into Greece must have been linked
to the immigration of blacks from Africa. The autochthonous origin of
Greek sickle cell anemia has been suggested, but in our present state of
kdowledge, nothing supports the assumption that a supplementary mu-
tation took place there. J. Bernard and J. Ruffie attribute the introduc-
tion of hemoglobin S into Greece "to the invading armies and the black
soldiers they hired.' ' 77 That introduction probably dates back to classical
antiquity. 78 #
The Role of Malaria

The geographic zones in which the several thalassemias are common and
those in which high frequencies of sickle cell anemia have been observed,
when taken together, strikingly overlap those parts of the ancient world in
which falciparum or malignant tertian malaria raged in historical times.
After a clever suggestion by the famous physiologist and geneticist John
B. S. Haldane, 79 the notion that hereditary anemias could depend on
malarial infestation took root. If these defects do provide a specific protec-
tion against malaria, it would be easy to understand how it can happen
that such a harmful gene is not eliminated by natural selection but rather
maintains itself at amazingly high frequencies among certain populations.
In 1954, Anthony C. Allison was the first to show that areas with a great
number of persons affected by sickle cell anemia coincided with ancient
malarial regions, and also that attacks of malaria were less common and
featured more benign symptoms in heterozygous carriers of hemoglobin S
than in normal individuals. 80 In a strongly malarial area, the loss that the
sickle cell gene causes the population through the heightened mortality of
homozygous carriers is compensated for by the advantage it offers hetero-
zygotes in the form of a better defense against the severest form of malaria.
That is just one aspect of the general biological phenomenon that geneti-
cists call "balanced polymorphism. " 81 Mathematical models reveal how,
in a hyperendemic area, accrued mortality and diminished fertility in the
normal population produce progressive increases in the proportion of de-
fective individuals. The rates increase from one generation to the next until
they stabilize at a balanced value that depends on the degree of malarial
infestation. When the selection pressure of malaria declines, the rate of
sickle cell anemia slowly diminishes, from one generation to the next; it
constitutes a portion of the genetic load that is on the way to extinction. 82
The same explanation was applied to the thalassemias, to hemoglobi-
nosis E, and to the glucose-6-phosphate dehydrogenase (G6PD) deficiency
(R. Cepellini, H. Lehmann, L. Brumpt, A. G. Motulsky, and others). In
all these "inborn errors of erythrocytic metabolism," falciparum malaria
is the external factor responsible for their heightened frequency in well-
defined geographical areas. 83
In regard to sickle cell anemia, the hypothesis that makes the selection
pressure of malaria responsible for its survival gains confirmation on a daily
basis. The same is not true for the thalassemias. Epidemiological inves-
tigations on this subject have not provided conclusive results. 84 Investi-
gations carried out in Sardinia (U. Carcassi and R. Cepellini, 1957;
M. Siniscalco et al., 1959) and in the region of Ferrara (C. Menini, 1970)
favor a correlation between the two pathological states. 85 In other coun-
tries, the results have been contradictory. In Greece, several detailed stud-
ies have confirmed the concentration of hemoglobinosis S and favism in
areas where malaria was once prevalent, but for thalassemia the correlation
has not been so plain. 86 The situation in the Arta (Epirus) region is partic-
ularly interesting in this regard. There is a kind of historical experiment
taking place there. The valleys were strongly malarial until 1946, while
those in the mountainous regions did not suffer from the disease. Rural
communities in the whole region cling to a sedentary and fairly archaic
way of life. A systematic examination of four groups of boys from different
zones within it (two groups from the villages of the valleys, and two in
places without malaria) showed that sickle cell anemia and G6PD defi-
ciency are indeed more frequent in the first two groups (favic gene: 17.6
percent and 16.4 percent; sickle cell gene: 14.3 percent and 6.4 percent)
than in the two others (4.4 percent and 2.8 percent for the former; no
cases of sickle cell anemia). For thalassemia, the differences are not signifi-
cant: 13.2 percent and 15.2 percent in boys from the valleys, 10.1 percent
and 9. 7 percent in the other two groups. 87 Hasty conclusions are inappro-
priate: this evidence neither confirms the hypothesis about the influence
of malaria on the genetic survival of thalassemia, nor does it refute it. The
simultaneous presence of sickle cell anemia and several thalassemic variants
is an embarrassment, since competition between these diseases makes in-
terpretation of the data difficult. We know that sickle cell anemia and
thalassemia /3-A2 tend to exclude one another, probably because of the
clinical severity of the heterozygous thalasso-drepanocytic combination.
Falciparum malaria is not necessarily the only external factor that comes
into play in the establishment of a balanced polymorphism of sickle cell
anemia and the thalassemias. Factors still unknown and suspect ones like
amebiasis, iron deficiency, and endogamy, must also be taken into
consideration.
The Distribution and Paleopathological

Significance
of PoroticHyperostosis
Ancient cases of porotic hyperostosis have been reported in the follow-
ing geographical areas: Greece proper and neighboring regions with mixed
populations (Macedonia, Anatolia, Cyprus), Italy (particularly the south
and the Po delta), France, and Hungary; Egypt (especially Nubia) and
tropical Africa; India, Indonesia, China, and Japan; South America (espe-
cially Incaic Peru), Central America (Mayan Guatemala and Mexico, es-
pecially in the Yucatan peninsula) and North America (pre-Columbian
Indians whose bones have been exhumed in several American states, es-
pecially Arizona, New Mexico, Utah, Arkansas, Florida, Illinois, and New
York). Despite numerous excavations and conscientious paleopathological

examinations, no typical example of porotic hyperostosis is known to exist
on bone remains from Scandinavia, Great Britain, Germany, or the Slavic
countries. In France, of the two cases that have been described, one comes
from the Mediterranean zone and is of Neolithic date, and the other
concerns a subject found in a Roman context. There is no trace of this
osteopathy in northern Europe or in Siberia. On bones exhumed in the
Soviet Union that were rich in mongoloid traits, Rokhlin never observed
lesions that could be interpreted as induced by chronic anemia. 88 This does
not accord well with the hypothesis of Caminopetros and of Brumpt on
the role of the Mongols in the introduction of the thalassemic gene into
Europe.
There is a danger of confusing porotic hyperostosis with other osteopo-
roses not accompanied by medullar hyperplasia. In my opinion, certain
isolated forms of cribra orbitalia (without concomitant lesions on the cra-
nial vault or the postcranial skeleton) that occur in Egypt as well as on
medieval Europeans derive from states of nutritional deficiency and per-
haps even from an inflammatory affection of the eye. In a study of 285
skulls exhumed in Nubia, D. S. Carlson and his co-workers interpret cribra
orbitalia as a sign of secondary anemia brought on by intestinal parasitism,
lack of iron, chronic diarrhea, and an abundance of childbirths. They stress
that the paleopathological evaluation of these lesions in the ancient popu-
lation should take account of several biological and ecological variables. 89
Indeed, anemia can have a wide variety of causes. In Egypt, the role of
schistosomiasis, an anemia-causing disease par excellence)must not be for-
gotten. And it is not at all impossible that cribra orbitalia and mild forms
of porotic hyperostosis are at least in part linked to malaria by way of the
anemic states that it causes. In that case, malignant tertian malaria is not
the only thing to blame. In the same vein, Calvin Wells found cribra
orbitalia on 6 percent of 200 Anglo-Saxon skulls he examined. As for a
diagnosis of their etiology, he hesitates between a deficiency disease and an
ophthalmic infection. 90
According to a communication by A. Marcsik and F. Kosa at the twenty-
fourth International Congress on the History of Medicine (Budapest, 1974),
cribra orbitalia were present on 19 percent of the adult skeletons and 58.8
percent of the infantile ones that had just been exhumed from the Avar
cemetery in Siikosd. The lesions resemble those found on the Nubian
skulls: they do not show the characteristics of a hereditary anemia but
probably result from a nutritional deficiency. By contrast, a medieval skull
from Kiszombor (Hungary) whose racial origins have not been specified
does have modifications on its vault that could be due to some form of
hemoglobinosis. 91 Another lesion that could be confused with symmetrical
osteoporosis of the vault is bilateral thinning of the parietals (the "carpet-
bag skull"). Hrdlicka's terminology has created the possibility of a confu-
sion between porotic hyperostosis and this specific osteopathy, which can-
not have any hematological significance. As an example, I cite the cranial
anomaly of the mummy of Thutmoses III or that of an aged woman who
lived in Berkshire (England) during the Roman period. 92
Within these specifications and limitations, the geographical distribu-
tion of porotic hyperostosis is essentially identical to the combined distri-
bution of the thalassemias and sickle cell anemia, which, in turn, is the
same as the historical zones of strong infestation with falciparum malaria.
To be sure, these distributions are only superposable for the Mediterra-
nean, African, and Asiatic foci. The New World is a separate problem; the
history of malaria in pre-Columbian America is a battleground. 93 It is very
likely that malarial infestation was introduced on that continent only after
the Spanish Conquest. Whatever the case may be, the hematological traits
of American Indian tribes do not entitle us to hypothesize the existence
of hereditary hemoglobinoses in the New World before the Spanish expe-
ditions and the institution of the slave trade.
Since porotic hyperostosis of the cranial vault has no single etiology,
there is no need to interpret similar lesions from different geographical
locations in the same way. The absence of foci of osteoporosis in the
postcranial bones of American skeletons, long ago stressed by Hrdlicka,
points to a hypochromic, nonhereditary anemia that could be due, for
instance, to prolonged lactation associated with nutritional deficiencies, or
to a parasitosis involving chronic blood loss. Several important paleopatho-
logical investigations of the spread of porotic hyperostosis among the pre-
Columbian American Indians have recently been undertaken, above all by
Mahmoud Y. El-Najjar. In his latest study, a sample of 3,361 skulls from
various zones on the American continent served to establish the distribu-
tion of frequencies according to cultural grouping, age, and gender. Sig-
nificant distinctions were reported. They clearly favor a hypothesis that
explains the lesions of the cranial vault or the eye sockets as the conse-
quence of a nutritional lack of iron and proteins combined with attacks
from infectious diseases beginning in childhood. The decisive factor seems
to be an almost exclusive dependence on corn for food and some tradi-
tional ways of cooking it. 94 Currently available data r~fute Zaino's hypoth-
esis, discussed above, according to which the Mayas disappeared because
of thalassemia.
The Osteoarchaeo/,ogy
of Mediterranean PoroticHyperostosis
A diagnosis of thalassemia was suggested by Ignazio Gatto in 1948 for
the bone remains of five individuals dating from the Upper Paleolithic
period that were found by Paolo Graziosi in the cave of San Teodoro near
Aquedolci in the province of Messina (Sicily).95 Skeleton 5, which is not
well preserved, showed signs of osteoporosis on the vertebrae and the
epiphyses of the long bones. Moreover, the general appearance of the

skulls, particularly the enlargement of their zygomatic arches, recalled the
facies that Gatto had noted in parents of his thalassemic patients. So he
concluded that thalassemia had existed among the aboriginal inhabitants
of Sicily even prior to the Greek colonization of the island. 96
But the lesions reported on the bones from the cave of San Teodoro do
not correspond exactly to the consequences of bone marrow hyperplasia.
The descriptive portion of Gatto 's own article leads one to doubt the basis
of his diagnosis. In 1973, Antonio Ascenzi reexamined skeleton 5 from this
paleolithic site. He discovered that its diploe is perceptibly enlarged but in
a way that suits senescence rather than hemopathy. Radiological examina-
tion gives no indication of anemia. The advanced age of the person in
question is unfavorable to a diagnosis of thalassemia, a disease that is only
rarely survivable. As for osteoporosis in the spinal column and the long
bones, it can be explained as a consequence of old age and arthrosic changes
in the joints. 97
During archaeological investigations at Valle di Treba (1922) and Valle
Pega (1953) in the Po delta, bone remains of the ancient inhabitants of the
Greco-Etruscan town of Spina (fifth to third centuries B.C., approximately)
were brought to light. Enrico Benassi and Antonio Toti reported the
presence on them of stigmata that they attributed to thalassemia: enlarge-
ment of the diploe of the cranial vault, prominent cheekbones, and ''wide
mesh" osteoporosis on the epiphyses of the long bones. Unfortunately,
the osteological material from Spina is exiguous compared with the corre-
sponding archaeological finds, a state of affairs that is as much a conse-
quence of the destructive effects of salt water from the lagoon as of the
underwater archaeologists' mistakes-they are more interested in art ob-
jects than human remains. As a result, the paleopathological data do not
lend themselves to statistical analysis. Relying on historical evidence of the
role of Greeks in the colonization of Spina, Benassi and Toti have inferred
that their paleopathological study is another argument in favor of the
Silvestroni-Bianco hypothesis on the Greek origins of Italian thalassemia. 98
A definite Italian case of porotic hyperostosis was presented by
A. Ascenzi at the First French Paleo pathology Colloquium (Lyon, 1973).
The skull is of an adolescent aged 13 or 14, and it has the exe1nplary
pathological morphology: frontal and parietal bumps, "hair-on-end" X-
ray profile, and typical cribra orbitalia. Unfortunately, it derives from a
relatively recent era: the skull was found in a medieval "cave-church" from
Gravina in Puglia. 99 In order to shed light on the question of the origin of
the thalassemic defect on Italian soil, A. Ascenzi and P. Balistreri per-
formed a systematic examination of the human remains found on archae-
ological excavations in Lucania, a region in Italy whose colonization by
Greeks is well known. In 227 skeletons from various periods (ranging from
the eighth century B.C. to modern times), they found 4 cases of porotic
hyperostosis, 3 of which go back to the High Middle Ages and one to

antiquity. The three medieval cases come from the same locale, Venosa
Trinita. The bone lesions of these three individuals are fairly typical for the
skull, but in the postcranial skeleton, the stigmata that suggest thalassemic
bone marrow reactions are lacking. Of the fourth individual, only an in-
complete, damaged skull remains. It was exhumed from a tomb in the
necropolis of Policoro, ancient Heracleum, which is dated archaeologically
to the third century B.C. The person in question may well be a Greek
settler, but unfortunately paleopathological analysis does not justify a firm
conclusion that he suffered from thalassemia. The lesions on the cranial
vault of this adult male consist only in a spongy enlargement of the parietal
diploe without the characteristic striation of the outer edge. Ascenzi
draws conclusions cautiously from his investigation: given the present
state of knowledge, paleopathology does not elucidate the problem of the
origin of Italian thalassemia. To progress in this field, new research is
necessary, and it must take account of all parts of the skeleton and con-
sider the possibility of establishing family links between subjects examined.
It must give the greatest attention to the bone remnants of chil-
dren.100
Following Ascenzi's recommendations, Gino Fornaciari and Francesco
Mallegni surprised specialists by identifying traces of thalassemia on bones
from an Etruscan region. During excavations at San Giovenale near Tar-
quinia in 1958, Swedish archaeologists discovered a group of tombs from
the third century B.C. containing three well-preserved skeletons. On two
of these (one of a woman aged 16 or 17, the other of a child around 5),
researchers from the Center of Paleopathology of Pisa reported porotic
hyperostosis on the cranial vault, cribra orbitalia, and osteoporosis of
the vertebrae and the long bones. The details of the lesions, especially
in the young woman, correspond perfectly to a diagnosis of thalassemia
major. 101
We owe the discovery of porotic hyperostosis in Greece to John
Lawrence Angel, an anthropologist at the Smithsonian Institution. In his
report on the first results of an anthropometric study of skeletons ex-
humed by J. L. Caskey's team at Lerna in the ArgoFd, Angel states that
"the problem of cranial osteoporosis is still unsolved" and reports, as a
puzzling and relatively frequent syndrome, combined porosities on the
inner surfaces of vault bones. 102At first, the nutritional origin of this lack
of proper ossification seemed fairly plausible. In the Scandinavian coun-
tries, internal porosities in the skull had been described and interpreted as
signs of malnutrition. Angel also set out on that route but took a different
turn. Although in 1964 he still has a question n1ark in the title of an article
on thalassemia in Greece, 103the hesitation is missing in a newer publication
on the same subject (1966), 104and certitude marks his monograph on the
ancient population of Lerna. Porotic hyperostosis on the skulls of Greece,
according to Angel, is a sign of thalassemia or sickle cell anemia, and

therefore its frequency is a precious index of the level of malarial infestation.
That is also the way he understands the pores and bumps on the inner
table: they are the osseous expression of an initial overabundance in new-
born homozygous thalassemics. 105 However, the overall progress of
knowledge in this domain and an enriched personal experience led Angel
to retreat somewhat, or rather to take a more qualified position. In his
latest publications, he admits that an iron deficiency anemia can and even
should be responsible for part of the Mediterranean cases of porotic hyper-
ostosis .106 While recognizing that the current state of paleopathological
research does not enable us to distinguish between a hereditary and an
iron deficiency etiology for these osseous lesions, Angel remains convinced
that the main role belongs to thalassemia, at least as far as the eastern
Mediterranean is concerned. His opinion rests on the diachronic and geo-
graphical correlation between the frequencies of porotic hyperostosis and
the degree of severe malarial infestation, as well as on the statistical prefer-
ence of the lesions for the skeletons of infants.
From 1937 to 1977, Angel examined the osseous remains of 2,334 indi-
viduals (1,750 adults, 584 infants and children) from the territory of the
ancient Greek world (Greece proper, western Turkey, and Cyprus). He
was able to determine topographic and diachronic frequencies of porotic
hyperostosis that seem significant. The severest cases of the ailment are in
infants, except for a few strong reactions on adult skulls from the Neolithic
period. On infants' skulls, the thickening of the diploe is especially signif-
icant at physiological ossification points. Transverse sections of the skull
reveal a complete restructuring of the diploe with radial canals and without
visible traces of the outer table. The sphenoid, maxillary, and malar bones
are often thickened as well, with signs of medullar hyperactivity. As an
example, I can cite the case of a girl about 2 years of age whose facial
skeleton contains spongy tissue in places where it is usually lacking (tomb
no. 133 from Lerna). However, on the ancient Greek bone remains we do
not find modifications comparable in severity to those that occur in certain
modern thalassemic children. In the case just mentioned from Lerna (skel-
eton of tomb no. 133), porotic hyperostosis of the skull is accompanied by
a disturbance in the overall development of the body and by spongy
hyperplasia of the two femurs and several ribs. 107 Analogous postcranial
lesions have been observed on some bones from N ea Nikomedia and from
Cyprus (in particular the ulna of a child about 5 years of age, no. 40-3 CCB
from Bamboula, which shows a thin new cortex and weblike trabeculae
around an "inner shell" representing the unmodeled earlier bone). 108 At
Bamboula, near the Cypriot salt marshes, a Bronze Age tomb was found
containing four typical cases of porotic hyperostosis. 109 It was therefore a
familial disease, either hereditary or due to an external pathogenic factor
that acted selectively on the members of a single family.
In Greece and its environs, porotic hyperostosis seems to have had a

predilection for marshy regions. The most striking example is its presence
in Lerna, on the fertile Argive plain, whose swamps play a role in the
legendary history of Greece. An archaic narrative situated the marshy hab-
itat of the Hydra there, the horrible, seven-headed serpent. Without wish-
ing to follow the reasoning of those who see this monster as the
personification of malaria and Hercules' labor to slay it as an attempt to
sanitize the area, I mention the legend only as an indicator of the impor-
tance of the swamps of Lerna and their dangerous reputation. In the same
region, on the coast of the Gulf of Nauplia, lies the Franchthi Cave, where
2 adult skulls (of 6 examined) and I child's (of 3) bear signs of a hemato-
poietic hyperplasia. These osseous remains belong to the Mesolithic ar-
chaeological layer (about the eighth millennium B.c.) and are among the
oldest examples of such an affection. 110
The frequency of porotic hyperostosis is very high in two Early Neolithic
cemeteries: <;atal Hiiyiik (about 7000 B.c.) in the plain of Konya in Ana-
tolia, and Nea Nikomedia (about 6000 B.c.) in the coastal zone of Mace-
donia, at the time a very marshy region. In the former, the frequency of
affected adult skulls is at 43 percent (of a total of 165 examined); in the
latter, it is about 60 percent (of 45 examined). The first site is significantly
less affected than the second-that is especially dear from a comparison of
the bone remains of infants and children. Of 40 children in <;atal Hiiyiik,
only 2 show the typical signs of porotic hyperostosis and 7 have relatively
slight traces of it (in sum, 22.5 percent); but of 23 children's skeletons from
Nea Nikomedia, 4 have marked signs and 9 slight ones (in sum, 56.5
percent). In adult samples from Nea Nikomedia, there are some especially
severe cases of this osteopathy. <;atal Hiiyiik stands at an altitude of 900
meters, which may well explain why the endemic was less serious there
than at Nea Nikomedia, which lies in the midst of a marshy delta. 111
Porotic hyperostosis is relatively rare and mild on bones from Neolithic
villages established in insular, rocky regions without standing water nearby.
For instance, at Khirokitia, an Early Neolithic site in Cyprus, the rate is at
only n percent, and the cases have slight traces, in a sample of 36; at
Kephala, a Late Neolithic site on the Cycladic island of Cea, the frequency
dips below 7 percent (2 slight cases of 32 examined). 112 During the Late
Neolithic period and the Early Bronze Age, the frequencies of porotic
hyperostosis retreat in all archaeological sites in the Mediterranean, but
the rate always stays much higher where the habitat is flat and humid (for
instance, about 50 percent of the skulls found from this era at Corinth,
near the marshy water channels) than when it is rocky and mountainous
(for instance, 11percent of the skulls in Karata~ in Lycia). 113
During recent excavations at Lerna in the Argolid, all the skeletons of a
necropolis were examined, and for the first time in Greece, not just in
order to obtain values for anthropometric parameters but also for a system-
atic study of pathological modifications. Of 157 skeletons from the main

part of this huge prehistoric cemetery-it belongs to the Middle Bronze
Age, about 2000-1600 B.c.-Angel found 10 cases (6.4 percent) with well-
developed porotic hyperostosis and 22 with slight traces of it. In sum, a
fifth of the population of Lerna seems to have been affected by the disease.
The distribution by gender is not significant; that by age deserves mention:
of 73 adult skulls, n cases of porotic hyperostosis were discovered (3 severe
cases); on 84 infants' and children's skulls, there were 21 cases with marked
lesions, of ,vhich 7 were especially severe. According to Angel, these 7
infantile skeletons (Lerna nos. 10, 61, 71, 103, 133, 136, and 204) "very likely
show results of homozygous thalassemia. " 114
Of a total of 31 Early Neolithic infants' and children's skulls from the
Greek world that can be given proper paleopathological examination, there
are 4 cases (14 percent), according to Angel, that are severe enough to
correspond to the homozygous form of thalassemia. Of n6 infants' and
children's skulls from the Middle Bronze Age, there are 9 (8 percent) that
may be such homozygotes. From these numbers, Angel generated gene
frequencies. Hardy and Weinberg's formula makes it possible to determine
the probable percentage of thalassemic heterozygotes. If we consider that
only a part, at most half, of heterozygous carriers of the genetic defect
suffer anemic disturbances capable of affecting the osseous system, these
numbers show that the percentage of mild forms of thalassemia in adults
should have been about 20 percent for the Early Neolithic populations
and a little less for the Bronze Age. This result is not too distant from real
observations on bone remains from the Middle Bronze Age, especially as
regards the population of Lerna, but it does not at all accord with the
frequencies based on adult skeletons from Nea Nikomedia and <_::atal Hii-
yiik. So Angel concludes that more than half of the Neolithic cases of
porotic hyperostosis must have a nonhereditary anemia as their cause.
Unfortunately, paleopathological examination of osseous lesions does not
yet permit a differential diagnosis between the homozygous and the het-
erozygous forms of thalassemia. This situation and also the still fairly re-
stricted number of skeletons examined combine to deprive this type of
statistical analysis of truly convincing conclusions.
Recently, two German specialists published a minute paleopathological
analysis of the osseous remains of a young woman that were exhumed at
Tiryns in the Argolid in a geometric tomb (900-700 B.c.). Although in
this case there are no traces of cranial hyperostosis, the presence of cribra
orbitalia and of a generalized symmetrical osteoporosis of the postcranial
skeleton fully justifies a diagnosis of thalassemia. 115 For the classical period
of Greek history (about 650-300 B.c.), Angel was able to examine 151adult
skeletons and 30 skeletons of infants and children. In this series, only a
single adult skeleton has the typical signs of porotic hyperostosis, and no
skeleton has well-developed lesions. Slight traces appear in 7 percent of the
adults and 13 percent of the infants and children (some of which have only
cribra orbitalia) . 116 These data, which are to be interpreted either as the
result of a decline in the malarial endemic or as an indication of good
nutrition, suggest a great improvement in the conditions of daily life in
comparison with the prehistoric period.
There is crying need for a systematic search for porotic hyperostosis on
bone remains from ancient necropolises in Yugoslavia, both along the
Adriatic coast and inland, particularly in Macedonia. The results of those
paleopathological investigations, whatever they may be, will have consid-
erable importance for hypotheses on the origin and expansion routes of
the hereditary anemias.
Hemqglobinoses in the Art and Literature of Ancient Greece

Not long ago, readers of the venerable scientific correspondence section
of the British Medical Journal were urged to respond if they knew of med-
ical observations about sickle cell anemia or thalassemia prior to those by
J. B. Herrick (1905). Dr. W. T. Menke reported to his colleagues at the
time that the oldest description of a hemoglobinosis occurs, in his opinion,
in the Hippocratic corpus, in chapter 32 of the treatise Internal Affections. 117
A literal translation of this text follows:
Another sickness of the spleen. It comes on mainly in the springtime and is caused
by the blood. The spleen becomes engorged with blood, which evacuates into the
stomach. Shooting pains in the spleen, the breast, the clavicle, the shoulder, and
beneath the shoulder blade. The body's coloration resembles lead. Sores form on
the leg and become large ulcerations. The discharges with the feces are bloody and
bluish green. The belly hardens and the spleen is like a stone. This one is more
murderous than the one before, and few survive it. 118
According to Menke, this passage refers to thalasso-drepanocytosis, that

is, to the clinical expression of the chromosomic state that typifies the
simultaneous presence in a single individual of the genes of thalassemia
and sickle cell anemia. In fact, the clinical picture provided by a Greek
physician of the classical period does resemble certain forms of thalasso-
drepanocytosis (or, perhaps, a ho1nozygous form of sickle cell anemia), 119
and actually there is no reason to oppose such a diagnosis. However,
several other diseases can furnish the same symptoms as those in the Hip-
pocratic account. This text can be considered an indication of the existence
of hereditary anemias in ancient Greece, but its ambiguity prevents us from
using it as a proof of the fact. Sickle cell anemia once seemed to me
recognizable in a passage from the Epidemics (book 7, chapter 52), in which
the Hippocratic writer describes two cases of a fatal disease of children
marked by abdominal pains, thinning down, and a strange affection of the
cranial vault. On reflection, I would now explain this observation by a
diagnosis of hereditary xanthomatosis.
There is also a third passage in the Hippocratic corpus that could relate
to hemoglobinosis find deserves mention in this context, namely, section
332 of the Coan IYenotions: '' At the age of 7, weakness and discoloration,
labored breathing while walking, and a craving for dirt herald corruption
of the blood and its resolution. " 120 An anemic syndrome seems to lurk
behind these lapidary phrases. Is it a hereditary one, specifically, a form of
thalassemia? The geophagy recalls instead a secondary anemia, a deficiency
disease. One called pica provokes a perverse craving for inedible substances,
especially soil. Modern studies have shown that it is associated with dietary
iron deficiency. 121 The link is etiological, in that either iron deficiency
anemia is expressed by such behavior, or because eating everything pro-
duces the deficiency in question.
Artworks provide an unhoped-for argument in favor of the presence of
thalassemia in ancient Greece. While examining terracotta figurines from
the Hellenistic era in the Louvre for a study of the artistic representation
of diseases that I have begun with Danielle Gourevitch, I was surprised to
come across some children's heads with mongoloid facies and the stigmata
of thalassemia. 122 Seven statuettes of this type come from Smyrna; one was
found at Troy. The heads have swollen faces with hypertrophied zygo-
matic arches and symmetrical protuberances on the fronto-parietal parts of
the skull, and the base of the nose is crushed. They resemble to an extraor-
dinary degree the heads of homozygous thalassemic children in Greece and
Turkey nowadays. Several other specimens of Smyrnean sculpture prove
that the artist's goal was the sculptural reproduction of an actual patholog-
ical state, not an imaginative search for the "grotesque."
The Evolution of Malaria in the Eastern Mediterranean

In studying the morbidity of ancient populations, it seems reasonable
to admit as a working hypothesis that a correlation exists between the
frequency of cranial porotic hyperostosis and the degree of malarial infes-
tation. The correlation may result from relationships between malaria and
hereditary hemoglobinoses as well as more complex relations between the
conditions of daily life, secondary anemias, and the malarial endemic. The
results of J. L. Angel's research are summarized in the accompanying table
(Table 2), which gives the diachronic distribution of the frequencies of
porotic hyperostosis on skeletons from Greece and Asia Minor. Admitting
that this distribution of frequencies reflects, if only in crude terms, the
evolution of malaria in the eastern Mediterranean, one is struck by the
existence of three periods of hyperendemic: the Neolithic period, the
Roman era, and the time of the Turkish occupation and the beginnings of
Greek independence. The corresponding low points are first in the Paleo-
lithic period, then toward the end of the Dark Ages (that is, right before
the so-called "Greek miracle"), and the Byzantine period.
TABLE 2. Frequenciesof PoroticHyperostosisin the Greek World
Adults Infants and Children Total
Approx.
Chronological Typical No. Cases Typical No. Cases No. Cases
Period Limits Traces* Lesions* Examined Traces* Lesions* Examined Traces* Examined
Paleolithic 2** - 50 - - - 2 50
9000
Mesolithic 17 16 6 33 0 3 33 9
7000
Early Neolithic 33 10 165 25 9 75 40 240
5000
Late Neolithic 18 3 63 9 5 22 19 85
3000
Early Bronze Age 11 1 332 7 5 163 11 495
(Doric invasion) 2000
Mid. Bronze Age 11 1 169 16 6 148 17 317
1500
Late Bronze Age 8 1 215 10 1 81 9 296
1100
Early Iron Age 6 1 114 16 0 51 10 165
650
Classical period 5 1 151 13 0 30 7 181
300
Hellenistic period 12 1 138 22 0 9 14 147
A.D. 150
Roman period 24 1 100 - - 3 24 103
600
Byzantine period 10 2 87 8 8 12 13 99
1400
Turkish occupation 45 2 53 - - 3 45 56
1800
Nineteenth century 36 1 200 - - - 37 200
1900
SOURCE: J. L.
Angel, "Porotic Hyperostosis in the Eastern Mediterranean," MCV Quarterly14 (1978): 10-16.
*These figures represent percentages.
**These are dubious cases.
Looking at the distribution of these frequencies over time, it is not just
the highs and lows that are arresting, but also the historical moments when
a rise begins. These are the points at which the equilibrium of the patho-
coenosis is broken, and they must be grasped and explained. According to
the table, then, malignant forms of malaria invade in the Mesolithic pe-
riod, and again during the classical period, and finally there is a recrudes-
cence of the endemic after the fall of the Byzantine Empire. For historical
times, this overall reconstruction of malarial infestation in Greece agrees
with conclusions that can be drawn from literary evidence. For the oldest
periods, it is in harmony with paleoparasitological and anthropogeograph-
ical considerations.
Malaria is a disease that left its mark on the history of Greece and Italy.
It was certainly responsible for severe economic, social, and even political
disturbances. A portion of Hippocratic medicine (for instance, the theory
of critical days) can be understood only in terms of the prevalence of
malaria in the classical Greek pathocoenosis. But malaria is not a disease
native to the Balkans or the Italic peninsula. Its cradle was very likely
tropical Africa. According to L. J. Bruce-Chwatt and P. F. Mattingly, the
equatorial zone of Africa more than any other place on earth favored
contact between hominids and the insects that are vectors of the various
species of Plasmodium. 123 At first, these microbes could not have been
exclusive parasites of humans, since small groups of isolated hunters could
not have assured their survival. The first human malaria persisted only
because there was a reservoir of hematozoa among other primates. It is
possible that during a large part of the Paleolithic period, the disease
sustained itself only in a very few foci of tropical Africa and the Far East.
There are good reasons for believing that malarial infestation radiated from
Africa toward Mesopotamia, and that it propagated along the Nile Valley
up to the banks of the Mediterranean. Climatic changes, a demographic
surge, and the beginning of agriculture combined to favor that
propagation. 124
It is important to keep in mind that malaria is not etiologically a unitary
disease. Its history is complicated by the fact that several different species
can cause it. They all belong to the genus Plasmodium) but are not identical
morphologically, nor do they provoke the same symptoms, nor do they
all necessarily have the same ancient pedigree. Humans nowadays are af-
fected by four species of parasites: Plasmodium vivax (causal agent of vivax
or benign tertian malaria, which predominates in temperate climates), Pl.
malaria,e (agent of quartan malaria), Pl. ovate (agent of ovale malaria, a
benign tertian form that is relatively rare and confined to tropical Africa),
and Pl. falciparum (agent of falciparum or malignant tertian malaria). Spe-
cialists in the biology of Haemosporidia, the group to which the genus
Plasmodiurn belongs, believe that Pl. vivax and Pl. malaria go back to the
Pliocene, a period of the Tertiary era, where they lived in some kind of
association with the ancestors of humans. These species, or rather their
ancestors, must have been relatively well adapted to severe climatic condi-
tions. The affection they caused was of long duration, and they were able
to survive despite the small size and wide dispersion of human popula-
tions. By contrast, Pl. falciparum) which is more sensible to cold and more
devastating to its host, may be a more recent mutation, dating from the
last, interglacial phases of the Pleistocene era. 125
Only falciparum malaria can be related to porotic hyperostosis. A Pale-
olithic skull with a significantly enlarged diploe was found in the Petralona
Cave in Chalcidice. 126 A diagnosis of thalassemia is far from certain, and
even if it were, that would prove nothing about the antiquity of falciparum
malaria, since the anemic mutation in humans must have preceded the
one in Plasmodium. Other cases have been described on bone remains from
an epi-Paleolithic necropolis at Taforalt in Morocco. 127 Their significance
is unclear. In the Greek world, porotic hyperostosis assuredly appeared in
the Mesolithic period and attained remarkable frequency in the beginning
of the Neolithic period. The hypothesis that the eastern Mediterranean
was invaded by falciparum malaria at the time can benefit from geological
data on a rise in temperature and archaeological evidence on the beginnings
of agriculture and sedentary life in low-lying, moist regions. Pl. vivax and
Pl. malariae need an ambient temperature of at least 15° C to achieve
sporogeny in hematophagous mosquitoes; Pl. falciparum must have a tem-
perature higher than 18-19° C to survive. In the Paleolithic period, the
stages of glaciation must have eradicated them from the European conti-
nent, if in fact they had gotten that far. But before the end of the Pleisto-
cene and the beginning of the Holocene, that is, around 8300 B.c., the
average temperatures in southern Europe rose to their current values. The
expansion of agriculture in Greece during the eighth century B.c. may well
have fomented malarial infestation, not just by creating conditions that
were especially favorable to the transmission of its pathogenic agents, but
also simply because of the demographic upswing and the concentration of
persons in fertile, that is, hot and humid, zones. It is highly likely that
at that time vivax and quartan malaria extended .their endemic zones
to the north, and falciparum malaria began laying waste to the Aegean
basin.
Though he realizes that climatic changes taking place on the verge of
the Neolithic period made for the spread of malaria into a large portion of
Europe, Julian de Zulueta tnaintains that all during the Neolithic period
and the Bronze Age the number of benign instances of this disease in
Europe and particularly in Greece was relatively small and that its malig-
nant form could not survive there. His argument is based on some traits
of Anopheles mosquitoes, vectors of the disease, and their distribution in
Europe. Experiments by P. G. Shute, C. Ramsdale, and de Zulueta hin1-
self have shown that the hematophagous species that is most important
for the maintenance of the malarial endemic in southern Europe, Anopheles
atroparvus) resists infection by the tropical strains of Pl. falciparum. 128 The
preferred vectors of malign tertian malaria are A. labranchiae and A. sacha-
rwi. Although it is true, says de Zulueta, that these two species still exist
today along the northern banks of the Mediterranean, they did not reach
the Aegean basin, Italy, and Spain until after the massive deforestation of
these lands in the Hellenistic and Roman periods. Current distribution of
A. sacharwi) he suggests, is the result of relatively recent diffusion by way
of sea trade. So prehistoric Europe suffered only a little, if at all, from
malaria, and it was completely spared the terrible ordeal of a falciparum
endemic. Even the presence of thalassemia in preclassical Greece is called
into question by this argument. 129
Actually, we know nothing about the presence or the absence of the
species A. labranchiae and A. sacharwi in the prehistoric eastern Mediter-
ranean. It is altogether possible that the rise in temperature that character-
ized the end of the glacial age and took effect well before the flowering of
Neolithic civilization incited the immigration from North Africa of A.
sacharwi and A. superpictus) both of which were perfectly capable of trans-
mitting Pl. falciparum. But there is no need to make this supposition, since
A. atroparvus) or even A. superpictus) could have served as the vector of
falciparum malaria in those days. Attempts to infect A. atroparvus with
current Indian and African strains of Pl. falciparum regularly fail, but its
infection with Italian or Rumanian strains easily succeeds. All this points
to the split between tropical and European strains of Pl. falciparum and the
antiquity of their geographical segregation from one another.
The disease that resulted from the first contact of the peoples of prehis-
toric Greece with this especially virulent germ must have been murderous.
That helps to explain the astonishing success of the thalassemic gene,
which offered an unhoped-for protection to its heterozygous carriers. Be-
ginning in about 5000 B.C., the frequency of porotic hyperostosis and
probably as well the degree of malarial infestation steadily diminish. And
this improvement in public health was not interrupted, as it might have
been, by the influx of new peoples, by destructions and political distur-
bances. The retreat of malaria seems to have been linked to a significant
drop in sea level, to the drying up of wetlands, to the perfection of agri-
cultural techniques, and to more judicious selection of ploughland. It was
a complex event some of whose main features still elude us. In any case,
Mycenaean civilization apparently did not suffer from malaria, and several
intersecting clues suggest that at the beginning of the archaic period, falci-
parum malaria had completely disappeared from the majority of Greek
lands, leaving the field open to the benign forms of the disease.
In a series of remarkable studies, William Henry Samuel Jones (1876-
1963) took advantage of new ideas formulated by specialists in the epide-
miology of malaria (he was especially inspired by the work of Sir Ronald
Ross and Angelo Celli) and of his profound acquaintance with ancient
Greek literature to validate the following two theses:
r. In early times, malaria, if it existed at all, was sporadic, rare, and not
severe in its clinical manifestations; it spread into Greece only after 500
B.c., during the classical period-in Attica, for instance, it was introduced
in about 430 B.c.
2. The malarial endemic "fell like a blight upon many fertile districts of
Greece"; it ruined health and changed people's nature, depopulating lands
essential for economic survival and, from the beginning of the Hellenistic
period to the beginning of the twentieth century, it never ceased raging
on; it was the decisive factor in the fall of classical civilization. 130
The first of these is based on the silence of literary sources prior to the
fifth century B.c. and on some general considerations. Jones maintains,
for instance, that the principal Greek city-states are situated in zones that
were especially exposed to malaria, which proves that their sites were
chosen before the existence of the malarial endemic. Hesiod seems igno-
rant of the disease. 131 Between him and Theognis, that is, from the eighth
to the sixth century B.c., the silence of literary witnesses is total. 132 Theog-
nis, a didactic poet from Megara, mentions in passing the term ep(a/,os 'fever
with chills,' which in most later texts refers to a bout of malaria. 133 There
is an allusion to malaria in Herodotus, but not on Greek territory. Aris-
tophanes is the first figure in nonmedical Greek literature to speak clearly
of fever accompanied by chills as a significant event in daily life. In the
Wasps)a comedy composed in 422 B.c., it is spoken of on three occasions.
There is another mention of it in the Archarnians (about 425 B.c.). 134 Jones
concludes that the Athenians of the last quarter of the fifth century B.C.
had special reasons for being interested in malaria, since it was a relatively
new disease in Attica. 135 He asserts that after Aristophanes and notably at
the time Plato was composing the Timaeus) the term puret6s designated a
paroxysm of malarial fever (that is, not fever in the general sense) in the
ordinary language of Athenians and in almost all nonmedical literature. 136
The introduction of the cult of Asclepius in Athen~ near the end of the
fifth century could also betoken the deterioration of public health, says
Jones.
This first thesis of his was well received and has been widely dissemi-
nated. It now constitutes what some handbooks consider a secure histori-
cal truth. However, the opposite view has also been vigorously defended
by several experts. Its main partisans are physicians with actual experience
of the ravages of malaria in Greece. 137 For them, this disease is especially
suited to that land and must have always flourished there: it arrived when
people did. However, although malaria existed in Greece from the most
remote times, it did not necessarily manifest itself with the same intensity
throughout history. In prehistoric times, it is admitted that its scope was

restricted; later on~ certain factors provoked endemic upswings that en-
larged infestation zones and made the clinical picture more severe. 138 By
supposing that one of these upswings began in the classical period, the
historical and philological arguments of Jones are rehabilitated. What re-
mains is to account for a hyperendemic at this critical moment in Greek
history. In Jones's perspective, everything is explained by the arrival of the
Anopheles mosquito and of a specific hematozoon on virgin soil. Those
who believe in the great antiquity of malaria in Greece instead see the cause
either in ecological factors (especially deforestation 139 by human hands or
a change in climate linked to great geophysical cycles140 ) or in factors
related to the biology of the germ that cause it to undergo cyclic variations
of its virulence. 141
What I have said above is enough to show why nowadays we must find
more subtle and complex explanations that combine and supersede both
the traditional form of the hypothesis about the constant presence of
malaria and the simple notion of its introduction de novo.In any event, we
have no choice but to treat the problem in a wider framework than that
of Greece and to consider the plurality of hematozoic parasites, their bio-
logical history and that of their vectors, human migrations, archaeological
data, and finally, recent discoveries in paleoclimatology and the physics of
the planet. Most of Jones's conclusions still evoke our assent, but on
condition that his hypothesis of the first introduction of malaria into
Greece be replaced by that of the reintroduction of falciparum malaria.
Jones neglected prehistory and did not take into account the difference
between the various malarial germs.
There can be no doubt as to the presence, at the time of Hippocrates,
of vivax, quartan, and falciparum malaria (the bilious remittent fever as
well as the invasive pernicious form). Writings in the Hippocratic corpus
are imbued with clinical experience based on the observation of patients
with these diseases. Malarial cachexia is well described in some treatises
and, especially in Airs) Waters) and Places) causally related to stagnant
water. Notice is taken of the wizened appearance and the intellectual and
moral weakness of those who live near swamps. The author of the Aristo-
telian R--oblemamnotes that they age prematurely. From the end of the
fifth century, malaria is the disease par excellencein the Greek world. But
its frequency and the variety of its clinical manifestations obscure its noso-
logical unity. It is made into a host of different diseases. Associated with
typhoid fever and dysentery, malaria is the source of clinical pictures that
are both varied and impressive.
We know now how a malarial hyperendemic can mark and transform
the pathology of a whole population. That is what started to happen with
Greek patients in the classical period; it was surely an essential characteristic
of the Hellenistic pathocoenosis. 142 But it is not easy to determine at what
point in Greek history malaria took this turn for the worse. Was it just at
the birth of Hippocratic medicine? Was it one or two centuries earlier? For
the moment, the answer to these questions cannot profit from paleopatho-
logical research, given the low number of skeletons from the period that
have been examined and their imprecise dating. Even if we assume that
the frequency of porotic hyperostosis gives a good approximation of the
degree of malarial infestation, it must be admitted that a substantial time
lag should separate these two pathological phenomena. Arguments from
immuology seem to me to favor the idea of a reintroduction of falciparum
malaria shortly before the composition of the first book of the Epidemics.
In addition to the philological and historical clues proposed by Jones, I
can point to the severity and peculiarity of several cases of malaria described
in the Hippocratic corpus. For Athens, there is the surprising way events
unfolded during the Sicilian expedition. 143 Falciparum malaria occasionally
took the form of a pestilential disease, and it is not impossible that its
commonness contributed to the triumph of the concept of miasma over
that of infection; it also may have been responsible for the success of
fumigations. Without actually "purifying" the air, fumigations do get rid
of the mosquitoes. The introduction of black bile into the system of the
cardinal humors may have been due to clinical experience with malignant
malaria, too. Jones draws attention to the role that the neglect of irrigation
and other agricultural works during the Peloponnesian War may have
played in the diffusion of malaria throughout Attica. 144 Moreover, the
Sicilian War could have been responsible for the transmission of a virulent
form of Plasmodiumfalciparum from Syracuse to Athens.
Once it had developed during the classical period, the falciparum infes-
tation took on amplitude during the Hellenistic period and reached its
apogee during the Roman era. For Jones, malaria depopulated Greece and
made its inhabitants morose, pessimistic, and apathetic. 145 The Periclean
Age did not keep its promises of a brilliant and triumphant future, but the
fault for that lay in a disease that broke the life force of the Greek people.
Pausanias seems to have voiced the same sentiment, without, however,
making clear to what disease he was ascribing the failed power of the Greek
city-states. ,
It is certain that a hyperendemic of malignant malaria can have cata-
strophic consequences for the birthrate, death rate, and overall vigor of a
people. Contemporary medical geography offers dazzling proofs of that.
Nevertheless, Jones's second thesis was heartily criticized by those who
believe that the economic, demographic, political, and moral degradation
of Greece is entirely due to social factors. For them, the expansion of the
malarial endemic is more an effect than a cause of social disorder: the
decline of city life brought on the disorganization of agriculture, the exten-
sion of swampland, the multiplication of mosquitoes, and thus the recru-
descence of malaria. 146 In any case, a vicious circle links malaria,
depopulation, poverty, and political disorder. Once the process has started,
it matters little which element came first. Deforestation, the decline in
agricultural techniques, and the neglect of hydraulic works were all impor-
tant factors in the diffusion and intensification of malaria from the begin-
ning of the Hellenistic era. There was also the rising level of the sea and
the change in climate, which was becoming ever more hot and dry. How-
ever, all the factors just mentioned came into play after the fifth century
B.C. They cannot, therefore, account for the first recrudescence of malaria
in the time of Hippocrates, but only its subsequent extent.
Chapter Eleven
THE HIPPOCRATIC CONCEPTION

OF DISEASE
An Exemplary Clinical Report
The seven books in the Hippocratic corpus called Epidemicsinclude several

series of clinical descriptions. 1 These lists of ancient patients, with the dry,
precise descriptions of suffering they entail, are of priceless historical inter-
est, since they have.served as models for generations of practitioners. Few
nowadays have direct knowledge of the clinical histories of Hippocratic
patients, but it remains true that all doctors know them and benefit from
them indirectly via the traditions of medical education. In their terminol-
ogy, logical structure, standard analytic procedure, and effort to escape as
much as possible from doctrinal underpinnings, the clinical observations
stored in the Epidemicsare still a model of the way in which the Western
physician looks at a patient and reports his findings.
The Caseof Philiscus

Here is the first case in the list of fourteen patients that constitutes a
kind of appendix to the first book of the Epidemics.2
Philiscus lived near the wall; he went to bed; the first day he had an acute fever,
sweating; a miserable night (r). 3
Second day, an overall worsening; in the evening, after an enema, good bowel
movement; a calm night.
Third day, in the morning and until the middle of the day, he seemed without
fever; then, toward evening, acute fever, sweating, thirst, and his urine was black
(2); miserable night, completely restless (3), confused about everything (4).
Fourth day, generally worse, black urine, a better night, urine color improved.
Fifth day, around midday, a slight dripping of unmixed blood from the nose (5);
THE HIPPOCRATIC CONCEPTION OF DISEASE 285
varied, irregular urinations, containing rounded, floating particles in suspension
resembling sperm; nq deposit. After a suppository, a little excrement with wind
(6). Miserable night; cat naps; talking, rambling; extremities everywhere cold,
impossible to get them warm again; black urine; brief drowsiness near dawn; no
voice; cold sweats; extremities livid.
He died around the middle of the sixth day. Toward the end, deep, infrequent
breathing, as though he was trying to recall it (7). The spleen stuck out, forming a
rounded swelling; cold sweats through to the end. The exacerbations on even
days.
Before proceeding with a modern medical interpretation of this clinical

case, some details of my translation need to be justified. Also, its spatial
and temporal coordinates qua historical account need to be specified as
well as its significance in the context of ancient medicine.
I have translated the Greek text as it was established by Hugo Kueh-
lewein4 and revised by W.H.S. Jones, 5 but I have also taken into account
Littre's apparatus criticus and Galen's commentaries. 6 Translation of an
ancient text into a modern language is a work of approximation that
necessitates interpretation and choice. The better to grasp the various ways
in which the details of this Hippocratic text may be understood, I have
consulted, besides the esteemed bilingual edition of Emile Littre,7 several
other French translations, 8 some German 9 and English translations, 10 and
an Italian translation as well. 11 My version partly reflects the clipped and
elliptical style of the original. The Hippocratic writer artfully joins precise
content to concise form. Translation into a modern language requires
more words, either to say the same thing as the Greek or in order to
interpret and clarify it. An explanation of several details follows:
1. In the description of the initial stage of Philiscus's disease, the punc-
tuation of the Greek text varies from one manuscript to the next, which
allows for different interpretations of the sequence of symptoms. I have
followed the punctuation of Littre that agrees with the way Galen presents
the events in his commentary, as follows: "The fever was acute on the first
day; then came sweating, which in no way arrested it; the night was
miserable. A new exacerbation arrived on the second day.''
2. For a medical evaluation of this case, it is especially important to
determine the exact meaning in context of the word milana) used three
times to designate a specific appearance to the urine. I have translated oura
milana as "black urine," in agreement with the accepted interpretation
(Foesius, Littre, Daremberg, Fuchs, Jones). However, the adjective could
also have a vaguer meaning and imply grayish or just dark-colored urine.
Such is the opinion of Muri and Diller, who translate ''Harn dunkel,''
like Martiny's "urines foncees." In this instance, it seems to me an exces-
sively and unjustifiably cautious translation. There is such a thing as black
urine. Since the most common meaning of the Greek adjective is medically
possible, it is in principle preferable to secondary meanings. On most
occasions when the Hippocratic writers speak of ''black urine,'' they are
referring to hemolytic syndromes. 12 Returning to the case of Philiscus, I

note that the translation "black urine" seems inevitable, partly because of
its consistency with the overall clinical picture, and partly because of one
of Galen's remarks. According to his commentary, the appearance of
"black urine" on the third day is a fatal sign that betokens the seriousness
of Philiscus's disease and allows one to predict rapid death. It would be
hard to admit that such a statement refers to urine that was only dark-
colored.
3. Despite their authority
,,
and number, I have not followed the scholars
who translate ouk ekoimetheas "did not sleep" or "without sleep" (Littre,
Fuchs, Jones, Diller, and others). A lack of rest is not the same thing as an
absence of sleep, as is well shown by Daremberg in his commentary on the
passage. 13
4. To account for the psychic disturbances that the patient suffered on
the third day, Littre wrote that he "eut des hallucinations sur toute chose"
[hallucinated everything]. In my opinion, that goes beyond the Greek
expression panta parekrouse.From a medical standpoint, it is clear beyond
doubt that the ramblings that often come up in Epidemics)I and III, have
nothing to do with psychotic disturbances. They are states of confusion
with toxic origins, that is, mental disturbances or delirious behaviors that
take place under high fever (for instance, during a malarial paroxysm or in
the acute phases of salmonellosis or shigellosis). Hallucinations are rare
among them, and besides, they are always rooted in states of mental con-
fusion. Littre's translation (or Fuch's "Hallucinationen") does not agree
with the medical interpretation and can be justified neither by the etymol-
ogy of parekrousenor by its usage in other classical texts. Other, more
satisfactory translations have been suggested, such as '' divagations de toute
sorte" [all sorts of ramblings] (Bourgey); "delire complet" [utter delirium]
(Martiny); "delirious on all subjects" (Adams); "completely out of his
mind" (Jones); "vollig besinnungslos" [totally insensate] (Muri); "ganz
von Sinnen" [altogether out of his senses] (Diller); and "perse del tutto la
ragione" [deprived of all reason] (Vegetti). I have opted for "confusion
about everything," which keeps to the relatively vague meaning of the
Greek. '
5. On the fifth day, Philiscus had, according to Littre, ''a small epistaxis
of very black blood," or on my interpretation, a "slight dripping of un-
mixed blood from the nose." I did not wish to introduce into the trans-
lation a Greek technical term (epistaxis) in a place where the Greek author
himself avoids it and says instead apo rhinfJn estaksen 'dripped from his
nostrils.' Other than this small formal difference that I mention only in
passing, there is a divergence semantically concerning the quality of what
drips from the nose. The hemmorhage is said to be akreton. At first sight,
the sense seems clear: "unmixed blood," that is, "pure," without any-
thing else mixed in. P. Berrettoni 14 remarks that from Homer onward this
THE HIPPOCRATIC CONCEPTION OF DISEASE
word is usually used to designate substances such as wine not diluted with
water (see in this regard Epid.) vr, 6, 7). In a medical context, it may take
on a more technical meaning and relate to a "raw," that is, undigested
state of fecal matter or humors. But Littre instead argues for a different
translation, citing Galen, who recommends that akreton here be translated
"dark black." I do not accept this explanation, at least on the level of
translation. The only thing that can be said concerning this passage in
Philiscus's history is that "unmixed blood," especially without a mixture
of "breath" or of bile, looks like "black blood." However, we have to be
very cautious about such assertions: an anachronistic grafting of Galenic
physiology onto Hippocratic clinical description can easily lead to histori-
cal misinterpretation.
6. The administration of a suppository on the afternoon of the fifth day
of the disease produced the evacuation of phus~dea smikra. I have translated
these words "a little excrement with wind." Bourgey speaks of "weak
emission of gas." In the translation of a text from antiquity, is it permis-
sible to use the technical term "gas," which was invented in the seven-
teenth century by J.B. van Belmont to denote a concept not formulated
until his time? This term carries with it connotations that are alien to the
pneumatic chemistry of antiquity. Albeit hesitantly-I am inhibited by
notions of purism in scientific terminology-Bourgey's translation of
phus~dea seems to me perhaps acceptable even though I believe the work
more probably implies flatulent stools rather than flatulence without either
solid or liquid stools. In regard to this same passage, Diller is surely mis-
taken when he speaks of blood loss accompanying defecation (' 'Stuhl
unter Blutungen"). The introduction of intestinal hemmorrhage into the
clinical history of Philiscus's disease would be very important in a discus-
sion of the modern diagnosis of this case if it were based on a valid histor-
ical or philological argument. But Diller's text probably results from a
typographical error, with the correct version being "Stuhl unter
Blah ungen. ''
7. Toward the end of this account, the Hippocratic writer notes that
Philiscus's breathing was "deep and infrequent, as though he was trying
to recall it." Deep and infrequent breathing immediately suggests Kuss-
maul's dyspnea, but the expression h~speranakaleomeno indicates that it is
rather paroxystic dyspnea of the Cheyne-Stokes type, a deep and noisy
respiration that gradually diminishes and gives way to an apnea that c·an
last as long as 20-30 seconds. During the apnea, one has the impression
that the patient has forgotten how to breathe; then he catches his breath,
as though by a conscious effort.
According to Galen's comments, 15 which are cited by Foesius and Littre,
the word anakaleomeno in this context should be taken in the sense of the
verb anamimnlskomai 'remember, recall,' and the phrase translated
"comme si le malade se souvenait de respirer" [as though the patient was
re1nembering to breathe] (Littre), "comme chez quelqu'un qui ne respire

que par souvenir'' [like someone who only breathes when he remembers
to] (Daremberg), "as though he were recollecting to do it" Gones), or
''wie einer, der sich immer darauf besinnen muss'' [like someone who
always has to remember] (Diller). This interpretation, in my opinion, does
not deform the clinical reality of what is meant on the level of observation,
but it still colors it by means of a dubious analogy. Galen's remarks are not
self-evident. There is no difficulty in dispensing with the notion of "re-
membering," since the metaphorical description of Cheyne-Stokes breath-
ing is all the more striking if we stick to the usual meaning of the verb,
namely, to bring back someone or something. 16 Classical authors used this
term to denote an invocation, a determined effort to summon someone
or something (such as a dead soul from Hades). So I have tried to express
in my translation the idea that the patient recalled his breath with a special
effort. These two interpretations are in a way complementary. Vegetti
combines them by commenting in a footnote on his translation (''quasi
dovesse richiamarlo" [as though he had to recall it]): the patient "was
forgetting'' how to breathe and had to replace the automatic breathing
process with a conscious effort. 17 Other interpretations of this passage have
been proposed, for instance, "wie bei einem, welcher wieder zur Besin-
nung kommt" [like someone returning to consciousness] (Fuchs), "wie
wenn er wieder zum Leben kame" [as if he were returning to life] (Muri),
and "comme clans quelqu'un qui revient d'une defaillance" [like someone
reviving from a blackout] (Martiny). To me they seem unacceptable.
The Placeand Date of the Diseaseof Philiscus

By a fortunate coincidence, the exact place and approximate date of
Philiscus's galloping disease can actually be determined: Thasos, a city on
the island of that name in the northern Aegean, during the last decade of
the fifth century B.c. It is usually accepted that books r and 3 of the
Epidemicsare the most authentic of the Hippocratic documents, that is,
they began to take shape starting in the fifth century under the hand of
Hippocrates (460-377 B.c.) himself or of an itinerant physician belonging
to the Coan guild of Asclepiadae. 18 The results of modern historical and
philological research do not refute the opinion of Galen, who says that
Hippocrates himself wrote the oldest kernel of the Epidemics)especially the
clinical observations and the kamstaseisthat are found in the first and third
books of this "nosographic catalogue." In Hippocratic parlance, a katas-
msis (conventionally translated "constitution") is the climatic and patho-
logical make-up of a whole year in a given place. 19 It denotes distinctive
climatic conditions and a fixed set of prevalent diseases. The first book of
the Epidemicsconsists of three constitutions, all of which refer expressly to
the island of Thasos and to years that are very close to one another and are
probably even successive. After them, the book consists of a series of

individual cases that begins with the case history of Philiscus. The third
book of the Epidemics)closely tied to the first in form and content, contains
two lists of patients and another, fourth constitution, for which no place
name is given. Although there is no rule about the relation between the
sequence of constitutions and the individual case histories, the case of
Philiscus can be attributed with certainty to the third constitution of
Thasos. There are two proofs of it. First, Philiscus is recalled by name as
an example of the patients who died during the third year despite having
had a nosebleed, which was thought to be a sign favorable to the outcome
of an ardent fever. 20 Second, his personal history is probably the basis for
the generalized description of the malignant disease that dominated Thasos
during the year of the third constitution. 21
Philiscus is said to have lived close to the fortification wall ofThasos. As
the result of archaeological research, we know the placement of that wall,
whose reconstruction began around 411 B.C. That date, by the way, pro-
vides a terminus post quem for the arrival of the Hippocratic writer. 22 It is
worth noting, in anticipation of the medical conclusion of my study, that
the place is well suited to the propagation of mosquitoes.
In the catalogue of Thasian theoro() 23 there occurs the name of a certain
Philiscus the son of Aristocleides who, according to Deichgraber, 24 may
well be the same person as the Hippocratic patient. In Thasos, the theorof
were high magistrates who held office for three years. Philiscus may have
done so around 410, probably from 411 to 408 B.C. If this identification is
correct, we must suppose that he succumbed to his disease around the end
of that period or shortly thereafter. The dating of the column of the list
with Philiscus's name is uncertain. Against the opinion of Fredrich ex-
pressed in the edition just cited of the Thasian inscriptions, Pouilloux dates
the column in question to the fourth century B.C. (between 390 and 360). 25
On that hypothesis, it is no longer possible to identify our patient with
the magistrate Philiscus. 26
The Meaning of l(ausos in Hippocratic Medicine
The Hippocratic description of the sufferings of Philiscus is not accom-

panied by a diagnosis; the disease in question is not named. The omission
is intentional. The writer wished to make his account as objective as pos-
sible, so he refused to introduce a name that would be a result of concep-
tual analysis, not just the clinical reality. But it would be incorrect to
conclude from this that Hippocrates did not countenance the intellectual
process of diagnosis and that he forswore the naming of diseases. The
author of this medical record knew that in the ancient system of disease
entities, Philiscus's disease was a particular kind of kausos.That is what he
calls it himself when he uses this case history to describe some traits of the
third yearly constitution ofThasos.
From the viewpoint of modern pathology, what exactly was the fever
that ancient physicians called kausos) or, in Latin, causus?It has become
customary to translate this term "ardent fever" (French "fievre ardente,"
German "Brennfieber"), 27 but that only displaces the problem, since the
expression is not part of modern medical vocabulary. Actually, the Greek
word kausosis not translatable into a modern term because it refers to a
superseded notion that has no equivalent in modern conceptualizations of
disease. ICausosis a clinical entity defined by a kernel of obligatory symp-
toms and a surrounding haze of optional ones. Its fundamental elements
are acute fever that appears abruptly, intense thirst, coated tongue, insom-
nia, transient states of confusion, abdominal distress, and "bilious" excre-
ments. The fever can be continuous or remittent. The patient's body is
"cold on the outside and warm within, " 28 which means that the subject's
sensation of fever is stronger than what the physician feels when he places
a hand on the patient's body. At times the limbs are cold, and they can
become livid. Among optional symptoms, the Hippocratic writers most
often mention the following: nosebleed, mild jaundice, diarrhea, spleen
enlargement, parotitis, bilious vomit, whitish or black urine, and deposits
on the eyes and mouth. ICausostends to transform itself into peripneumonfa.
It does not spare children and can bring on convulsions in them. Although
it is an acute, febrile disease, kausosoften appears sporadically. It is observed
occurring in winter as well as in the heat of summer; the summer and fall
forms can be very severe. 29
This clinical picture is easily recognizable as a form of toxic-infectious
febrile dehydration affecting particularly the central nervous system and
the digestive system. The origin of such a disease is certainly microbial, but
states similar to this one can be brought on by various germs. Several
concrete cases of kausosdescribed in the Hippocratic corpus suggest hypo-
thetical diagnoses 30 that vary from one patient to the next: salmonellosis,
malaria, rickettsial infection, acute food poisoning, puerperal septicemia,
and, less securely, leptospirosis, relapsing fever, and acute appendicitis. So
kausosis not a disease sui generis but a nonspecific syndrome. Renate Wit-
tern is right to say that " attempts to identify it with a definite disease
entity in modern nosology have ended in failure. " 31 At times through
retrospective diagnosis we can recognize the true nature of a specific case
of kausos)but it is methodologically incorrect to persist in making a specific
pathological identification of it when we are dealing with texts in which
this disease name has general value.
From the Middle Ages up to the nineteenth century, physicians thought
they recognized Hippocratic kausosin febrile diseases as varied as malignant
intermittent fever, typhoid, typhus, ephemera, "bilious fever," and,
against the epidemiological· evidence, even plague and yellow fever. 32 In

1803, Jean-Baptiste Germain posed the problem of the modern sense of
kausos very nicely, but the nosography of acute fevers v1as still in such a
confused state that he had to content himself with comparing the ancient
descriptions of it with observations by his contemporaries of diseases that
a modern clinician can recognize as typhoid fever and the other salmonel-
loses. 33 The notion becomes more precise in Francis Adams, who sees
Hippocratic kausosas "the bilious remittent fever of Sir John Pringle" (an
obsolete clinical entity whose essential component was typhoid fever). 34
Emile Littre and Emile Beaugrand compare kausos"with the pseudo-con-
tinuous bilious fevers of tropical lands as described by African army physi-
cians or the British in India" (translated into modern medical parlance,
the reference here is to invasive forms of falciparum malaria). 35 Obstinately
desiring to preserve the nosological unity of kausos) Carl A. Wunderlich
avers that, when all is said and done, no modern disease corresponds to
the clinical picture provided by Hippocrates, so the disease in question
must have become extinct over the course of time. 36
Twentieth-century medical historians were forced to break up the an-
cient concept and differentiate the identifications according to context.
W.H.S. Jones says that Hippocratic kausos designates a group of several
infectious diseases, with typhoid fever leading the way. 37 Georg Sticker
distinguishes between kausosproper and puretos kaus1Jdes: the first is "Fleck-
fieber" (exanthematic typhus, rickettsiosis) and the second "Riickfallfie-
ber" (relapsing fever or borreliosis). 38 Sir William MacArthur asserts that
at least one form of kausos described in Epidemics) I, greatly resembles
modern relapsing fever and that Thasian kausos should also include epi-
demic typhus. 39 Each of the identifications given here has some truth in
it, but the whole undertaking is faulty from the start. ICausosis certainly
the name of a disease for Greek physicians, but in terms of underlying
nosological realities, it is a syndrome with multiple etiologies whose only
unity lies in a common pathogenetic mechanism, namely, a particular
disturbance in the body's water and electrolyte equilibrium.
For ancient practitioners, kausoswas a disease in the strongest sense of
the word. 40 Galen is anxious to insist that it is a disease and not just a
symptom, 41 but that is already evident from some of the Hippocratic texts,
in particular the treatise Affections and the Appendix to the Regimen of
Acute Diseases.42 The author of this appendix firmly believes in the etiolog-
ical unity of kausos)·for him, it is a special improper mixture of bile that
was overheated after having been drawn through the small vessels. This
same physician, or rather the group of physicians who profess the ideas he
is codifying, distinguishes between two clinical forms of kausos)·the distinc-
tion was later adopted and developed by some Greek medical writers of
the Ron1an period and their Byzantine and Arabic successors. 43 Although
at first this subdivision did not affect the unitary etiological explanation,
later authors had a tendency to reduce one of the two forms of kausosto
disturbances in bile and the other to those in phlegm.
At the beginning of the tenth century, Rhazes classified Hippocrates'
kausosamong the tertian fevers, a decision that doubtless reflects his own
experience with malarial infestation in the Near East. 44 That famous Per-
sian physician studied and commented on the Hippocratic text concerning
Philiscus. The translation of it that he gives appears to be a good one. 45
His interpretation of the symptoms tends toward an implicit diagnosis of
an especially brutal intermittent fever. It may be that his decision to classify
Hippocratic kausoswith the tertian fevers was influenced precisely by his
knowledge of the case history of Philiscus.
Hippocraticl+ognosisand Its Relationshipto Diagnosis

It has been said that Hippocrates dismissed diagnosis and was interested
only in prognosis. That method is supposed to be especially apparent in
the "authentic" treatises of the Coan School, to which group the text
under discussion belongs. Thus Monique Vust-Mussard does not hesitate
to declare that
the purpose of Epidernics)I and II, is to teach physicians to form a correct prognosis,
one that will allow for appropriate treatment and will show them how to take
account of the diverse elements at their disposal, how to relate the patient's habits,
the symptoms he/she presents, the disease, the climate ... In fact, as numerous
signs suggest, the orientation of Epidemics)r and III, is heavily toward prognosis.
They make it possible for Hippocrates to reveal what diseases a physician in certain
atmospheric conditions can expect and how regularly such diseases will recur. This
interpretation alone accounts for the specific character of Hippocratic medicine, in
which prognosis has the central role while diagnosis as it is now conceived is not
utilized. 46
This opinion is widespread among current specialists in Greek medicine.

But the idea that Hippocrates consciously wished to dispense with diag-
nosis has been blown out of proportion by 0. Temkin and several classical
philologists who are well versed in textual analysis ,but have no personal
experience of the practice of medicine (Deichgraber, Edelstein, Diller, and
others). 47
There is certainly some truth in the assertion that the author of the
Epidemicsheld prognosis more dear than diagnosis. More precisely, I would
say that he preferred statements in prognostic form to statements in diag-
nostic form. But we should not overestimate the importance of this pref-
erence on the practical level and start to believe that the Hippocratic
physicians actually abandoned the art of diagnosis. The opposition be-
tween these two modes of clinical investigation is clear and complete only
in abstract formulations. It gets blurred once we look at the concrete
intellectual strategies of physician confronting patient. In truth, modern

diagnosis implies pr6gnosis, and Hippocratic prognosis is partly diagnosis
in disguise. In that it offered an immediate way to verify the professional
know-how of an itinerant practitioner, the prognostic art of the Coan
physicians certainly had a specific social usefulness; but from the stand-
point of medical logic, its main role was taxonomic. Prognosis gave the
physician a simple and effective way to distinguish and articulate typologi-
cal regularities in the jumble of a still very crude nosological taxonomy. It
was the way in which Hippocrates sought to differentiate and classify the
complex and confusing clinical reality that lay before him. As a tool for
making subdivisions and nosological comparisons, Hippocratic prognosis
functioned the way diagnosis does for modern physicians. With the help
of his prognostic method, Hippocrates pretended to '' discover beforehand
and, in the presence of the sick, to reveal the present, the past, and the
future" (lt-ognostic)1). Foresight is not the only goal. The Hippocratic
method should enable the physician to make judgments about the pa-
tient's present status and also his past. Essentially, it was a matter of
recognizing the diachronic regularity of diseases.
The subtle logic of the Greek natural philosophers shattered the rigid
framework of the on to logical concept of disease that characterized archaic
medicine and replaced it with a dynamic interpretation of all pathology.
For Hippocrates, disease is not a being but a process. It consists of a
disturbance in the equilibrium of the body's humors, a dynamic upset of
the organism's natural harmony. So disease should be studied and de-
scribed in its temporal dimension. 48 Clearly, from this standpoint, there
are diseases beyond counting. However, they do have structural traits that
make it possible to group them; there are pathological events that recur in
a determined sequence. Prognosis is probably the most flexible way of
grasping their structural regularities in the temporal dimension, so it is the
preferred way, if not the only one. Actually, none of the writers in the
Hippocratic corpus was really able to abandon the classificatory method of
traditional diagnosis.
To perfect his nosology through prognosis, Hippocrates divides the
clinical forms of an acute disease into those that cure it and those that end
badly, that is, in relapse, or by becoming chronic, or in death. In the
course of the disease, he observes events whose importance is not plain to
the uninitiate but that he notes with care so as to relate them to the
outcome. For instance, Hippocrates remarks that in the cases of kausos)
blood loss via the nose is an auspicious sign, while black urine betokens
approaching death. By interpreting the message that these signs relate,
Hippocrates takes into account the whole individual history of the patient,
the chronological sequence of pathological phenomena, and especially the
exact day on which each clinical change took effect. 49 For example, when
a patient sleeps on the stomach, that is a bad sign, but only if it was not
customary before he or she fell ill. Sweats are good or bad depending on
the number of the day in the sequence on which they appear and the
chronological relationship between the sweat and the bout of fever or its
end. In the Hippocratic dynamic vision of nosological reality, signs that
function as precursors, discovered as such through minute clinical obser-
vation, actually play the role of differential symptoms.
Let us consider how our author proceeds to elicit nosological teaching
from the case of Philiscus. For instance, in his description of the third
constitution ofThasos, he writes as follows:
In the beginning of spring, the ardent fevers began, and they continued until the
equinox and through the summer. Now those who fell ill in springtime or early in
summer in most cases got well, though a few died; but when fall came and with it
the rains, the affection turned dangerous, and more patients died. The course of
the ardent fevers was such that patients who bled copiously and properly from the
nose were the most likely to survive. In this constitution I do not know a single
case that proved fatal when proper bleeding occurred. In fact, Philiscus, Epami-
non, and Silenus, who died, had only slight nosebleeds on the fourth and fifth
days, in small quantities, and they died. Most of the patients had chills before the
crisis came, especially those who had not suffered a bleeding. 50
For the Hippocratic physician, one feature of the case history of Philiscus
was surprising and instructive, because it seemed to contradict a general
rule: in the course of the disease, he had an epistaxis and despite that
favorable sign, he died. The physician sought for and found an explana-
tion: the epistaxis in a patient sick with kausosis auspicious only when it is
copious and occurs on a favorable day. Instead of falsifying the rule, obser-
vation of this case confirms it, but the price is a restrictive modification of
the rule.
In a third stage of medical reasoning by the author of the Epidemics)a
model description of the disease detached itself from a series of similar
clinical reports. This changeover from the description of a concrete case
associated with a patient's name and address, to a syndrome that is so
generally conceived as to dispense with comments about individuals, can
be illustrated by a comparison between the case history of Philiscus and
the description of the autumnal form of kausosin the third Thasian consti-
tution. Here is the Hippocratic passage: ·
The patient affiicted with ardent fever showed early signs that indicate a fatal issue.
From the start there was acute fever with slight chills; insomnia, thirst, nausea,
some sweats on the forehead and the collarbones, but never on the whole body;
frequent nonsensical talk, fears, depression, very cold extremities, toes and hands,
especially the latter. Exacerbations occurred on even days. In most cases, the pain
was very severe on the fourth day; the sweats were usually cold; it was impossible
to warm up their limbs, which remained livid and cold; they ceased being thirsty.
Their urine turned black, and was small in quantity, and thin; bowels were consti-
pated. In none of the patients with these symptoms was there a nasal bleeding,
only a slight loss of blood. ~cf one had a relapse, but all died on the sixth day,
s\veating. 51
Although this description has a broad scope, it is undoubtedly modeled

on the notes the physician took over the course of Philiscus's disease,
which were then filled out by observation of similar cases. In comparing
these two passages from Epidemics) I, one about a patient and the other
about a disease, one cannot help noticing a number of resemblances in
form and content that, as Monique Vust-Mussard correctly points out,
''are unquestionably too striking to be attributed to coincidence or
chance. " 52 However, the logical process by which the Hippocratic physi-
cian constructs his clinical picture only appears to be a matter of pure
induction. His glance is not as "virginal" or "objective" as numerous
historians of medicine say it is. Actually, his nosography is organized on
the basis of the various theoretical presuppositions and risky hypothetical
generalizations that also underlie his nosology.
The Hippocratic effort ends up with concise judgments, aphorisms even,
on the supranosological reality in the sequence of pathological events.
Here is an example of them from R--on-heticon) r: "Sweating, especially on
the head, in acute diseases, with pain, is bad, especially when the urine is
black; if dyspnea is added, that is bad. " 53 An aphorism from R--ognostic
likewise stresses the fatal significance of black urine. 54 Since that treatise is
probably by the same author as the first book of the Epidemics)the clinical
experience accumulated by this itinerant physician during his stay in Tha-
sos must have either confirmed or been the basis of his apprehensions
about patients presenting this symptom.
BlackwaterFever: The First Case in History

In ten of Hippocrates' patients, 55 the urine is described as black. All
suffered from a very severe febrile ailment. In terms of modern diagnoses,
one can entertain several possibilities depending on the special features of
each case, including first malaria and septicemia, then relapsing fever, lep-
tospirosis, and typhoid fever. Only three of the ten patients escape death.
I omit from consideration patients whose urine shades toward black (hu-
pomilana) or whose sediment or suspensions are blackish. Hippocrates had
reason to fear the appearance of black urine in patients suffering from acute
fevers. In the majority of cases in which ancient physicians encountered
urine of that color, it was due to the presence of reduced he1noglobin.
The clinical context makes it possible to exclude alkaptonuria, melanuria
in the strict sense, and porphyrinuria. Confusion with simple hematuria is
possible, at least in some specific cases. However, it does not seem to me
likely in the patients whose sufferings are related in Epidemics)I and III.
The Hippocratic physicians recognized perfectly well and explicitly noted

the presence of blood in urine, particularly in the course of diseases they
attributed to the kidneys. I should stress that "black" urine left no de-
posit, unlike the majority of urines called "red" or "shading toward
black.'' That is an important differential sign. Another misunderstanding
to avert would be, in my opinion, considering as hemoglobinuric dark
brown urine, which is easily described as "black"; actually, its color is due
to a urobilinuria that can arise, for instance, from an attack of typhoid. In
sum, hemoglobinuria is the pathological phenomenon that best corre-
sponds to the Hippocratic "black urine," and that entitles us to consider
it an expression of a hemolytic crisis.
In order to diagnose the idiopathic disease that brought on the hemo-
lytic crisis in Philiscus, it is useful to recall the circumstances and symptoms
by expressing them in modern medical parlance. This Thasian fell ill at the
beginning of the rainy season, in fall or toward the end of a summer that,
according to Hippocrates, was marked by an unusual drought and by
intense and unremitting heat. The climate and geographical structure of
this region, an insular coastal plain looking out on the nearby shore of
Thrace, favored vectors of all sorts of swamp hematozoa. With its saline
soil and typically Mediterranean coastal ecosystem, this part of the island
of Thasos has all the traits of the biotope that M. Sorre has characterized
as "a bioclimatically determined pathogenic complex. " 56 The kausos of
Philiscus did not arise sporadically, but, as Hippocrates notes in the third
constitution, it was the exacerbated form of an endemic disease.
Stricken by a sudden, rising fever, Philiscus is forced to remain bedrid-
den from the start of his sickness. He perspires freely, but his temperature
does not fall. He suffers throughout his body; then the fever becomes
remittent, with paroxysms like those of tertian fever. During the attacks,
he is agitated and confused; when they remit, he is weak and drowsy.
From the third day on, he suffers from thirst and a coated tongue. He is
constipated, which is treated with an enema and laxatives. On the third
and fourth days, his urine becomes black. (It should be noted that the
Hippocratic physicians do not observe urine right after micturition.) Then
the urine becomes clear but with whitish flakes, an ,attenuation that lasts
barely a day, after which it becomes black again. On the fifth day, there is
a mild nosebleed. During the last days of the disease, the limbs become
cold and livid, and there are cold sweats, Cheyne-Stokes respiration, and
significant spleen enlargement. Six days after the first bout of fever, the
patient is dead.
I recognize this as the first description in history of malarial hemoglobin-
uria or blackwater fever. It is very likely that the patient's quick demise
was caused by acute renal failure. In that case, his Cheyne-Stokes breathing
was due to uremia. I hasten to add that this diagnosis of Philiscus's disease
is couched in broad terms that are nowadays outdated. Malarial hemoglo-
binuria or blackwatet fever is a severe, paroxystic hemolytic syndrome that
is often fatal; it occurs in victims of Plasmodium fakiparum. My diagnosis is
of a clinical entity recognized in the nineteenth century that is now being
split up into several pathological states with similar symptoms that are
probably caused by differing pathogenic mechanisms.
There are writers, such as Galen and the Byzantine physicians Theophi-
lus Protosparathius (seventh century) and Johannes Actuarius (fourteenth
century), who mention as a fatal symptom the appearance of black urine
in certain acute febrile states (at times there is mention of jaundice and an
enlarged spleen). But their remarks are so vague that it would be rash to
insist on interpreting them by a modern diagnosis, especially because it is
hard to distinguish the personal clinical experience of these writers from
what they repeat of traditional Hippocratic lore.
In the modern era, the earliest certain descriptions of this disease go
back to the first half of the nineteenth century. First a few cases were
observed by French and British physicians among inhabitants of the west
coast of Africa; later there were reports by local practitioners in the United
States and Greece. 57 This pernicious form of malaria was first conceived as
a clinical entity sui generis during the third quarter of the nineteenth cen-
tury. Physicians in the French navy on the Comoro Islands and in Mada-
gascar isolated the syndrome under the name ''fievre pernicieuse icterique''
[pernicious icteric fever] (Lebeau, 1847, in his report on the health service
on Mayotte). Once the close link to malaria was recognized, the main
thrust of differential diagnosis was to distinguish this fever from "vomito
negro" or yellow fever. Three sets of symptoms were especially scruti-
nized: (1) remittent or unremitting high fever; (2) jaundice and bilious
vomiting; and (3) black urine. 58 The strange urine color was attributed to
massive renal hemorrhage, but some colonial physicians expressed doubts
as to the origins of this pigmentation. Nineteenth-century medical termi-
nology takes account of the malarial character of this disease as well as its
trilogy of symptoms. But there is no name that prevails for the majority of
physicians. 59
The first modern publication of a certain case of blackwater fever in
Greece dates from 1858. The clinical observation was made over a decade
before by A. Antoniades in Nauplia in the director of the School of Agri-
culture in Tiryns. 6 ° From then on the disease was reported fairly often in
the main wetland regions of continental Greece and the Peloponnesus,
less often in Crete and Cephalonia, and rarely among inhabitants of the
Aegan islands. Numerous sporadic cases have been described, but there
have also been true epidemics of malarial fevers with a predominance of
the hemoglobinuric type over all other pernicious complications. In the
last century, such epidemics took place, for instance, in Phthiotis in 1858-
59, in 1864-66, and in 1870. Perhaps one can even speak of a real recrudes-
cence of this disease in Greece during the second half of the nineteenth
century. "This type of malaria," writes Stephanos in 1884,
was apparently relatively rare or unknown around the first few years of Greek
independence in many places where it is relatively common today. Many older
physicians are certain that they only observed it for the first time in the past few
years. The evidence from these Greek practitioners is not trivial, but one must be
careful not to draw hasty conclusions from it, since their reports coincide with the
introduction of a new form of Western medicine in Greece and with the nosolog-
ical recognition on a global scale of blackwater fever. 61
Whatever the case may be, blackwater fever did not disappear from Greece
in the twentieth century, and the highs and lows of its frequency in mod-
ern times seem to correlate with variations in the degree of malarial
infestation.
Modern Greek physicians were the first to bring to light some etiological
factors and to clarify the real nature of black urine. In his article of 1858,
Antoniades was already citing the opinion of some of his Greek colleagues
who had observed massive "hematurias" in malaria victims taking qui-
nine; they wished to reduce the new nosological entity to an intolerant
reaction to chemotherapy. Antoniades rejects the hypothesis and asserts
that there can be "hematurias" of nonmedicinal origin during bouts of
malaria. 62 In a lecture published in 1859but delivered in Paris the same year
that Antoniades' article was published, S. Verettas took the lead in extoll-
ing the pathogenic explanation of blackwater fever in terms of the selective
toxic action of quinine. He had observed in his own father a special sensi-
tivity to the drug whose chief manifestation was the production of blackish
urine. According to Verettas, blackwater fever was the prerogative of ma-
laria sufferers who were subject to repeated bursts of fever and had been
vigorously treated with extracts of cinchona bark.
Observation of the undesirable consequences of antimalarial chemother-
apy spread. Even leading lights like S. Tommaselli and R. Koch did not
hesitate to consider blackwater fever as the intoxication by quinine of an
organism previously weakened by chronic disease. In retrospect, a signifi-
cant source of error in these opinions on the cause of malarial hemoglobin-
uria, especially in Greece, was confusion between it and the hemolytic
crises caused by glucose-6-phosphate dehydrogenase (G6PD) deficiency.
At the same time as they found proof of the triggering action of quinine
in numerous cases of malarial hemoglobinuria, some Greek malaria spe-
cialists at the end of the nineteenth century, particularly Karamitsas,
Koryllos, and Palladios, did note that identical attacks also took place in
the absence of quinine chemotherapy. According. to Karamitsas and Corre,
another sporadic cause of these attacks was the exposure of a chronic
malaria patient to moist cold. 63 In 1878, Karamitsas realized that the black
urine of these patients contained not blood proper but free hemoglobin
from the intravascular destruction of red blood cells.64
To explain the clinical features of blackwater fever, some raised the
possibility of a specific etiology. According to them, it was a disease radi-
cally different from malaria, produced by a special germ (they tried to
identify it with certain species of Babesia)spirochetes, and chlamydia), and
tied to malaria only by ecological and secondary epidemiological circum-
stances. But these attempts at explanation ended in total failure. 65 In 1892
two Italian researchers, E. Marchiafava and A. Bignami, 66 described Plas-
modium falciparum and distinguished between malignant tertian malaria or
falciparum malaria and quartan malaria. It was shown soon thereafter that
blackwater fever mainly arises in patients infested with falciparum
hematozoa. 67
In the twentieth century, knowledge of malarial hemoglobinuria is be-
coming complex. The most recent handbooks of hematology and malaria
admit that the pathogenesis of its hemolysis is as obscure as ever. Nowa-
days, the disease has been reduced to a syndrome, since there is good
reason to deny the unity of the pathogenic mechanism of hemolysis. In
the nineteenth century and the beginning of the twentieth century, it was
asserted that malarial hemoglobinuric crises occurred above all among in-
digenous inhabitants of hyperendemic zones. Current handbooks have a
different opinion: that the complication is more frequent in persons who
have recently arrived in endemic zones, for instance, among Europeans
landing in regions where tropical diseases rage. On the one hand, experi-
ence has shown that the administration of quinine and other analogous
medications can trigger a hemolytic crisis in patients with malaria. But on
the other, careful observations, such as those made during World Warn,
have shown that there are bouts of falciparum n1alaria accompanied by
severe hemolysis that cannot be attributed to the intervention of chemo-
therapy. Besides, the discovery of enzymatic deficiencies and the progress
of our knowledge on immunological mechanisms have laid the ground-
work for a deeper understanding of malarial hemoglobinuria.
To impose some order on this chaotic situation, it was necessary to
revise the nosological content of blackwater fever or malarial hemoglo-
binuria in the broad sense. In recent publications by French specialists,
a firm distinction is drawn between at least three disease entities that dif-
fer from one another in their pathogenesis as well as their clinical charac-
teristics. Each of these entities, however homogeneous it may be from
the standpoint of symptoms, is not necessarily produced by a single phy-
siopathological process. The three entities are as follows: (r) the acute
hemolytic syndrome in which hemoglobinuria immediately follows che-
motherapy; it is basically linked to a hereditary deficiency of erythrocytic
glucose-6-phosphate dehydrogenase; (2) the nonmedicinal hemolytic
syndrome, which occurs during an attack of invasive falciparum malaria;

and (3) malarial hemoglobinuria in the strict sense, arising in chronic ma-
laria patients usually, but not necessarily, after a quinine treatment and
manifesting itself by the classical profile of blackwater fever, with liver and
spleen enlargement and massive hemoglobinuria. 68
Concerning the first of these syndromes, I should stress that it is not a
purely coincidental encounter of three independent factors (hereditary en-
zymatic deficiency, malarial infestation, and quinine treatment). In 1957,
Motulsky showed that a falciparum malaria endemic favors, by the phe-
nomenon of balanced polymorphism, the frequency of inherited red cell
enzymopathies. According to U. Bienzle (1972), the biological advantage
of glucose-6-phosphate dehydrogenase deficiency resides in the accrued
resistance of heterozygous women to malarial infestation. Whatever the
case may be, this defect was and still is widespread in Greece and Italy. To
be sure, in the classical age in the Mediterranean there was no use of
antipyretics derived from cinchona bark or, still less, from 8-aminoquino-
line; but we should not forget that broad beans, which were consumed as
food and not for therapeutic reasons, can induce hemolytic crises in defi-
cient individuals. 69
Most of the recent explanations of the pathogenic mechanism of classical
malarial hemoglobinuria are inspired by the autoimmunization hypothesis
proposed in 1946 by J. Gear. 70 According to it, certain individuals who are
particularly sensitive or who find themselves in "immuno-allergic" situa-
tions that are still poorly understood develop autoantibodies that act
against their own antigenic constituents. In support of the existence of
autoantibodies in the blood of such individuals, one can point to the
positive results of Coombs' test. It has been suggested that the presence
of parasites changes the antigenic structure of individual erythrocytes and
that the parasite and its host both contribute to the formation of the
hemolytic factor. It may also be true that malarial hemoglobinuria is an
acquired enzymopathy. Probably it is necessary to invoke a third factor to
explain the triggering of the hemolytic crisis: it can be quinine, but it
seems to me that in our present state of ignorance, there are insufficient
reasons to restrict still further the definition of blackwater fever by exclud-
ing cases in which chemotherapy does not intrude as a secondary factor.
Modern Diagnosisof the Diseaseof Philiscus

Since it is not my task to unravel the tangled web of current hypotheses
about the mechanics ofhemolysis in infectious disease, now is the n1oment
to return to the main theme of this chapter, Philiscus's disease. The Hip-
pocratic presentation of it is in perfect accord with the clinical profile of
malarial hemoglobinuria, as sketched by nineteenth-century clinicians, ex-
cept for two details, both of which relate to "bilious" disturbances. Phil-
iscus does not seem to have suffered from jaundice or vomiting, and this
presents an especially troubling problem in historical pathology: what sig-
nificance should be given to silence about certain phenomena? Does the
omission of a symptom necessarily signify its absence? There is no need to
take literally the rhetorical claim that W.H.S. Jones made about Hippo-
crates' "clinical histories": "Nothing irrelevant is mentioned; everything
relevant is included. " 71 It is probably more true to historical reality to
believe, with R. Joly, that the Hippocratic descriptions contain insignifi-
cant details, since "by the very inadequacy of his knowledge of causes, the
writer is constrained to put everything on the same level of significance,"
and "above all, one should set aside the idea that Epidemics)I and III, are
devoted to purely objective observation. Several a priori's reveal them-
selves: even observation ... depends on a fleeting state of mind. " 72
In an area infested with malaria and amid a population whose skin was
tanned by the sun and sea air, a slight case of jaundice could easily escape
notice if someone was not carefully looking for it; only a pronounced case
would impress itself on an objective observer. But the same is not true of
vomiting, a symptom to which Hippocrates attributed great importance
and one he would have not failed to notice if Philiscus had suffered from
it. In the course of the generalized description of this disease (the kausosof
the third constitution), nausea is not forgotten. However, in studying the
evidence favoring the existence of blackwater fever in ancient Greece, I
prefer to hold to concrete cases and avoid reference to a synthetic descrip-
tion that may actually be an amalgam of what we now consider several
diseases.
H. Foy and A. Kondi 73 propose a diagnosis of blackwater fever for five
other cases in Epidemics)I and III (Silenus, Hermocrates, Pythion, Hero-
pythus, and Apollonius). Putting aside Philiscus for the moment, the only
Hippocratic case in which the diagnosis seems justified is that of Pythion.
Hermocrates' disease is probably leptospirosis icterohemorrhagica (Weil 's
disease). As for Silenus, Heropythus, and Apollonius, the most likely
diagnoses are invasive malaria and typhoid fever. In summary form, here
are the disease symptoms mentioned by Hippocrates in the case of Py-
thion, who lived on Thasos near the sanctuary of Heracles: acute intermit-
tent or remittent fever of the tertian type, sudden onset, chills, weakness,
periodic coldness in the extremities, short intervals of mental confusion,
shorter than normal breathing, nausea, vomiting, heaviness in the stom-
ach, black urine without sediment, death on the tenth day of the disease. 74
Vomiting is a common but not obligatory symptom of blackwater
fever. Its absence in the case of Philiscus is not material to a differential
diagnosis. Since the hemolysis that occurs in this disease is generally intra-
vascular, the jaundice that accompanies it is generally mild. It appears only
after the transformation of free hemoglobin into bilirubin. All other things
being equal, the mention of flamboyantly yellow skin coloration in an
ancient clinical description suggests Weil's disease rather than malarial

hemoglobinuria.
The Hippocratic description of Cheyne-Stokes respiration is concise and
beyond doubt. 75 In modern medicine, this type of dyspnea was described
by the Irish physicians John Cheyne (1818) and William Stokes (1854) in
patients suffering from heart failure. Near the end of the century, Sir
William Osler, one of the most famous American physicians of the time,
observed this type of breathing in a patient suffering from an acute hyper-
thermia that was soon fatal. The diagnosis of the living patient was just
"heat stroke"-a sad mistake for such a clever clinician-but the autopsy
showed malignant malarial infestation and thus established a link between
that etiology and the clinical observation of Cheyne-Stokes breathing. 76
Subsequent investigations have shown that in such cases the paroxystic
dyspnea is due to uremia resulting from acute renal failure, which is very
common in blackwater fever or in malignant falciparum malaria. 77
What diagnoses for Philiscus's case have been proposed up to now by
medical historians? In the beginning of the nineteenth century, Hippo-
cratic commentators likened the febrile diseases reported in Epidemics) I
and III, to nosological entities that are obsolete today but correspond to
various clinical forms of typhus and typhoid fever. The case of Philiscus
was interpreted as a special variant of typhoid. 78 Then advances in colonial
medicine improved knowledge of tropical diseases, and E. Littre, relying
on the works of M. Maillot, W. Twinint, and J. Johnson, was able to
assert in 1840 that "Philiscus's disease as related by Hippocrates has noth-
ing to do with typhus or typhoid fever." Instead, it is "a remittent fever
from the tropics. " 79 A glance at the history of the disease of a soldier
named Devos, whose case is cited by Littre as the modern counterpart of
Philiscus's, is enough to confirm that his diagnosis corresponds to what is
now called pernicious malaria. 80 In 1859, Carl A. Wunderlich, in his hand-
book on the history of medicine, expressed the consensusomnium that
Philiscus's disease is a classic example of malaria. 81 A retrospective diagnosis
of malarial hemoglobinuria was proposed above all by Greek physicians of
the second half of the nineteenth century. In 1861, A. Antoniades, who
provided the first modern description of this syndrome in Greece (see
above), drew attention to the resemblance between his own observations
and the Hippocratic descriptions. C. Stephanos remarked with caution
that "hemospherinuric malarial fever" was "perhaps observed by Hippo-
crates on Thasos (Epidemics)r, third constitution, first patient). " 82 J. P.
Cardamatis supported his identification with enthusiasm as well as sound
knowledge of the subject. 83 But difficulties began to arise with the hypoth-
esis of the obligatory intrusion of quinine to trigger the hemolytic crisis.
In 1909, in his ground-breaking monograph on the role of malaria in Greek
history, W.H.S. Jones refused to accept the existence of this syndrome in
the Hippocratic writings on the grounds that the ancient Greeks were
ignorant of treatment by extracts from cinchona bark. In the case of Phil-

iscus he opts for simple malignant malaria. This argument is lame and was
soundly criticized by H. Foy and A. Kondi 84 and by H. Scott; 85 even so it
reappears from time to time in recent publications.
There is no doubt that it would be impossible to contemplate a diag-
nosis of blackwater fever for Philiscus if by definition that name were
restricted to hemoglobinuric crises brought on in malaria victims by the
administration of certain medications. That is not where the problem lies.
The issue is whether some Hippocratic descriptions, in particular the cases
of Philiscus and Pythion, correspond more closely to the modern symp-
tomatology of malarial hemoglobinuria than to a severe attack of' 'bilious
remittent" malaria. It is conceivable that without quinine the triggering
substance was something eaten-for instance, broad beans. In a recent
work, M. Martiny made the following remarks on Philiscus's case: "Amid
the various acute ailments possible, a severe form of invasive malaria could
provide this clinical profile. " 86 This verdict is a reaffirmation of the cau-
tious opinions of Jones and Littre. In my opinion, the enlarged spleen,
the periodic drops in fever, and the seriousness of the urine modifications
make a diagnosis of blackwater fever more likely than one of invasive
malaria.
The two possibilities considered here both view Pl. falciparum as the
main pathological agent. But there is the rare medical historian who casts
doubt on that etiology for Philiscus's disease. According to G. Sticker, his
is a case of typhus recurrens, that is in modern medical parlance, an espe-
cially severe form of relapsing fever (borreliosis). Since the patient died on
the sixth day, which is usually crucial in that disease, its main symptom,
the recurrence of the febrile state, could not manifest itself, but Sticker
believes that his diagnosis is justified if other Hippocratic texts are taken
into account. 87 I reject this way of amalgamating cases. It is necessary first
to prove the nosological identity of descriptions that relate to different
patients, not to presuppose it. In Philiscus's case, taken in isolation, the
following characteristics are clearly against an identification with relapsing
fever: the remittence of his hyperthermia (not a fever on a continuously
high level, as seen in the initial phase in patients afflicted with relapsing
fever); the color of his urine; the absence of severe muscular and joint
soreness; a swollen spleen that seems soft instead of hard.
A physician who is otherwise· very sensible, R. E. Siegel has suggested
exanthematic typhus, but indirectly: he proposes it not explicitly for Phil-
iscus but in connection with the generalized description of his disease
included in the third constitution of Thasos. 88 That is really attempting
the impossible, and to do it Siegel has to resort to subterfuge. In citing
the passage from Hippocrates, he quietly replaces a key phrase ("exacer-
bations on even days'') with an ellipsis ( ... ) and tries vaguely in his
comments to explain this periodicity in the fever as the irregular remissions
that mark the fever curve during the first ten days of an infection with
Rickcttsiaprowazekii. There is a single point in favor of Siegel's hypothesis,
the coldness and lividity of the extremities. Against it, there is a decisive
detail: the exanthem that regularly appears on the fourth or fifth day of
exanthematic typhus is mentioned neither in the case history of Philiscus
nor in the generalized description of the kausosof which he perished. In
any case, other, more likely alternatives can be imagined, such as Weil's
disease, acute miliary tuberculosis, or a severe form of streptococcal septi-
cemia. Nor should we dismiss the possibility of a strange coincidence: a
crisis of favism in a patient suffering from falciparum malaria.
In conclusion, exegesis of the case history of Philiscus provides strong
evidence in support of the existence of falciparum malaria in the eastern
Mediterranean toward the end of the fifth century B.c. Information and
considerations of another type also point to its recrudescence and, in come
areas of Greece, to its introduction at this time. This recrudescence and
the concomitant observation of "melanuric" fever are probably not unre-
lated to the transformation in the concept of black bile that was taking
place right at the time that the main treatises of the Hippocratic corpus
were being composed. 89 Just like favism, blackwater fever was probably
present in Greece for at least the 2,500 years between Hippocrates and the
nineteenth century, even though it was invisible to the eyes of physicians
owing to a combination of sociological and doctrinal factors. The lot of
this Thasian man makes me suspect that blackwater fever could well be the
clinical expression of malarial infestation encountering a still unidentified
enzymopathy. At this juncture, however, the historian of medicine must
hand the matter over to the clinicians and experimenters who live in the
present and for the future.
Chapter Twelve
THE CONSTITUTION OF A
WINTER IN THRACE
The "Cough of Perinthus)J
The Hippocratic report on what has been called, since Littre, the '' cough
of Perinthus" is a basic text that should be considered a paradigm in any
historical, philological, or medical study of Greek epidemiology in the
classical period, alongside the "plague of Athens" and the katastaseisof
Thasos. 1 This narrative, part of the sixth book of the Epidemics) is very
carefully composed and of exceptional richness from a clinical standpoint. 2
I append a translation of it: 3
Coughs began around the winter solstice, fifteen or twenty days after there had
been frequent changes in the south wind, the north wind, and the snow wind (1).
Some [of these coughs] lasted a short time, others long; then there were frequent
pneumonias (2). Before the equinox (3), most had a relapse, usually on the fortieth
day counting from the start [of the illness]. In some, the relapse was hardly severe
and its crisis was mild; in others, the throat was inflamed, and in others there was
angina (4); in some, there were paralyses (5), and others, especially children, had
trouble seeing at night (6). The pneumonias were not very severe (7). Trouble
seeing at night (8) replaced coughing in those who subsequently coughed a little
or not at all; sore throats were not severe, especially in those who saw poorly at
night (9). Anginas and paralyses (10) brought with them hard, dry [sputum], or
little and slightly concocted, or in some highly concocted (n). Those who had
used their voice a lot or suffered a chill were finally stricken (12) mostly with angina
(13). Those who exerted themselves with their arms had paralyses only in their
arms. Those who rode horseback or who walked a lot or who tired out their lower
limbs in some other way had paralytic disturbances (14) in their loins or lower
limbs; also aches and pains in their thighs and legs (15). The harshest and most
violent [coughs] brought with them paralyses (16). All this happened in patients
with relapses, rarely in the first stage. In several patients, the coughing subsided in
the middle (17) without stopping altogether; and it came back during the relapse.
In those who lost their voice in fits of coughing (18), most did not even have fever
and a few had a very little bit. None of them suffered at all from pneumonia or
paralysis; they had no other symptoms than the crisis in their voice (19). Problems
with night vision arose in the same way as those that arose after other initial
manifestations (20). They mainly affected children; as for eyes, especially those
whose eyes were black with lightly colored spots and small pupils, generally those
in which black was dominant (21). There was a preference for people with big eyes,
not small ones; most of them had black, straight hair. Women did not suffer the
same way from the cough; only a few had the fever, and of those, very few fell
victim to pneumonia; they were old people, and all recovered. I attributed (22)
that (23) both to the fact that they do not go out like men (24) and to the fact
that, even in other diseases, they are not affected the way men are (25). Only two
free-born women had anginas (26), which were benign to boot. Slave women were
more severely affected, among whom there were some violent cases that soon
proved fatal (27). However, many men were stricken, and some were cured, while
others perished. As a rule, the disease was benign and bearable for those who
merely couldn't swallow (28), but it was troublesome and long-lasting in those
who also spoke indistinctly; among those who in addition suffered from swollen
veins in their temples and neck (29), it was pernicious; finally, it was a very grave
disease in those whose breathing was elevated (30), since they had the fever. The
train of morbid symptoms was as I have just described it; those described at the
beginning appeared without those described afterward, but the later symptoms
did not occur without the earlier ones. The quickest to die were those who were
overcome by a feverish chill. In all these patients, nothing was gained either by
massage (31), or purging the abdominal cavity, or bleeding, all of which I at-
tempted. I even tried making an incision beneath the tongue (32); and to some I
administered medication orally (33). These diseases persisted throughout the sum-
mer along with many others that burst on the scene. First (34), during the drought,
painful ophthalmias predominated.
Commentaryon the HippocraticText and Its Translation

Every modern translation of an ancient text like this one is actually only
an interpretation. It is therefore incumbent on me to explain and to try to
justify my version. I have taken Littre's edition of the Greek text as my
point of departure, along with his apparatus criticus and a collaction of the
two best manuscripts, Marcianus 269 (M) and Vaticanus 276 (V). 4 The sixth
book of the Epidemics was the subject of commentaries by Galen, Palladius,
and John of Alexandria. 5 Galen's commentary is by far the most impor-
tant, but unfortunately the part on the description of the "cough of
Perinthus" was not preserved in its original version. Some editions of the
works of Galen contain a Latin translation of this part of his remarks; but
we should not be fooled-that text is a forgery from the sixteenth century.
However, we now have at our disposal a modern German translation of
the medieval Arabic version of Galen's commentary. 6 As for the two Al-
exandrian commentators from the end of the sixth century, Palladius and
John, the farmer's work is available to us in a well-edited Greek text, while
we have the latter's by way of a recent critical edition of it in a Latin
version. 7 As for translations of this Hippocratic passage into modern
THE "COUGH OF PERINTHUS" 307
languages, I have consulted those of Littre, Puschmann, Fuchs, and

Kapferer. 8
Explanatory remarks on several details of the Hippocratic account follow:
1. The epidemic cough appeared right at the start of winter, after an

autumn marked by frequent weather changes. Puschmann is incorrect in
translating the passage to mean that the epidemic began "the fifteenth or
twentieth day after the solstice." It is important to keep in mind that
these events took place in northern Greece, where the climate is colder
than on the Mediterranean islands that were probably the homeland of
the physician who wrote this report. It is easy to understand why the
Arabic translator of this lemma in the works of Galen interpreted the snow
wind as an "icy wind." In connection with this information about the
winds, Galen theorizes that when the south wind shifts to the north, it is
especially harmful to the head, since it is uncovered. He explains the
pathogenesis of all the forms of sickness that will be mentioned in the
Hippocratic text as stemming from an extreme cold that descends from
the head and becomes fixed in the throat, the lungs, the eyes, or the
limbs. The south wind gathered up the moisture in the head, and the
north wind impeded it from being properly dissipated, whence, says Galen,
the catarrh, the cough, the pneumonia, the sore throat, the vision prob-
lems, and the paralyses.
2. Peripneumonfais presented in this account as a complication arising
immediately after the initial disease and not as a relapse. For the now
conventional translation of the ancient Greek word peripneumonfaor peri-
pleumonfa (Latin peripneumonia) by the modern term "pneumonia," see
above, Introduction, "Semantic Constants and the Difficulties of Retro-
spective Diagnosis," and chapter 4, "The Oldest Medical Evidence on
Purulent Inflammation of Internal Organs."
3. The spring equinox takes place on March 21 or 22. According to
Galen, the reading in the old manuscripts was both "before the equinox"
and "after the equinox." The former agrees better with the interval be-
tween the initial disease and the relapse mentioned in the next clause.
4. In Greek, the term is kundngkhai. It can be translated "anginas,"
in the general sense of a strangling pain. I will return to this issue. This
passage is especially instructive in regard to it since it draws a clear distinc-
tion between kundngkhai and inflammation of the throat, pharungesephleg-
rnenan. Puschmann's translation ("Rachenentziindung" [sore throat] as
against "Halsentziindung" [laryngitis]) is false medically and a theoretical
anachronism, since the Hippocratic writers did not yet make the distinc-
tion that modern anatomy does between the pharynx and the lary_nx.
5. The Greek has paraplegikga)but that is not n1odern paraplegia. For
the Hippocratic writers, it is indeed paralysis of the limbs, but without
specifics as to which.
6. I understand Hippocratic nuktaliJepesto mean night blindness. That

agrees with Galen's and Palladius's commentaries, which are very explicit
about it, but it is not the opinion of Littre or still less that of Fuchs, who
relies on a statement in J+orrheticon)n, to support his translation of nuktal-
opsas "day blindness." To confirm his preconceived diagnosis and cut this
Gordian knot with a single stroke of his pen, Puschmann states that it is
merely a case of photophobia ("die Lichtscheu im Allgemeinen") proba-
bly caused by influenza. This view does not stand up to critical analysis of
the ancient evidence on nuktalops)·and medical arguments make the tran-
slations of Littre and Fuchs very unlikely.
7. Another possible translation is that they were of short duration. In
any case, this statement seems to contradict what was just said on the
importance of the pneumonias that followed the coughs. Palladius ex-
plains correctly that the contradiction is just an apparent one, since only
in relapse were the pneumonias minor.
8. Here, the text of the recentiores(with the exception ofC, a sixteenth-
century manuscript that usually matches V) and the vulgate is corrupt;
instead of nuktalopesthere is discussion of pharunges) which from a medical
standpoint is clearly not harmonious with the context. Galen's commen-
tary has a lacuna at this point; Palladius and John speak of "nyctalopia"
in reference to this passage. Careful examination of the M manuscript
(f. 422v) proves that the error goes back to its scribe: he skipped a passage
and began his phrase with the word pharunges) which is repeated a little
farther down in the Hippocratic text; then he realized what he had done
but only partially corrected his mistake. The result is that M and its apo-
graphs omit nuktalopes while the word pharunges appears in them twice
instead of once. The correct reading can be restored from the V manuscript.
9. I cannot follow Littre here, who translates "sore throats that were
mild and less than the nyctalopias."
10. The word paraplegikawas eliminated by Puschmann, who found it
troublesome on his medical interpretation. According to him, the word is
lacking in Palladius, which is incorrect. Puschmann read Littre's apparatus
incorrectly. It says "Pall. n'a rien sur cette phrase" [P. has nothing on this
sentence], that is, as opposed to John of Alexandria, he does not comment
on it. According to Galen, paralyses were mentioned here, even though
the context cannot relate to them if his medical theory is the right one.
n. Since this sentence is elliptical in Greek, my translation is of necessity
a conjecture. What do the two adjectives sklera kai xera and the verb
andgousairefer to? According to Foes's Latin translation, they should refer
to the anginas and paralyses: anginae autem et partium siderationesaut durae
aut siccaeerant aut parvae et raro matura educentes.Littre does not accept this
interpretation. He construes the adjectives with sputum and refers the
verb to the cough, a word that is to be understood: ''The writer was able
to omit [it] the more easily since it was the chief phenomenon of this
THE "COUGH OF PERINTHUS"
epidemic and as such was always present to mind." From this argument
results the folloWing translation: "The anginas and paralyses declared
themselves in the coughs which brought up hard, dry matter, or little and
slightly concocted, or even in some cases highly concocted.'' My transla-
tion is fairly close to Littre's, but it differs from it by refusing to introduce
in explicit terms the notion of the cough. I am inspired mainly by Galen's
remarks, \vhere he speaks of dry, hard matters, or slightly concocted ones,
that are rarely, though sometimes profusely, spit up. Although I know of
this remark only via the Arabic and German versions, it is certain that the
solution adopted there goes back to Galen himself. He clearly affirms that
the terms in this sentence refer to sputa that are associated only with
angina, not paralysis.
12. The tradition recognized two variants to this passage: according to
the first, the patients were stricken with angina, but according to the
second, they died of the disease. This divergence was already in existence
in the time of Galen, who considers the former reading correct, though he
stresses that the angina was indeed malignant.
13. According to Galen, the Hippocratic writer is insisting on his discov-
ery. The humors that derive from the head have a predilection for the
organs that were previously weakened. There is a general formulation of
this clinical observation in Aphorisms (rv, 33).
14. John of Alexandria explains that the expression akras(aiparaplegika(
denotes paralytic impotence. Littre translates it "intemperies paralytique"
[paralytic humoral disturbance], which construes the first term of the
Greek expression as a derivative of krasis'mixture ofhumors.' The humoral
disturbances in question are probably those expressed by the loss or sub-
stantive lessening of voluntary movement. According to Galen, the im-
potence in question is such that the patient lacks the strength needed to
support himself on his legs and remain standing without wavering.
15. In drawing attention to this painful sensitivity, the Hippocratic writer
provides us with key evidence for a differential diagnosis. Galen under-
stood the importance of this observation. A total paralysis, he tells us,
occurs when feeling and mobility are lost at one and the same time. He
reports his own clinical experience that a limb that has lost its mobility can
still be painful. We know that this dissociation actually exists in certain
paralyses, for instance, in those that are due to diphtheria or poliomyelitis.
16. In Greek, the subject of this sentence is not given, a fact noted by
all the comm en tat ors. Galen refers the sentence to the pains that were
mentioned in the previous one, but he must then admit that the writer
made a blunder when he used the feminine form instead of the masculine.
Palladius basically agrees with Galen, and he tries to explain the unlooked-
for feminine by understanding the synonym odunai as standing in place of
p6noi. Neither solution is satisfactory, for their medical content is dubious,
even unlikely. Cornarius and Foes, followed by Littre and Fuchs, decided
to introduce the word bekhes'coughs' as the subject of the sentence. I join

them.
17. By using the expression en t{Jimisoi) the authory probably meant that
the coughs subsided in the interval between the relapse and the initial
phase of the disease. Other interpreta~ions are possible. According to
Fuchs, the cough increased in the middle of the relapse. Palladius under-
stood something completely different, namely, that compared with other
troubles, the cough was of middling severity.
18. The translation I have adopted here is Littre's. It seems to me that
this passage is describing concisely a symptom that is encountered, for
instance, in patients with whooping cough, namely, the sudden interrup-
tion of speech by coughing fits. The Arabic translation in the indirect
tradition is really different: "in those whose voice is lost due to coughing."
The Arabic translator, Hunain ibn Ishaq, took his cue from Galen's com-
mentary on the passage. Galen explains that in these patients, since all the
pathogenic matter is deposited near the larynx and trachea, the voice is
ruined. If this interpretation is accepted, the disease is one marked not by
sudden, unforeseen fits of coughing, but by a kind of laryngitis with
aphonia.
19. It is worth stressing that the Hippocratic writer is now isolating very
clearly a group of patients who suffered from a special type of cough that
was relatively benign in both its accompanying symptoms and its outcome.
20. Littre translates, "like those arising from altogether different
causes.'' At issue is the pr6phasis)an ancient notion customarily under-
stood to mean "occasion" or "immediate cause." In the classical medical
writings, this term usually keeps its original sense and indicates the initial
aspect or the very first expression of a state, especially a state of disease. 9
That meaning applies here very well and elucidates the Hippocratic state-
ment: the hemeralopias in relapse were not different from those that arose
in the beginning of the disease. Littre's translation causes difficulties in the
medical interpretation of the text, since its author allows for only one
cause of night blindness.
21. I have tried to preserve the elliptical style of this sentence in my
translation. This whole text is a set of notes. Greek physicians used writing
to save the essence of their observations and thoughts in concise format;
information committed to memory was to be transmitted by word of
mouth as a supplement to the notes, to render them complete. According
to Galen, the Hippocratic description is referring to the appearance of the
eyes, which on this occasion tended to be afflicted with night blindness.
An individual observation has been made over into a general rule, and
Galen struggles to provide a physiopathological explanation of it. Accord-
ing to him, black eyes, small pupils, eyes that change color, and black,
straight hair constitute a predisposition to night blindness because they are
the outer signs of a moist temperament and, in particular, of an inborn
abundance of moisture in the head and the eyes. The same explanation
had been given more succinctly but in almost identical terms in a treatise
of Aristotle. 10 Galen certainly knew the passage, and moreover I think it
not unlikely that Aristotle's theory of vision and its failings was inspired
by the clinical reports of the Hippocratic physicians. In particular, it ap-
pears that Aristotle knew, either directly or indirectly, the observations
contained in the sixth book of the Epidemics.Palladius is distancing himself
from both Aristotle and Galen when he remarks in regard to this line that
the eyes became black, the pupils small, and the eyeballs swollen under
the influence of the ailment. On his interpretation, the Hippocratic de-
scription does not refer to the patients' state before the disease but instead
to the way their eyes looked during it. Littre took a stand against Palladius
and similar to Galen's, which was not yet known to Western scholars since
it is preserved only in the Arabic version. I think it necessary to qualify
both views: the Hippocratic writer did indeed have predisposition in mind
and not the changes brought on by night blindness, but in practice he was
unable to foretell who would be affiicted by this disease; so he had to base
his observations on actual patients at the same time as he believed that
they had had the traits given above before falling ill. He mistook what is
actually a symptom of the disease for a normal somatic trait.
The Greek text contains a strange word: hupopofkila.Galen seems to
have understood this to mean that the eyes had changing colors. To a
modern physician, that is nonsense. Littre translates, "of somewhat varied
color." That is satisfactory from a lexical standpoint, but it introduces a
certain contradiction, namely, that eyes susceptible to night blindness were
at once black and variegated. To skirt the problem, Puschmann offered
the following translation: "in den Augen erschien das Schwarze verschie-
denartig gefarbt." If hupopofkilareally denotes shimmering eyes, or eyes
with lightly colored spots, it may be the first notice of a distinguishing
symptom of xerophthalmia, the organic side of night blindness. At the
time Littre was translating the Epidemicsand writing his commentary on
them, little was known about the anatomopathological substratum of night
blindness, and its etiology was completely unknown. In 1863 Bitot pub-
lished the first description of the whitish or, rather, pearly gray, iridescent
spots that he observed on the conjunctiva of undernourished children in a
Bordeaux orphanage; he noted their correlation with disturbances in night
vision. 11 Later on the real nature of these xerophthalmic spots was under-
stood: they are the clinical manifestations of the cornea degeneration that,
along with failing eyesight in weak light, is caused by vitamin A deficiency.
In conclusion, it seems that the Hippocratic writer saw and noted in
summary fashion Bitot's pearly spots, but that their presence was wrongly
taken to be a normal somatic trait that was a predisposition to night
blindness.
22. One should note the use of the verb in the first person singular.
The text is certainly an individual's handiwork, and it bears the marks of

an original version composed when the events in question were not re-
mote. Its author is a physician whose opinion has weight in the Greek
intellectual community, a weight of which he is indeed conscious.
23. Littre's translation of this passage as "j'attribuai cette immunite" [I
attributed this immunity] is a serious anachronism. Neither the term nor
the concept of immunity in its modern medical sense existed in Hippo-
crates' time. The Greek text simply has the word touto 'this thing.' Palla-
dius believes that it refers to menstruation: the women were less subject
to the cough, fevers, and pneumonia than men because they did not go
out as much and because they benefit from menstrual purgation. Probably
Palladius's strange interpretation stems from his desire to absolve Hippo-
crates of a reproach by Galen.
24. According to Galen, the author meant that women are less suscep-
tible than men to bad weather and abrupt changes in air. The Hippocratic
explanation is only partial, he says, because one should add the benefit
accruing to women from menstruation.
25. This sentence looks like a marginal comment that later slipped into
the text. It is missing from the V manuscript. But if it is an interpolation,
it is an old one, since Galen knows of it.
26. According to Galen, Artemidorus Capito, an editor of the Hippo-
cratic corpus, suppressed the word "two" here, but most commentators
believed that it belonged in the original text.
27. By Galen's time two readings of this passage were known. The
second of these was correctly translated by Littre: "clans les cas ou elles
furent violentes, elles causerent tres promptement la mart" [when they
were violent, they were almost immediately fatal].
28. According to medical teaching in Galen's day, difficulty in swallow-
ing was a very serious symptom. This text is therefore an embarrassment
to him, and he resorts to all his skill in argument to explain why it is
associated here with a mild ailment. In this case the advantage clearly lies
with the relatively unsystematic, flexible theory of the Hippocratic
observer.
29. The Greek word is aukh~n. Its usual meaning.is "nape of the neck."
Curiously, the indirect tradition (Hunain ibn Ishaq) here speaks of "eyes."
I translate the word "neck," since there are no blood vessels in the nape
of the neck or the eyes that would be visible when swollen. For Galen,
such swelling is a harbinger of suffocation. It occurs when bad humors are
so abundant that they cannot be evacuated and overflow their place of
abscession.
30. Exactly what is meant by the expression pneurna xunemeteorfzeto?It
is some kind of breathing, but Littre calls it "elevated," Fuchs "superfi-
cial," and Puschmann "accelerated." According to Galen, breathing was
elevated in these patients because of laryngeal constriction. In modern
medical terms, this means laryngeal dyspnea. There is no doubt that the
Hippocratic account is an increasingly precise clinical picture of croup.
31. In the medieval manuscripts of the direct tradition, the text of this
passage is obscure. The chief difficulty is to understand precisely what is
meant by the expression oude anastasei.And how should one translate the
term piezomenousin relation to it? Foes thinks it has to do with the repres-
sive effects of re percussive medications. Calvus understands anastasis to
mean defecation, and he is followed therein by Littre, who, ''without
mistal(ing just how obscure this passage really is,'' decides to translate it as
follows: "ces malades, que la presentation sur le siege ne fatiguait meme
pas" [these patients, who were not the least bit indisposed by the calls of
nature]. From a medical standpoint this is really ridiculous, so Fuchs sug-
gests we understand the passage to mean that the sick were tortured by
the need to defecate. That translation suits the demands of logic, but it
does violence to the Greek language. Cornarius tried another solution:
suppress the words oude anastasei. This is exactly what Hunain ibn Ishaq
did in the Arabic translation of the Hippocratic lemma of Galen's com-
mentary. His justification for doing so comes from the commentary itself.
Galen says that there were two readings. I have adopted the one he dubs
"ancient" and whose medical value he explains very clearly.
The Hippocratic writer is here alluding to the treatment of swollen
tonsils by local massage: the physician used his fingers, which had been
smeared with ointment. Since these porous glands had become swollen
through the absorption of phlegmatic humor, massage was indicated in
order to squeeze out the liquid. Galen states that he used this method
himself in combination with the administration of astringents. Manual
intervention should be prudent and moderate. It is useful, he says, when
the swelling affects only the glands. If the throat itself is swollen, massage
can actually be harmful. According to Galen, some editions of the Hip-
pocratic text had another reading of the passage in question, to wit, that
there was no use in torturing ("squeezing") the patients with "forbidden
things." Those two words have always piqued commentators. Galen tells
us that according to the "new physicians" (these are "the ones who lived
after the death of Alexander"), it was a matter of abstaining from certain
foods, but that the "ancient physicians" had taken no position on the
subject.
32. The Hippocratic writer practiced a special form of bloodletting.
Note the insistent, repeated use of the first person singular of the verb.
33. Littre translates "que je fis vomir" [whom I induced to vomit].
That is indeed simpler and clearer. But the Greek text has a lack of certainty
that I wished to preserve by translating it as literally as possible. Usually,
says Galen, ano epharmakeusameans ''I induced vomiting by medications,''
and when Hippocrates speaks of medicating ''from below'' and ''from
above" without further detail, the former implies the use of purgatives
and the latter the use of emetics. In this case, however, the usual interpre-
tation runs up against a medical objection: to induce vomiting in a patient
with a swollen throat who is in danger of suffocating is counterindicated.
That is why Dioscorides, a Hellenistic editor of the text of Hippocrates,
chose to replace anoby kdto. If that is the correct reading, our healer
purged his patients instead of making them vomit. Galen sees the virtue in
ridding the text of vomiting, but he does not feel entitled to modify such
a venerable document without a historical or philological motive. So he
keeps the reading "from above" and mends the text by explaining that
the Hippocratic writer is referring not to the use of emetics but to topical
medications, especially gargling and the application of drugs to drive out
phlegm through the nostrils.
34. This instance of proton'first of all' followed by a truncated clause
and without a matching "second," shows that the narrative aborts and
has not been preserved in its entirety. The only consolation is that, in
beginning to speak of the summer, the author seems to have completed
his statement on the diseases that ruled in winter and in spring over the
space of a year in a specific Greek settlement.
The Authory Datey and Localeof this EpidemiologicalReport
Although this text contains no hint of the name of the place in which
the Hippocratic physician made his observations, it is certain that the
whole story is connected with the city of Perin thus. In a chapter from the
same book of the Epidemics)there is talk of "a cough that visited Perin thus
during the winter. " 12 Moreover, in the Hippocratic treatise Humors) there
is a remark that the pathological humors settle in parts that were abused
before the outbreak of the disease, "as was the case with those who in
Perinthus suffered from cough and angina. " 13
Situated on the Propontis, Perinthus was an important fortified port in
Thrace. In the course of its troubled history, it changed its name to Her-
acleum. Today it is known as Eregli, a Turkish town on the Sea of Mar-
mora. The town of Perin thus is mentioned by nar:ne seven times in the
Hippocratic corpus. The oldest of the citations is probably the one in
Section 3 of book 2 of the Epidemics)in which there are descriptions of the
diseases observed and treated by a group of physicians immediately upon
their arrival in Perinthus near the solstice of a certain year-which year, we
are not told. A passing remark in the same book stresses the influence of
the regularity of the seasons on the nature of diseases: an aphorism of
broad applicability is illustrated "by what is seen at Perinthus." Three
other citations concern patients who are said to be inhabitants of Perin-
thus: a certain Antigenes, his wife, and a swarthy patient who lived in the
home ofTemenes' niece~14 Antigenes' wife was among the patients visited
by the Hippocratic physicians at about the time when the cough afflicted
the Perinthians. Her case is very strange from a medical standpoint, but
for the moment it interests us only because it seems to offer a way to date
the stay in Perinthus of the author of the notes about the cough.
The consultation in question took place at about the same time as two
extraordinary natural events that were noted by the itinerant physician in
a kind of preamble to his clinical report: "At the winter solstice, the
appearance of a rather large star; the fifth and sixth days thereafter, an
earthquake. " 15 Littre relates this text to a notice in Aristotle that "in the
archonship of Eucles, -the son of Molon, at Athens, a comet appeared in
the north in the month of Gamelion when the sun was in the winter
solstice. " 16 Two men named Eucles were archons at Athens, one in 426,
the other in 402. Historians of astronomy and modern commentators on
Aristotle favor 427-426 B.C. as the date of the appearance of this comet.
That is also the opinion of Littre, who furthermore recalls that during the
winter of 427-426, according to Thucydides, there was a recurrence of the
"plague" at Athens and earthquakes were felt in Attica, Euboea, and
Boeotia. 17
After the philological and historical analyses of Deichgraber, however,
Littre's dating became very troublesome, because the group consisting of
Epidemics)II, IV, and VI, appear to be later than the group consisting of
Epicemics)I and III, which is dated around 410 B.C. That is why Grense-
mann, who considers 400-399 B.C. as the tenninus post quem of Epidemics)
IV, identifies the "rather large star" in the Hippocratic text with the comet
of 373-372 B.C. 18 Just like Littre, Grensemann relies on the authority of
Aristotle, but he evokes another passage in which there is talk of "the
great comet that was seen at the time of the earthquake in Achaea and the
coastal flood ... This large star appeared in winter during the frost and
when the weather was calm in the west, under the archonship of As-
teius. " 19 This really seems to refer to the winter of 373-372, as Gundel
allows, but it is worth noting that in the astronomical lists of Pingre and
Bigourdan, this same comet is dated to 371 B.c. 20
Di Benedetto disputes Grensemann's argument and returns to Littre's
view. According to him, the star seen at Perin thus was the same as the
comet seen at Athens under the northern sky in January of 426 (the month
of Gamelion began that year on the day that corresponds to January 13 of
our calendar). Di Benedetto's choice is due to his conviction that, from a
doctrinal standpoint, the group Epidemics)II, IV, and VI, is older than the
group Epidemics)I and III. 21
Looking closely at both sides of this dispute and putting aside all other
considerations about the Hippocratic texts, I find myself forced to admit
that there is no sound reason to identify one as against the other of
Aristotle's comets with the large star of Perinthus. Actually, the situation
is even more uncertain. For the period from 450 to 350 B.c., no other
comet observation was recorded in historical documents in the West, 22but
we are not entitled to think that written records exhaustively mirror the
reality of astronomical phenomena observable in Greece during the time.
For example, we know by calculation that in 392/391 B.C. Halley's Comet
returned, but nothing is said of it in the ancient sources. That date hap-
pens to be very well suited to our Hippocratic report. Unfortunately, the
description there is couched in such vague terms that we cannot be sure if
it is really reporting a comet and not a supernova or even a simple nova.
The Hippocratic writer only speaks of an astir ou smikr6s 'no little star,'
not a kontitesaster 'long-haired star,' as Aristotle calls it. He does not tell
us if the star in question was fixed or moving in relation to other stars, nor
does he indicate its place among the constellations or give any information
on the length of time during which it was seen. If the Hippocratic star was
a nova or a supernova, which is perfectly compatible with the text, no
chronological conclusions can be drawn. 23
Ingenious philologists have sought to date Epidemics)II, IV, and VI, by
using various other clues, such as names that could relate to persons known
from other sources. A patient is said to come from the village of Medo-
sades.24 Xenophon in the Anabasis mentions a certain Medosades to whom
Seuthes, king of Thrace, ceded some villages not far from Perinthus. Xen-
ophon's narrative on this subject is linked to the return to Greece of the
10,000, and Deichgraber therefore infers that the story told in Epidemics)
IV, 45, must be later than 400/399 B.C. 25It seems to me that Deichgraber's
interpretation arbitrarily fixes this terminus post quem at too late a date.
Xenophon simply says that in 400/399 the Greek army supplied itself "in
the villages that were given to Medosades by Seuthes." The account sug-
gests that they were ceded prior to the army's encampment in the region. 26
Consequently, the terminus post quem should be put back a few years, to
the time of Seuthes' rise to power around 410-405 B.C.
If the Cyniscus mentioned in Epidemics)IV and VI, is the same person as
the Spartan ,vho, according to Xenophon, commanded in the Thracian
Chersonnese, the Hippocratic reports on him shoul,d be dated to around
400 B.c. But the identification is very uncertain. There is also a patient
who is identified as "coming from Alcibiades"' (but the phrase could also
mean "coming from Alcibiades"). 27 He was probably a member of the
retinue of an important personage. 28 Deichgraber does not hesitate to
allow that the reference is to the famous Alcibiades, who in fact spent time
in the vicinity of Perinthus until 404 B.C. Surprisingly, the celebrated
philologist concludes that "we must assume that this event and along with
it the accounts in Epidemics)II, IV, and VI, took place a little after 399. " 29
It is a doubly surprising conclusion, since on the one hand, Alcibiades had
already been assassinated by that date and the posthumous use of his name
in the context of the Hippocratic report would be very strange; and sec-
ond, fixing the tlate of one account in no way demands that all other
medical observations in these three books of the Epidemicsbe contempo-
rary to it and, still less, that they all be later than it.
In my opinion, the upshot of Deichgraber's arguments, if the identifi-
cations he proposes are accepted, is that the case history in book 4 that
mentions the village of Medosades must be dated sometime after 410-405
B.c., and that the mention in book 2 of Alcibiades is prior to 404 B.C. It
is even possible to accept the date 399 as a tenninus post quem of the story
about the village of Medosades without fearing, as Deichgraber does, that
it prevents us from accepting the date 404 as the latest for the story of
Alcibiades' man. The contradiction disappears if we admit that the "med-
ical files" gathered up in Epidemics)II, IV, and VI, were made over a series
of years. A date accepted as the tenninus post quem for one specific case
history also secures, to be sure, the chronological limit of the composition
of the whole of book 4, and perhaps also that of books 2 and 6 in their
current form, 30 but it would be a mistake to apply it to each medical
observation in the 2-4-6 group.
Accordingly, we can say that the "cough of Perinthus" may well have
been observed before the end of the fifth century, but it is unlikely that it
occurred much before or much after 400 B.c. 31 It is known that Perinthus
joined the revolt of the colonies against Athens in 411 and that it was
reconquered in 410 by Alcibiades. According to Oberhummer, a specialist
in the history of Perinthus, the acute epidemic fever mentioned in the
second book of the Epidemicsarose at around this time, if not even earlier. 32
This summer fever certainly preceded the epidemic winter cough, but by
how much we do not know: it could be a single year, a few years, or even
more. But we do know that the appearance of the summer fever coincided
with the arrival in Perinthus "around the time of the summer solstice"
(that is, in the second half of June) of a group of itinerant physicians. 33
Who were they? The tradition is unanimous in associating them with the
medical center at Cos. Could Hippocrates himself have been their leader?
My analysis of the chronological clues does not exclude the possibility,
and some circumstances even make it relatively plausible. I think it is a
reasonable hypothesis that the description of the summer fever and that of
the winter cough of Perin thus go back to the same author as the one who
compiled Epidemics)I and III. The latter texts are generally thought of as
among the most authentic: they derive from Hippocrates himself or from
a Coan physician who was his close contemporary. 34 To be sure, this
attribution is not necessarily correct for all the clinical observations and
medical notes that make up Epidemics)II, IV, and VI, in their current form.
To a modern reader, the heterogeneity of Epidemics)II, IV, and VI, is
surprising. F. Robert's description is apposite: "A formless mass of disor-
dered notes, some of which had previously been carefully edited, implying
the desire to produce a real book, but the vast majority of which are
unchanged from the rude state they were in when jotted down by a
physician returning from a patient's bedside. " 35 I do not believe in their
unity of authorship, if by "author" is understood the physician who
actually stood beside patients and noted his clinical observations and re-
flections and generalizing conclusions; but I freely admit the unity of the
compiler of this whole. The contents of these three books are spread out
before us like a collective dossier based on the personal experience of
several practitioners.
Nikitas has brought out some differences between books 2 and 6 as
against book 4. They are a matter of diction and style (for instance, the
use of articular infinitives, adjectives in -odis) more or less learned expres-
sions, and turns of phrase in more or less high style) as well as some
nuances in their teachings and in the way medical observations are used.
Since he did not wish to see that such differences could also be detected
between the several chapters within each of the three books, Nikitas con-
cluded that book 4 is the most recent in its group and that its author was
a second-rate physician capable only of applying a fixed doctrine to what
he observed instead of making new inferences from it. 36 But this attempt
to split the group 2-4-6 did not convince specialists. F. Robert explains
the differences mentioned above as the result of' 'a more or less advanced
state of work in the same author" and, naturally, the part whose form is
the farthest from final form, namely book 4, seems to him to be the oldest
part of the whole. 37
Although Robert and Di Benedetto may be right in refusing to separate
book 4 from books 2 and 6, the arguments in this whole debate reveal
deep inconsistencies in the composition of the texts in question. It seems
to me that books 2, 4, and 6 of the Epidemics)constitute a mosaic in which
the lines that separate the various pieces do not correspond to the tradi-
tional book division. By means of a sophisticated study of the addresses of
the patients, Robert was able to show that from books I and 3 to books 2,
4, and 6, there is an observable reduction in the interest of the physician
in the habitat of each patient, a drift away from the etiological teachings
espoused in Airs) Waters) and Places and toward those found in the treatise
Hunwrs. There is no way of mistaking the author (or, to put it more
cautiously, the editor) of books 2, 4, and 6 for the author of books I and
3. However, this demonstration loses in value if there is a distinctive fea-
ture limited to a part of books 2, 4, and 6, in particular the general
descriptions that rise above the individual case histories and integrate them
in to epidemiological profiles.
But that is exactly the case for the two chapters II, 3, 1, and VI, 7, 1. In
the midst of a formless mass, they stand out; their formal perfection con-
trasts sharply with the unfinished notes and hasty clinical remarks that
surround them. The method used for describing the diseases that prevailed
,f'
in Perinthus during the summer and winter of one or two years around
400 B.C. is exactly the same as the distinctive intellectual procedure that
makes the katastaseisof Thasos so valuable. Between the two chapters just
cited and the epidemiological descriptions in books I and 3, there is an
extraordinary similarity in style, composition, and above all in medical
teaching and "research program. " 38 From my analysis, which primarily
takes into account the effectiveness of the investigative techniques and the
quality of the medical content, the two "constitutions" of Perinthus are
39
in no way inferior to those of Thasos. There are two possible explanations
for this similarity: either all these texts had the same author, or the obser-
vations at Perinthus are the work of a disciple or close collaborator of the
author of books I and 3. If the latter is true, it must be admitted that the
disciple's genius for research and acute clinical observation was not less
than his master's.
According to Galen, learned men in Hellenistic and Roman times attrib-
uted books I and 3 of the Epidemicsto Hippocrates. It was commonly held
that their form was particularly pure since the Coan master himself had
prepared them for publication. 40 Littre says there is no reason to doubt
this important contention. As for books 2 and 6, Galen aligns himself with
the majority of ancient commentators, who held that they contain per-
sonal notes by the great master that his son, Thessalus, edited, having
found them on his father's hides or tablets. Unfortunately, Thessalus filled
them out with his own abundant store of notes. 41
A reading of books I and 3 reveals that Cyzicus on the Propontis was the
furthest from home that Hippocrates ventured, and probably also the last
place he visited to practice his art during the time-frame of the two books.
But Perinthus also lies on the Propontis, right across from the peninsula
on which Cyzicus was situated. What could be more natural for traveling
physicians than to pass from one to the other of these cities? If we accept
that Hippocrates left Thasos in 408 B.C. (a date made credible by the
patient research of Dugand) and that he subsequently visited Cyzicus, his
arrival at Perinthus, a nearby port city and as such a likely stopover, could
not be sooner than the summer of 407 and was probably a bit later. 42
A RetrospectiveDiagnosisof the cccough of Perinthus))

When Emile Littre published his edition of Epidemics)II, IV, and VI, in
1846, he committed an error that has made retrospective diagnosis of the
disease of Perinthus difficult if not impossible; until recently, it even led
astray the philological exegesis of these books as a whole. His error was in
supposing there was one disease of Perinthus, that is, that the description
in VI, 7, 1, refers to one specific epidemic disease. Supposing that to be
true and that the disease in question was very special and rare, Littre
brought to bear on this one historical event scattered notes and observa-
tions that had in common the mention of certain symptoms. He sketched

a synthetic pathological profile and was proud he had reconstituted it using
many remarks scattered throughout the Hippocratic corpus:
Several passages that were very obscure are elucidated by simple cross-reference;
some very short case histories that are barely skeletal become completely intelligible
once they are fleshed out with the general descriptionto which they must be referred;
isolated fragments come together under a single heading; seeing the way remarks
and facts all belongingto a singleaffection are scattered about, we can understand the
nature of these notes that were taken by the Hippocratics and that have survived
to our day under the rubric of the books of Epidemics)·three aphorisms (Aph '.)rv,
31, 33, 54) are attached to them that were inspired by the diseaseof Perinthus) and we
can thus witness the research of the Hippocratics and the elaboration of their
ideas. 43
In short, Littre took VI, 7, r, to be the description of an epidemic in the

sense that term had in the medical language of the nineteenth century
(which is essentially the same as that of our own time) and not in the sense
proper to the Hippocratic writings themselves. In the Hippocratic corpus,
the word "epidemic" designates the totality of diseases observable in a
given place during a given time. 44 A disease that is said to be epidemic, for
instance, the epidemic cough, is a disease that visits a region from time to
time; its appearance is strictly linked to the change of seasons and to
climatic variations.
Analysis of the medical content of this text in the light of modern
epidemiological knowledge proves beyond doubt that the Hippocratic
description of the "cough of Perinthus" has nothing to do with a partic-
ular, specific epidemic in the modern sense of the word; rather, it refers to
what is conventionally called an "epidemic constitution." The Hippo-
cratic writer is offering an overview of the climatic conditions and the
seasonal endemic and epidemic diseases that prevailed in Perinthus over a
specific winter and spring. So the "cough of Perin thus" is analogous, not
to the "plague of Athens," but to the "constitutions" of Thasos. I con-
tend that the Hippocratic description refers to a harmonious ensemble of
pathological phenomena, in other words, to a historical moment in the
pathocoenosis.
From this standpoint, it is possible to foresee the difficulties that awaited
any attempt at a retrospective diagnosis of the "cough of Perinthus." At
least in his first publication on the subject, Littre did not hesitate to choose
a solution that most historians of medicine today would consider a sad,
evasive admission of ignorance: that Hippocrates had observed and de-
scribed an epidemic disease unknown to modern 1nedicine. In his words:
In my reading I have found nothing resemblingthe epidemicof Perinthus) that is, an

affection that most often presents itself as a cough that falls into remission, then
relapses and in its relapsing form is accompanied by either nyctalopia, anginas, or
paralytic weakness in the limbs, or else it attaches itself to a preexisting fever and
produces feebleness or abscesses in the limbs as well as deposits in the ears. Accord-
ingly, until we are 0etter informed, it is necessary to group this disease with those
epidemicsof which no other exampleexists)·it swells the list of such diseases, which are
already rich in singular features; and in this respect, it is of real interest to historical
pathology. 45
When he wrote these lines, Littre was still unaware of the existence of
diphtheritic paralyses. Actually, such paralyses had been mentioned by W.
Piso in the seventeenth century and by C. W. Hufeland in 1811. But not
until 1855, when a certain Dr. Herpin, a physician in Tours, was himself
affected by paralytic complications of a diphtheritic infection, did Pierre
Bretonneau, at that time the leading authority on diphtheria, reveal to the
world in precise detail the existence of this relatively common clinical
feature. 46 At the time that he was preparing his edition of the Hippocratic
text on the cough of Perin thus, Littre could only have known of Breton-
neau's previous article on diphtheria (1826), a masterly work in which the
disease was accurately described for the first time as a properly defined,
specific, contagious clinical entity; but a discussion of the relapses accom-
panied by paralysis of the soft palate and the limbs did not yet exist. 47
Though buried in work, Emile Littre, tireless master of positivistic phi-
losophy, classical philology, and French lexicography, still succeeded in
keeping up with the progress of medical science. He read the clinical
lessons of Trousseau and the monograph of Maingault, and so modified
his diagnosis of the epidemic of Perin thus. 48 In 1861, in a number of the
Gazette medicatede Paris) he declared that it could no longer be considered
an extinct disease. Its symptoms-fever, angina, and cough with consecu-
tive, not concomitant, paralyses affecting mobility (first of the soft palate,
then of the limbs) and perception (notably, disturbed vision)-suggested,
not to say guaranteed, that the disease in question was diphtheria. 49 Shortly
after the appearance of this journal, Littre took up the subject for the third
time in the last look back ("Dernier coup d'oeil") that accompanied the
final volume of his edition of the works of Hippocrates. There he says that
the retrospective diagnosis of diptheria is indeed alluring to a modern
physician, but he no longer considers it obligatory. Having consulted a
monograph by Gubler, a leading authority at the time, Littre had learned
of the existence of paralyses that followed nondiphtheritic fevers. 50 Here
are his final reflections and conclusions on the subject:
Once I had identified everything in the Hippocratic corpus that pertained to this
epidemic and realized by a retrospective diagnosis that it was not a series of dis-
jointed accidents relating to various pathological states but connected accidents
linked to a single pathological state, I stopped and could go no further ...
Today the additional information I was awaiting has arrived. In the world of
science, when facts are unique, they trouble the mind n1erely because of their
isolation. But when they become part of a group of facts, and their links with
something more general can be glimpsed, one is reassured, and one gains confi-
dence in the clarity that such connection brings. That is the case for the epidemic
of Perinthus, since new research on consecutive paralyses in diphtheria or fevers or

certain inflammations has made it possible to order the disease described by Hip-
pocrates alongside others described by modern authorities . . .
These are the four principal traits of the disease: the appearance, during more
or less advanced convalescence, of paralysis of the soft palate, followed by paralysis
of the limbs and impairment of vision in various degrees and in various forms ...
In a related observation, the Hippocratic writer, who was disturbed by the
complications that were arising, took the trouble to say that neither the face nor
the intelligence underwent alteration. Apparently, he feared some brain affection
or other; but when he saw that there was none, he also understood the link
between what he called "the cough" and the paralyses. Nor did he allow himself
to be fooled by the minimal relationship that seemed to exist between the original
stage of the disease and its sequels. His guiding principle was to consider all the
phenomena that occurred in the epidemic as tied to each other. The cough, the
nyctalopia, the paralyses, the angina, even the pneumonias (for he noted several
that were concurrent with the disease), all appeared to him as a whole springing
from a single cause, which in the event was a single constitution. Thence came the
confidence with which he recognized what was before him, not as paralyses of
encephalitic origins, but as paralyses due to the influence of environment . . .
Poor speech articulation and difficulty in swallowing suggest paralysis of the soft
palate, which is the ordinary sign of invasive diphtheritic paralysis. And if it were
true that nyctalopia and consecutive paralysis were exclusive symptoms of diphthe-
ria, it would be difficult to disregard the similarity between the epidemic of Perin-
th us and diphtheria. But the problem is more complex.
The study of consecutive paralyses has just taken a further step. In an important
report, Dr. Gubler has shown that impaired vision and paralyses can occur consec-
utively after convalescence in fevers, inflammations, even simple sore throats, and
that basically even though these complications are more common in diphtheria than
anywhere else, in diphtheria they are the specific expression of a general rule . . .
If we return to the epidemic at Perinthus with that in mind, it is certain that its
arrival was signaled by cases of nyctalopia and consecutive paralysis that occurred
after the remission and then the relapse of a cough. Although the cough in question
was severe, with violent expectoration of dry matter and fatal, suffocating anginas, it
remains to be seen whether the disease was diphtheritic, since consecutive compli-
cations, as Gubler has shown, are not exclusively diphtheritic. So we must stop at
the most general level, at which they are recognized as being part and parcel of the
chief acute diseases. Accordingly, we must consider the epidemic of Perin thus as just
another instance of that fact. The Hippocratic writer has preserved for us an impor-
tant page in medical history: twenty-two centuries ago a physician noted the connec-
tion of an acute disease with consecutive paralysis and nyctalopia. This observation,
which proves the consistency of organic reactions over an extensive period of time,
was buried in obscurity because a mind alive to the relation between original effects
and consecutive ones was lacking for so long. 51
Read a century later by a professional historian of medicine, this lengthy
citation from Littre, for all its clarity, subtlety, and precision, immediately
calls for criticism. First there is the methodological error discussed above:
all the pathological states mentioned in Epidemics)VI, 7, r, are not neces-
sarily the result of a single disease in the modern sense of the word. To be
sure, it is appropriate to attempt a reduction of the diagnosis to a single
clinical entity, but it is not necessary to be obstinate. If the structure of
the medical description points toward a plurality of clinical profiles, it is
dangerous to wish to impose upon them at all costs a unitary and simplistic
nosological account. In the case of the katastasisof Perin thus, the complex-
ity of the underlying pathological events is apparent, at least in the view of
a twentieth-century physician. But Littre's historical setting forced him to
disregard this complexity. The concurrence of sever~ diseases was of no
interest to him, since it would essentially nullify the logical procedure of
Hippocrates. Littre is overjoyed that the establishment of "epidemic con-
stitutions,'' a method that he sees as typical of Hippocrates of Cos and
opposed to the "Cnidian" nosology, provides such unexpectedly impor-
tant results as the discovery of ~tiological links between angina and paral-
ysis. "True Hippocratic medicine" triumphs through its unifying, global
quality, which is the one Littre retains as a model for his own medico-
historical research.
My two other criticisms are the direct result of subsequent advances in
medical knowledge. The diphtheritic toxin can in fact bring on accom-
modation paralysis and oculomotor weakness, but the specific disturbances
that result from it-notably, eyestrain, double vision, and strabismus-are
not to be confused with weak night vision. Although night blindness has
been reported in the course of some modern diphtheria epidemics, it is
now established that it is due to a concomitant vitamin deficiency that
does not belong to the diphtheritic infection itself. Littre understands the
"nyctalopia" of this Hippocratic text as a poorly defined vision impair-
ment, which offers him two advantages: he does not have to decide be-
tween the two classic, contradictory senses of this term (thus avoiding the
need to amend lton-h ·> n, 33) and in addition he can easily include it with
the symptoms of diphtheria. Today, this interpretation seems to me un-
tenable. The text of Epidemics>VI, 7, 1, says explicitly that troubled vision
arose especially in those patients who coughed a little or not at all and who
suffered from mild sore throats. In a diphtheritic intoxication of the eye
muscles, the Hippocratic writer would no doubt have noticed just the
opposite, namely, a heightened frequency of disturbed vision in those who
previously had suffered severe laryngitis or pharyngitis. If we carefully
examine the five Hippocratic passages that speak of "nyctalopia," sound
medicai reasons emerge for restricting the retrospective diagnosis to night
blindness in the modern sense of the word, more exactly, to the acquired
form of night blindness, which is a defect in the eye's adaptation to dark-
ness that results from vitamin A deficiency. 52 Danielle Gourevitch has just
confirmed this identification by a very precise philological investigation. 53
Moreover, it is worth stressing that the Hippocratic physicians knew of
and described in different terminology the pathological states produced by
paralysis of the eye muscles. 54
My third and final criticism of Littre's opinion concerns the limb paral-
yses consecutive upon nondiphtheritic fevers. Specifically, he cites a clinical
observation by the British physician Eade published in 1859 in the Lancet:
the man in question was "down for a month with influenza," recovered,
then got sick again and had symptoms of paralysis exactly like those ob-
served in diphtheria. 55 According to Littre, this case belongs with others
like it collected in Gubler's monograph. He adopts Gubler's opinion that
paralytic complications consequent upon fevers and anginas are not spe-
cific, that even in diptheria they are "just a special instance of a general
rule." But that is incorrect. The paralyses described by Gubler do not arise
after any given infection but, if we exclude diphtheria, they only otherwise
arise in the presence of a neurotropic virus. Significantly, true influenza
does not result in disturbed mobility and sensibility. Reading today the
nineteenth-century clinical histories referred to by Littre, I recognize some
cases of acute idiopathic polyneuritis (otherwise known as the Guillain-
Barre syndrome) and, in particular, an outbreak of poliomyelitis with par-
alytic complications in western Europe right at a time when the general
health was improving.
A retrospective differential diagnosis between diphtheritic paralysis and
that caused by a neurotropic virus is not impossible to achieve. In the first
place, the massive occurrence of paralytic disturbances, even in the form
of small epidemics in a region like that of ancient Greece, is itself a point
in favor of diphtheria. As a rule, the Guillain-Barre syndrome arises sporad-
ically. For very specific epide~iological reasons, poliomyelitis could not
have brought about numerous, grouped cases of paralysis in a place with
the demography and hygiene of fifth-century Greece. Epidemic encepha-
litis has to be considered; I will discuss it further below. The set of symp-
toms makes it possible to distinguish effectively diphtheritic paralyses from
viral ones. We will see how the clinical profile of the latter is a poor fit to
the Hippocratic description of the cough of Perinthus.
Such considerations as these would have quieted Littre's doubts. The
upshot is that the "cough of Perinthus" was indeed diphtheria or, to put
it more precisely, among the diseases mentioned in Epidemics) VI, 7, 1,
there were several cases of diphtheria. A neurologist, Achille Souques, has
adduced additional arguments in favor of this hypothesis. According to
him, a long and strange clinical description included in book 2 of the
Epidemics)56 whose meaning was obscured by the Hippocratic writer's ob-
session with the position of cervical vertebrae, is actually a report of a
diphtheria epidemic. 57 Littre took the description literally and believed
that "the vertebrae of the neck projected forward" in these patients, a
notion that led him to postulate a diagnosis that is totally unacceptable to
modern medicine. 58 There is no such thing as "a spontaneous luxation of
the cervical vertebrae'' that is an acute fever occurring epidemically. If we
shed this error and interpret the so-called projection of the vertebrae as
swollen cellular tissue between the spinal column and the posterior face of
the pharynx, then no other disease than diphtheria is recognizable in the
minute Hippocratic description of tonsilitis with immediate, severe symp-
toms (angina, submaxillary
,, lymphatic reaction, difficulties in swallowing,
speaking, and breathing, fever, neck deformity, and so forth) and its para-
lytic aftereffects (paralysis of the soft palate, occasional paralysis of the
limbs, and difficulty standing up). It is even possible that the observation
logged at Epidemics)II, 2, 24, was made at Perinthus and that it has a direct
link with the katastasisof Epidemics)VI, 7, 1, but of this we have no proof.
The kinship between the two is certain: the pathological events that they
describe cannot be distant from one another either in time or in space.
Given that, the diptheritic nature of the anginas and paralyses at Perin thus
becomes still more likely. 59
However, diphtheria cannot by itself account for all aspects of the
"cough of Perinthus." What about the pneumonia, for instance? Ever
since the first half of the nineteenth century, epidemiologists 60 were con-
ceiving the possibility that the winter epidemic at Perinthus had the same
nature as a Roman plague that they thought was simultaneous to it (dated
by Livy 341-342 A.u.c., that is, 407-406 B.c.). 61 It was thought to be an
influenza epidemic, since "many fell ill, but few died." It must be stressed
that Livy's account is not precise about a single symptom of the disease.
Moreover, at the time of Littre's edition of Hippocrates, the clinical pro-
file and epidemiology of influenza were very poorly understood. A correct
and almost exhaustive symptomatology of it was first established during
the pandemic of 1889-90. In several European cities the incidence was said
to be 40-50 percent, and from 0.5 to 1.2 percent of the population died-
which speaks clearly of its importance. Naturally, the epidemic reached
Greece, and an Athenian physician, Gerasimos Phokas, who fought it on
his own turf, recalled the epidemic of Perinthus and did not fail to relate
the Hippocratic description to the epidemiological experience he had just
lived through. 62 His identification of the "cough of Perin thus" with influ-
enza was enthusiastically accepted by esteemed clinicians like Alexandre
Laboulbene in Paris and renowned medical historians like Theodor Pusch-
mann of Vienna. 63 And their opinion should not be taken lightly, since
they were able to benefit from direct, fresh knowledge of a specific pan-
demic of so-called Asiatic influenza. Phokas was an anticontagionist, and
even in 1892, well after the work of Pasteur, he still thought that the
Hippocratic remarks on the epidemiological role of climatic changes were
perfectly relevant to an influenza epidemic. The main argument of those
who favor this diagnosis is the overall impression created by the clinical
picture that Hippocrates sketches. The following symptoms and compli-
cations of influenza are found in the Hippocratic description: cough, the
fever curve, weakness and pain in the limbs, loss of voice, angina, pneu-
monia. "Nothing is missing," exclaims Phokas, expressing surprise at
Littre's ignorance. "Nothing is missing, not the etiology, not the diversity
of phenomena that mark the various forms of influenza.''
But things are never that simple. The elements of the "cough of Perin-
thus'' that favor a diagnosis of influenza are not pathognomonic. With

regard to the "pneumonias" that were relatively mild and whose outcome
was not serious, it is proper to think of viral bronchopneumonias rather
than pneumococcal or streptococcal ones, but the diagnosis cannot be
restricted to the influenza virus. Besides, in the influenza epidemics of 1889
and 1918,patients did indeed suffer from serious and dangerous pneumo-
nias that were caused by bacterial superinfection. Some details of the Hip-
pocratic text are even less relevant to true influenza. How can one explain
relapses on the fortieth day? Phokas obscures the issue by discussing the
nature of the troubles and not the moment they appeared. The "nyctalo-
pia" has to be transformed into photophobia, if the symptoms of influ-
enza are to be recognized at all costs. Another dubious argument: the
predominance of male victims of the disease. As for the paralyses, their
connection with influenza rests on a confusion with other diseases that
present the same syndrome, namely poliomyelitis and encephalitis.
In 1923,Angelika G. Panayotatou, a physician/hygienist in Greece, posed
the following question: '' According to the observations of modern sci-
ence, ... couldn't this epidemic [the cough of Perinthus] be related to
epidemic encephalitis?" The question is rhetorical, since it is accompanied
by arguments that favor a positive answer. Panayotatou singles out the
paralyses, the eye problems, and the cough as decisive symptoms of epi-
demic encephalitis. 64 Writing in 1923, she has in mind a very specific dis-
ease, the encephalitis lethargica of von Economo, which appeared in
Europe in severe epidemic form during W odd War I. Small epidemics
were reported in several countries, with Greece among them, up until the
end of the 1920s. Apparently, this disease no longer exists. No one had
doubted that it is viral, but the virus responsible was never isolated. Other
viral forms of encephalitis have been proven to exist. Their clinical profile
is so varied that a specific diagnosis is impossible without recourse to
laboratory tests. Current medical belief is that certain viral forms of en-
cephalitis, particularly those originating from mumps, herpes, or entero-
viruses, should have cropped up from time to time in the ancient
Mediterranean population. Still, none of the diseases described in antiq-
uity really corresponds to encephalitis lethargica. In particular, the "cough
of Perin thus" is a poor fit to such a diagnosis. The Hippocratic text makes
no mention of lethargy, convulsions, or Parkinsonian tremor. Paralysis set
in not immediately, but after a relatively long interval. Eye problems dis-
cussed in the text concern vision, not eye movement. And as for the cough
and angina, they have specific qualities that can just as easily (if not more
easily) be referred to other diseases like diphtheria and influenza.
To Henri Favier, the disease of Perinthus is dengue, a viral fever. Al-
though this disease exists nowadays in the Mediterranean, its set of symp-
toms hardly resembles the Hippocratic picture. To begin with, dengue
does not produce a cough or angina. It has a fairly typical way of unfolding
that only partly matches the Hippocratic data. The author of this surpris-
ing identification ~ealizes that "if we are content to compare [the symp-
toms of dengue] as a whole with the description of the disease of Perin thus
in the sixth book of the Epidemics)the analogies are hard to grasp; but if
we study the observations in the second and fourth books, by way of
which the description in the sixth was constituted, numerous analogies
appear. " 65 Favier's hypothesis does not stand up to medical or textual
analysis.
An enthusiastic specialist in infectious diseases, Edward Wilberforce
Goodall, allows that the clinical description in two passages of the Epidemics
(VI, 7, 1, and VI, 7, ro) "sounds very like influenza," but he isolates some
cases of"paraplegia" in them that he feels "suggest acute poliomyelitis." 66
His opinion is not shared by an American physician, John Rodman Paul,
the author of a monograph on the history of infantile paralysis. According
to his study of the Hippocratic texts, acute anterior poliomyelitis is recog-
nizabl~ only in the description of acquired clubfoot (Art.) 62). 67 I pass
rapidly over the hasty and arbitrary judgment of Arcangelo Ilvento, who
sees the events at Perin thus as ''an episode of typhoid fever or exanthe-
matic typhus. " 68 Most historians of medicine nowadays look askance at
this Hippocratic evidence and opt either for diphtheria, or influenza, or a
combination of the two. 69 In the last analysis, the diagnosis of diphtheria
nowadays seems unshakable, but that of influenza remains problematic for
the lack of basic epidemiological information about Perinthus. Here is
what the virologist Pierre Lepine says about the matter:
Nothing permits us to assert that influenza did or did not flourish in anti-
quity. People have singled out a description in the sixth book of the Hippocratic
Epidemicsof a contagious disease that raged five centuries before Christ in the north
of Greece ... The clinical description reminds one of influenza, whose symptoms
it recalls; but the lack of precise data on the incidence of the disease in the popu-
lation as a whole, on the duration of its evolution, and on the mortality rate,
prevents us from eliminating with certainty another disease with an etiology differ-
ent from influenza, such as an adenovirus infection or para-influenza. 70
I had already been immersed in the complexity of this katastasisof Perin-

thus, both in the philological exegesis of the text and in the modern
medico-historical commentaries, when I came across an eighteenth-century
study that surprised me for its originality and the exactness of its assertions.
In a way, it anticipated my own conclusions. In a report to the Societe
Royale de Medecine in Paris, Dr. Chamseru referred to Epidemics)vr, 7,
and stated, "The picture Hippocrates offers indeed resembles what hap-
pens every year in La Roche-Guyon." At first sight, such a comparison
between classical Greece and rural France near the end of the ancien regime
brings a smile to one's lips, but what follows proves that his opinion is not
unfounded. 71 For Chamseru, Hippocratic "nyctalopia" is "night blind-
ness,'' a disease that he knows well from having observed it regularly each
spring in a poor district in France. His zeal for the study of Hippocrates
led him to collate manuscripts in the Bibliotheque du Roi, where he
discovered that the codex R 2254 contains a negative in the definition of
"nyctalopia" that occurs in R-orrheticon)II. So it really concerns "those
who do not see at night." From his own medical experience, Chamseru
concluded that the omission of the negative in most of the manuscripts
"could only have originated in a copyist's error." 72 The "cough of Perin-
thus,'' according to him, was not the result of a single disease:
These winter coughs no doubt answer to several degrees of cough that we moderns
distinguish by name, as rheum, whooping cough, catarrh . . . Some of them were
afflicted with paraplegia, a disease inappropriately confused with paralysis and hemi-
plegia, that was peculiar to the district that Hippocrates was observing, and that
we believe was a kind of rheumatic enfeeblement, very similar to the Indians'
beriberi . . . The text of Hippocrates relates the progress of other diseases of
the same constitution, and specifically sore throats which carried off several vic-
tims in a few days and which Galen called pestilential. Actually, one recognizes
therein, with as much likelihood as in the Syriac ulcers of Aretaeus, the signs of
gangrenous or malignant angina, of which we had several examples last winter
[that is, in 1784-85] in Paris and various provinces of the realm. 73
So Chamseru contends that the epidemic constitution of Perin thus is com-

posed of several diseases, the main ones being, to give them their modern
names, night blindness, influenza syndrome, whooping cough, broncho-
pneumonia, beriberi (or another form of polyneuritis), and diphtheria.
The diagnosis of whooping cough, passed over by more recent commen-
tators, seems perfectly reasonable for a group of diseases that are isolated
from the rest in the Hippocratic account: those "who lost their voice in
fits of coughing'' and who had no fever whatever, or very little, and
suffered neither pneumonia nor paralysis. Chamseru does well to distin-
guish the paralyses of Perin thus from apoplexy, but, since he knows noth-
ing of the existence of postanginal diphtheritic paralyses, he offers a risky
hypothesis: beriberi. In his day, it was a disease practically unknown in
Europe. However, evidence on the nutritional habits of ancient Greece in
no way favors the diagnosis of a lack of thiamine. Beriberi in Perinthus
seems out of the question. But there may have been B group avitaminoses
(or rather, hypovitaminoses) that facilitated the neurological complications
of infectious diseases.
Some Individual Case Histories from Perinthus
In the welter of Epidemics)II, IV, and VI, several passages contribute to

a better appreciation of the nature of the "cough of Perinthus." But they
must be cautiously used. Relying on a certain similarity of symptoms,
Littre and Deichgraber associated Epidemics) VI, 7, r, with various other
Hippocratic texts. "Once I grasped the essential trait of this epidemic as

the author had represented it to himself," writes Littre, "that is, a cough
and paralyses consecutive to it, it was possible for me to discover several
scattered passages which, in relation to it, provide added information. " 74
The procedure is as seductive as it is dangerous. It assumes that the Hip-
pocratic account in Epidemics) VI, 7, 1, is restricted to a single disease.
Furthermore, the disease must be a kind whose epidemic expression is
both rare and localized. Otherwise, we have no right to interpret isolated
symptoms as "debris" from the "cough of Perinthus," that is, as obser-
vations that of necessity relate to a specific place and a specific year. Littre's
first impression, that the epidemic disease in question was exceptional and
extinct, served to justify his way of relating texts, but he should have
become more cautious once he came up with the diagnosis of diphtheria.
In those days in rural Greece, it could only have held sway as an endemic
spreading sluggishly, heating up now and then. Isolated cases and even
other small epidemics must have occurred in the vicinity. The same reason-
ing applies to viral diseases like influenza, which as a rule sweep over very
large areas and return at irregular intervals as small epidemic spurts in the
. .
win tert1m e.
Since the group of itinerant physicians making observations over one
winter in Perinthus was especially struck by certain symptoms-in particu-
lar, a relapsing cough and instances of paralysis after angina-they and their
disciples cannot have failed to speak of them on other occasions. Mention
of the "cough of Perinthus" in the treatise Humors leads me to believe
that the katdstasisof VI, 7, 1, served as a model text for pupils to imitate. 75
So it may well be that even within Epidemics)n, IV, and VI, the references
to it are not contemporaneous with it.
In three successive groups of the seven Hippocratic books called Epidem-
ics) I believe there is a change in overall structure that reveals an evolution
in method. First, in books I and 3, the katastdseisare plainly inferred from
concrete cases that are catalogued apart in lists (those lists are incomplete);
then, in books 2, 4, and 6, there are, side by side, katastdseis)then apho-
risms of broader scope, then concrete clinical descriptions that illustrate
the two previous categories of medical discourse instead of preceding them
on the level of investigative method; and last, in books 5 and 7, there
remain only some vestigial katastdseisburied in the presentation of actual
cases and, though the casuistry has become richer and more varied than
before, there is a failure to formulate new, general rules of prognosis (and,
by implication, of diagnosis).
As for the "cough of Perin thus," it is impossible to prove with certainty
that any concrete case reported in the Hippocratic writings was used to
define the "epidemic constitution" of VI, 7, 1, that is, that it was prior to
it from either a logical or a chronological standpoint. However, even
though the links of certain passages with the "cough of Perinthus" are
not as direct as has been maintained, a medical historian can infer supple-
mentary information from them concerning the diseases present in north-
ern Greece at the time of Hippocrates and use that information indirectly
for retrospective diagnosis.
Let us look closely at a specific instance in which an identification with
the "cough of Perinthus" was made without sufficient justification and
probably incorrectly: 76 "The female who had her right arm and left leg
paralyzed following a cough (she coughed only briefly and insignificantly)
suffered no other change, nothing in her face or her intelligence. Not very
severe. On the twentieth day things took a turn for the better, at about
the same time as the patient had her period, perhaps for the first time,
since she was a young girl. " 77 In this clinical report, the principal fact is
the crossed paralysis of one side's arm and the other's leg. Described here
for the first time in the history of medicine, this peculiar detail distin-
guishes the pubescent girl's case from all the diseases that make up the
"cough of Perinthus." Although a cough is mentioned in the story, that
is only to emphasize its distinct insignificance. We are not told anything
about the space of time between the cough and the appearance of the
paralysis. To be sure, the physician who composed this report was espe-
cially attentive to the cough-paralysis sequence. Supposing that the link
between relapsing cough and paralysis was not discovered until the "cough
of Perin thus,'' then this particular observation must have followed the
composition of the katastasisin VI, 7, 1. To me that seems very likely, but
not proven, since it is not impossible that the discovery of postanginal
paralysis preceded the masterly description of it in Epidemics)VI, 7, 1. Given
the insignificance of her upper respiratory problems, the peculiar location
and relative mildness of her paralysis, and also the ways in which it was
cured, it is possible to exclude diphtheria as a diagnosis for this young girl.
The disease in question might be acute anterior poliomyelitis, 78 but prob-
ably the most likely diagnosis is conversion hysteria. 79 The physician's
distrust concerning anamnesis, expressed in connection with the patient's
menarchy, says a good deal about the customs of this society. From a
medical standpoint, the remark about the absence of change in the face or
intelligence of the patient is a valuable sign that betrays the broad clinical
experience of this particular physician. Without it, one would have diag-
nosed an organic brain lesion.
Here is the passage that, more than any other in the Hippocratic corpus,
resembles cases included in the katastasisof Perinthus: "Among the pa-
tients who were coughing, those who worked with their arms, like the
child who twisted withes and the son of Amyntas, had only their right
arms paralyzed. In both cases, that ended, and then they suffered, having
the cough. Those who rode horseback or walked, in the loins, in the
thighs. Most were dry, and if not, strong in any case. " 80 My translation of
THE ''COUGH OF PERINTHUS'' 331
this Hippocratic account differs a good deal from those of Littre and
Fuchs. The difference is due in part to emendation of the text, and in part
to the preservation of obscurities that are inherent in its clipped style. The
first sentence is not without ambiguities concerning the age and social
position of the victims mentioned, since pats can mean "child" as well as
"slave," and ho Amunteo is a vague expression that can refer either to a
member of Amyntas's household or specifically to his son. Beginning with
the second sentence, the text can be interpreted and completed in such a
way as to make it resemble the account of Perin thus as closely as possible,
for instance, "in both cases . . . the cough ended, then they suffered from
paralysis)having the cough oncemore. Those who rode horseback or who
walked had paralyses(or, alternatively, had pains) in their loins and thighs.
Most of the coughswere dry, and if not, in any case they were strong.''
In comparing this passage from Epidemics)IV, 50, with the analogous
part of VI, 7, 1, one cannot help being struck by certain similarities in the
'
symptoms mentioned and especially in the words used to describe them.
For Littre and Deichgraber, there was no doubt whatever about their
having a common source. As far as they are concerned, Epidemics)IV, 50,
simply reports observations made during the "epidemic" of Perinthus.
Since it is a more concrete description, it is somewhat prior to the com-
position of Epidemics)VI, 7, 1, or, at the very least, contemporary with it.
This conclusion is troubling to Nikitas, however, who, after a detailed
philological examination, concludes that IV, 50, is a kind of generalized
reprise of VI, 7, I. For him, the former is later than the latter, though it
derives from observation of the same pathological event. 81
In my opinion, the common historical origin of the two texts in ques-
tion is in no way proven. To begin with their medical content, it is neither
apparent nor even likely that the two reports derive from observations
made during a single epidemiological event. Philological comparison obliges
one to conclude that the author of one knew the other's formulations, or
else that they are both by the same author. It is easier to explain the
situation if we assume that Epidemics)VI, 7, 1, is older than IV, 50. It is not
necessary to conclude, as Nikitas does, that the latter is an extract in more
generalized form of a prior description that has come down to us piecemeal
through VI, 7, 1. On the contrary, IV, 50, appears, medically speaking, to
be a first observation made by someone whose knowledge of VI, 7, 1,
provided him with vocabulary and an orientation. It appears as though
the author of IV, 50, was anxious to make known what he had observed in
a similar situation to the katdst:asisin VI, 7, 1, and to make even more clear
the relationship between coughing, physical exertion, and paralysis. This
method of investigation is exactly the same as the one that the author of
the account of Philiscus of Thasos used to study the clinical importance of
an epistaxis occurring in the midst of an acute attack of fever. In Epidemics)
IV, 50, the circumstance that engages the writer's interest is that the paral-
ysis is restricted to the right arm. Both patients "worked with their arms,"
but "had only their right arms paralyzed." Nothing of the kind was
noticed during the "cough of Perinthus," though it is just the sort of
observation that the Hippocratic physicians highlight in general epidemi-
ological presentations. So here is a new fact that was not observed within
the framework of the katastasisin VI, 7, 1, and it was important to bring it
to light in order to refine the prognosis. What is said about patients who
walked or rode looks like a repetition or confirmation of previous experi-
ence. Finally, I point out that most of the coughs in the kattistasis of
Perinthus do not seem to have been severe; on this minor point, the two
texts also seem to diverge.
The symptoms enumerated in Epidemics)IV, 50, are too summarily pre-
sented to allow for a retrospective diagnosis. The disease could be polio-
myelitis, diphtheria, the Guillain-Barre syndrome, and so forth. Both
victims were very young, but they are spoken of in a vague way that leaves
one wondering about their age. Some philologists and historians relate the
following story to IV, 50: ''Preferably the positions that give relief; like the
person who wove or twisted withes with his hand and suffered intense
pain when lying down, but by grasping an ankle over his head was
soothed. " 82 From the standpoint of its medical content, this case has
nothing to do with the previous ones. The only common feature is the
patient's type of work, an activity that was certainly not rare in Greek
villages. The problem here is not motor function (the pain is relieved by
grasping an ankle), but a painful condition in the arm. By translating the
Greek word huperoduneon [very painful] 'souffrant cruellement,' Littre
adopted literary rather than technical language that happens to facilitate
his interpretation of the case as a paralysis of the arm related to those in
Perinthus. Actually, the diagnosis of this very sharp pain is not difficult: it
is an occupational cramp. 83
The sixth book of the Epidemicscontains a paragraph that is situated not
far from the katastasisof VI, 7, 1, and is surely related to it. It looks like an
afterthought by the person who observed the "cough of Perin thus." The
nosological conceptualization process is identical to the one that marks the
first and third books of the Epidemicsand the treatis~ R-ognostic:observation
of specific cases and meteorological conditions, classification of clinical
profiles and their integration into katastaseis)and, finally, differentiation of
syndromes through prognosis. After writing his composite sketch (VI, 7,
1), the same physician continues to elaborate the empirical data in two
complementary directions: by generalizing them (formulation of rules in
the form of "aphorisms") and by limiting them (formulation of excep-
tions). Both tendencies emerge clearly from the following text:
For consumptives, the fall is bad. So also the spring, when the leaves of the fig tree
are as long as crow's feet. In Perinthus, most [fared poorly]. The accompanying
cause was a cough that stayed through the winter. The same thing happened for
the other chronic diseases, whose presence was confirmed where heretofore it had
been unclear. Nevertheless, there were chronic diseases in which that did not
happen, for instance, in those with kidney pains and also in some other patients,
like the one I was brought to by Cyniscus. 84
The sentence that adverts to the events at Perinthus is puzzling. It cannot

be made out unless one looks at the aphorism it exemplifies, the one right
before it, which speaks to the exacerbation of consumption. 85 Littre sug-
gests that this text be read in the light of accounts in Epidemics)IV, 47; rv,
49; II, 2, 9; and IV, 53. I think an association willful for the first two
citations, likely for the third, and practically certain for the last. The pa-
tient whom Cyniscus brought this physician to see suffered from an acute
disease that grew worse on the seventh day and better on the fourteenth. 86
The Hippocratic description of his case suggests pneumonia or an influ-
enza syndrome.
Epilogue:DiseasesR-esentin Perinthus around 400 B. c.

The clinical profile of the "cough of Perinthus" can be understood in
the framework of modern pathology only if its unity is broken. Among
the various nosological elements of which it is constituted, one can recog-
nize some viral diseases (influenza syndrome caused by rhinoviruses and
enteroviruses, perhaps influenza in the strict sense, too), bacterial infec-
tions (notably, severe forms of diphtheria and pulmonary tuberculosis as
well as a small-scale epidemic of whooping cough), and deficiency diseases
(above all, vitamin A deficiency, whose pathognomonic expression is night
blindness; perhaps also Band C hypovitaminoses).
Did influenza in the strict sense of the term, "the last great plague,"
flourish in the Mediterranean in classical times? To attempt an answer to
this question, it is necessary to consider not only the varied clinical aspects
of the influenza syndrome but also the results of recent research on the
properties of the influenza virus, the cyclical variations in the immunolog-
ical relations between it and humankind, the regular recurrence of influ-
enza pandemics, and the role played therein by demographic density and
animal stocks. The germ of this disease, M yxovirusinfluenzae) which was
isolated in 1933 by Smith, Andrewes, and Laidlaw, provokes a potent im-
mune reaction in humans that determines the halt of its great epidemic
waves. However, the reaction is not enough actually to eliminate the
disease, since the virus undoes the immunological defenses by a continual
hybridization process between human strains and those that are parasitic
in animals. Influenza is a disease of domesticated animals and of birds that
live in provimity to mankind. In modern times, the general rule is that the
periodic genetic recombination of the human influenza virus takes place
in Asia, in the vast frontier lands between Siberia and western China.
There the pandemic waves arise and proceed with astonishing speed to
infect in cycles all of humanity. Not a single historical influenza epidemic

began in the Mediterranean-even the so-called Spanish epidemic is
wrongly named, since it, too, came from Asia. 87
We now have precious historical information about influenza from se-
rological archaeology. It has been proven, for instance, that the pandemic
of 1889 had the same immunological properties as the "Asian flu" pan-
demic of 1957, which followed the ''Spanish'' one. The Asiatic and Spanish
types alternate with some regularity. Every sixty or seventy years, one gives
way to the other, and each cycle is broken around the midpoint by an
influenza pandemic brought on by a change of immunological subtype.
Going back in time, it is possible to apply the modern epidemiological
schema to former epidemics with the features of influenza, but no earlier
than the fourteenth century. 88 The cyclical scheme seems to be invalid for
antiquity, and at least in the case of Rome, a city whose plague records are
relatively complete, the absence of regularity cannot be explained away by
a lack of historical sources.
To a historian of medicine reading the current literature on influenza,
the relevant point is that its clinical manifestations are not specific. No
description of symptoms, then, can guarantee its retrospective diagnosis.
Instead, the course of its epidemics and their periodicity are the disease's
characteristic signs. The description of the "cough of Perinthus" corre-
sponds in part to the influenza syndrome, but we know today that a wide
variety of viral infections can produce such a syndrome. The overall nature
of the epidemiological events at Perinthus is not at all favorable to a
diagnosis of influenza. Three other texts have been cited to support the
presence of this disease in ancient times. They are from three historians of
Rome, Dionysius of Halicarnassus, Livy, and Cassius Dio, and they con-
cern plagues that ravaged the city in about 488, in 412/413 (or 407/406),
and in 43 B.C. 89 But in fact all three accounts are so laconic and vague that
no retrospective diagnosis is possible. The M yxmnrus influenzae has such
genetic suppleness that it is sure to have undergone profound changes
since antiquity. Its ancestor in classical times was not necessarily as patho-
genic for humanity as the strain that decimated the world's population in
1918. It is even legitimate to wonder whether the demographic density in
Hippocratic times was great enough to sustain the germ in its virulent
form.
It is certain that the ancient Greeks and Romans suffered from coryza or
the common cold. 90 The Hippocratic writers were careful to note that a
runny nose is unfortunate when it occurs alongside some other distur-
bances, particularly in the lungs. 91 But these symptomatic coryzas are not
colds in our sense of the word. The true "head cold" is described in the
Ancient Medicine. This text informs us that coryza is one of those things
"that everyone often experiences and will continue to do so," and also
that the Hippocratic physicians consider it a kind of model to use in
l'
explaining the notion of crisis. Right when the cure begins, an abrupt
change takes place in the organism: what flows from the nose becomes
thicker, and the inflammation subsides. 92 The author of the treatise Airs)
Waters) and Placesconsiders colds a physiological purging of the brain that
must take place when winter ends in spring; without them there is a risk
of serious disease, especially hemiplegia. 93
In modern societies, coryza or acute rhinitis is as trite as can be, a
common disease that is an embarrassment in several ways. It is a benign
affection for which there is no means of cure or prevention. Strictly speak-
ing, it is not a disease at all, but a syndrome that can belong to the clinical
profile of a specific disease like influenza, whooping cough, poliomyelitis,
and others, or it can stand alone in a variety of forms affecting the several
levels of the respiratory tract. Experience and common opinion tell us that
its main cause is the "weather, that people more often catch cold after a
local or general chill. For medical science, chills are only a factor that favors
this ailment, not something sufficient or indispensable to its occurrence.
The necessary condition for the common cold is a viral infection. Nowa-
days we know of at least a hundred viruses that can cause it. They belong
to the M yxmnrus) Adenovirus) Entermnrus)and especially Rhinovirus groups.
Contemporary Western man is subject to such an infection an average of
two to four times a year.
Although the cold has been a relatively common affection since classical
times, it is reasonable to wonder if its frequency in the past was as great as
it is now. No statistical comparison is possible, but it is interesting to note,
as a token of the lower incidence of colds in ancient times, the absence of
allusions to acute rhinitis in Greek and Roman comic writers. They only
mention runny noses in the elderly.
Whooping cough or pertussis, a disease caused by the bacillus Hemophi-
luspertussis (discovered in 1906 by J. Border and 0. Gengou), is older than
most manuals of the history of medicine allow. For chroniclers and physi-
cians of the past, it was often an inapparent disease, since it usually kills
newborns and young children. Because of that, whooping cough was
confused with other, undifferentiated causes of infant mortality.
As I indicated earlier, none of the diseases described in antiquity corre-
sponds to encephalitis lethargica (von Economo's disease), but it is certain
that several other forms of viral encephalitis appeared early on in the his-
tory of the Mediterranean world. In particular, there are notable derma-
tological accounts that testify to the ancient existence of the Herpesvirus
group. 94 We now know that a viral agent of the herpes group is the cause
of two diseases that are very different in their clinical manifestations: chick-
enpox and herpes zoster (shingles). The former is a generalized disease that
probably corresponds to the viral primo-infection, while the latter is the
localized manifestation of a nervous lesion that seems to be due to late
reactivization of the latent virus. In a passage in the Hippocratic corpus,
there is talk of "herpetic eruptions above the groin, developing toward

the side and the pubis. " 95 A modern physician could not find a more
concise way to describe certain typical forms of herpes zoster. We must
suppose, then, that chickenpox was also a disease that the Hippocratic
physicians encountered in the exercise of their art. In modern times, it is
the mildest of the fevers that cause skin eruptions, but it can take on severe
forms and produce fatal respiratory and neurological complications, espe-
cially when there is a secondary bacterial infection. Although the tradi-
tional view of medical historians is that chickenpox was unknown in
Hippocratic times, 96 I think that the new interpretation of a Hippocratic
text by Paul Potter is plausible. According to him, there is a nosological
description in Diseases)111, that essentially corresponds to chickenpox with
a bronchopneumonic complication. 97 What favors this diagnosis is chiefly
the moment when the exanthema appears and its exclusive localization on
the back and chest. Under the epidemiological conditions of archaic and
classical Greece, there is no need to think of chickenpox as exclusively a
childhood disease.
There is no reason to believe that the ancient inhabitants of the Medi-
terranean knew of measles and rubella, two infectious, eruptive diseases
that are extremely contagious and caused by specific viruses very close to
the one that causes mumps. Measles is thought to have existed in China
from time immemorial, but no trace of it is to be found in Greek and
Roman documents. However, the presence of mumps (epidemic parotitis
brought on by a paramyxovirus) in the Greek city-states is known from a
local epidemic of it observed on the island of Thasos in about 410 B.C. 98
The mention of orchitis following parotid swelling guarantees the diagno-
sis. That the majority of victims were adolescents proves that mumps was
no longer a new disease there.
If acute anterior poliomyelitis (Heine-Medin disease, caused by viruses
of the Poliuvirusgroup) did exist in the Greco-Roman world, it must have
been endemic, about the way it was recently in the Near East or Latin
America: a generalized, immunizing infection in children, with a few seri-
ous paralytic cases. It is a well-known paradox that the great prevalence of
this viral disease goes hand in hand with its clinical invisibility. Some
Hippocratic descriptions of paralyses that occur after a catarrh syndrome
could relate to poliomyelitis, but the diagnosis is very dubious. They are
better explained as diphtheritic intoxication or even, in one case (Epidemics)
11, 2, 8), as conversion hysteria. There is no certain classical reference to
the acute phase of poliomyelitis, but it is possible that the disease was
responsible for some chronic deformities. That is probably true for the
porter named Ruma whose stele is preserved in the Ny Carlberg Glyptotek
in Copenhagen. The atrophy of his lower right limb looks exactly like a
result of poliomyelitis. 99 So this Egyptian monument from the Pharaonic
period makes likely the existence of poliomyelitis in the whole Mediterra-
nean from the archaic period on. However, it is far from certain that this
disease is the cauS'e of the foot deformities described in the Hippocratic
corpus, particularly those in chapter 62 of the treatise Joints. John R. Paul,
the greatest specialist in the history of poliomyelitis, believes such a diag-
nosis, 100 but in my opinion he has let himself be misled by the English
translation of the Hippocratic work in question. The author of Joints
speaks of congenital clubfoot and clubfoot acquired through trauma, and
he appears to be ignorant of clubfoot acquired through paralysis. 101
The paralyses observed at Perinthus were flaccid, and certainly the result
of damage to peripheral neurons. An Aristotelian text proves that spastic
paralysis was also known in antiquity. In the Nicomachean Ethics) the con-
flict between reason and impulse in an incontinent man is illustrated by an
analogy to the body: the soul, says Aristotle, finds itself exactly in the same
situation as the body of some paralyzed persons whose "crippled limbs,
that they wished to move to the right, instead swing to the left. " 102 This
passage allows us to recognize the existence in classical Greece of Little's
disease, that is, of spastic paraplegia with clubfoot and scissor gait as the
result of a precocious encephalic affection. 103
As for the anginas that the Hippocratic writer describes during his stay
in Perin thus, it should be made clear that the term kunangkhe (or sun-
angkhe) in ancient medicine designates "any obstacle to breathing or swal-
lowing in a body-part, whether it be the back of the throat, above the
lungs, or in the stomach." 104 Basically, it is a constrictive discomfort that
can result from a local infection of the oropharynx and the larynx as well
as a disease of the bronchi or of the heart. What I have just said clarifies
the pathological situation for a modern physician, but it also introduces
into the explanation of the ancient term some anachronistic concepts and
some connotations that it cannot have borne: the Hippocratic writers were
not only ignorant of the role of heart ailments in angina pectoris, they did
not even distinguish between the oropharynx and the larynx. More recent
physicians, especially Aretaeus and Galen, made that distinction perfectly
clear, but no one in antiquity succeeded in understanding symptoms orig-
inating in the heart. Clinical descriptions from the classical period include
various ailments under the term kunangkhe) including membranous phar-
yngitis, pharyngeal and peritonsillar abscess, ulcerative-necrotic pharyngi-
tis, herpetic pharyngitis, and diphtheria of the pharynx and larynx, to say
nothing of tuberculosis of the ·upper respiratory tract and angina pectoris.
(There is an important infectious disease that appears to be excluded from
this list: scarlatina or scarlet fever. No certain case of it is known in Europe
before the last centuries of the medieval period.)
Diphtheria seems to dominate ancient descriptions of angina. It is a
fairly contagious acute disease caused by a specific microbe, Corynebacterium
diphtheriae or the Klebs-Loffler bacillus (discovered in 1883 by E. Klebs).
The germ first attacks the upper respiratory tract and produces a sore
throat with a characteristic formation of false membranes covering the

tonsils and obstructing the larynx. It secretes a toxin that can damage
tissues far from its original focus and so create myocardial lesions, paralyses,
digestive disturbances, and so forth. In modern times, before its treatment
with serum (1890), diphtheria was a scourge in Mediterranean countries.
In the prevalence and severity of its forms, it was particularly devastating
in Egypt, Syria, and Greece. Aretaeus of Cappadocia, who has provided us
with a masterful description of diphtheritic sore throat, explains that the
disease was called Egyptian or Syrian ulcers, since it was particularly ram-
pant in those two lands. 105 Aretaeus's account dates from the first century
B.c. It was followed by some very pertinent observations made by Rufus
of Ephesus and Aetius of Amida. But the history of diphtheria in Greece
is older still. Aphorisms 24 and 31 of the Hippocratic treatise Dentition
concern this disease, since the first asserts that "in ulceration of the tonsils,
the presence of something resembling a spider's web is not good," and
the second speaks of the voice change that appears when such ulcerations
extend over the uvula. 106 The subject here is the appearance of paralysis of
the upper palate after an ulcerous sore throat, a phenomenon carefully
noted by the ancient writer in its clinical consequences but poorly under-
stood in its anatomical extension. In my opinion, the second type of
angina described in the Appendix of the Regimen of Acute Diseasesdefinitely
includes, among other throat ailments, laryngeal diphtheria, just as the
first type of angina described in Diseases)n,' contains, among other forms
of pharyngitis, pharyngeal diphtheria. 107 This reinforces the credibility of
the diagnosis of diphtheria in the framework of the "cough of Perin thus."
One final word on the subject of Hippocratic "night blindness": its
origins in nutritional deficiency are evinced in the therapeutic success of
ingesting raw beef liver. 108 Diseases do not associate with one another in
an arbitrary way. We now know that nonspecific wintertime infections of
the upper respiratory tract are aggravated by the lack of vitamin A. In
ancient societies, night blindness regularly occurs in the presence of bron-
chitis and sore throat. Vitamin B deficiencies facilitate the neurological
complications of toxic-infectious diseases. Chronic alcoholism (that is,
consumption of wine accompanied by a low-protein diet) is not a negligi-
ble factor for a true etiological appreciation of the '' epidemic constitution''
of Perinthus. To be sure, Hippocrates does not breathe a word about it,
but that is because the cultural acceptability of such a diet made it intellec-
tually invisible. Finally, I add that Bretonneau long ago stressed that cases
of diphtheria are exacerbated during an epidemic of influenza syndrome. 109
How is it that a practitioner of the eighteenth century, Dr. Chamseru,
had clearer vision in his interpretation of the "cough of Perinthus" than
all the physicians and philologists who succeeded him? The reason is that,
unlike Littre and most recent commentators, Chamseru avoided the mis-
take of fixing his gaze in a· single direction, of insisting on the preservation
of the nosologicaI unity in the modern sense of pathological specificity.
He also, by virtue of his personal experience in a rural environment not
yet tainted by the Industrial Revolution, was able to grasp another kind of
unity in the Hippocratic picture: the balance of symptoms in the seasonal
appearance of a pathocoenosis.
Chapter Thirteen
A DIALOGUE BETWEEN A
PHILOLOGIST AND A PHYSICIAN
It has been my privilege to collaborate with Fernand Robert, a professor

of ancient Greek literature and civilization at the Sorbonne, in the prepa-
ration of an edition of the Epidemics for the Collectiondes Universitesde
France. Our collaboration began shortly after the Hippocratic colloquium
at Strasburg in 1972, in which we were both participants. The colloquium
had allowed philologists and physicians interested in Hippocrates to meet
and judge the benefits of working together. As far back as 1933, K. Deich-
graber had concluded a fundamental work on Hippocrates with an appeal
in that vein: henceforth, he asserted, philologists need to call on physicians
for help if they are to progress in their study of the Hippocratic corpus in
general and of the Epidemicsin particular. 1
This chapter faithfully reproduces, with minimal changes to suit the
conventions of written exposition, selected parts of the actual dialogue we
held as we prepared the edition (with commentary and translation) of
Epidemics)VII. 2 The point is not so much to publish secure results before a
work in progress is completed-since the paleographic research is in pro-
gress, there has not as yet been a review of some more recent manuscripts,
to say nothing of the questions posed by Arabic and Latin traditions-as
to bring out the problems encountered, to illustrate the n1ethod followed,
and to show how the work of both participants is intertwined in the
translation and even in the choice of variants to retain in establishing the
Greek text. 3
Among the texts that we have already studied together, we chose (1) a
short but complete text that allows for detailed discussion; (2) a very long,
remarkable text-we could only touch upon its highlights; (3) several con-
cise texts that are so rich in content that the modern physician can fairly
340
DIALOGUE BETWEEN PHILOLOGIST AND PHYSICIAN 341
easily postulate a diagnosis within the framework of modern pathology;

and (4) a group of texts in which it is possible to show, from either a
philological or a medical standpoint, that the original account of the ob-
servation has come down to us in truncated forn1.
A Case of Chronic Tetanus

I offer below a translation of chapter 8 of book 7 that we have agreed
upon and that we intend to justify. It is based on the Greek text edited by
Littre, 4 which was carefully checked and revised on the basis of current
knowledge of the Hippocratic manuscript tradition.
In the woman who lived over the Gates, already elderly, there was a slight fever,
and, at the moment wheh it was about to end, a pain in the nape of the neck
extending down the back to the loins; she did not have complete control of these
areas; moreover, clenched jaws, and squeezing the teeth together too hard to let a
probe pass through; her voice articulated badly because her body was inert, with-
out movement and without strength; but she was conscious. With fomentations
and lukewarm mead, she improved on about the third day; and then, with barley
gruel and soup, she returned to good health. Took place at the end of autumn.
Our discussion follows:

Iwbert: For this text we have only one ancient manuscript, which is M,
since there is a very long lacuna in V (it stretches from chapter 5 to chapter
11).5 The M manuscript allows for several solutions here that accentuate
the rapid, somewhat telegraphic style. Our translation should keep the
look of notes taken in haste or, more precisely in this case, of slips put in
final form with an extreme concern for conciseness. The place where Littre
proposed a conjecture (epi tous instead of heoutous)has the same impossible
reading in M as Littre found in the manuscripts that he knew. His conjec-
ture is an excellent one, and setting it next to the actual reading is enough
to explain how the mistake happened. As for the rest, the problems in the
text or in the translation will depend-I say this from the outset-on the
medical interpretation. It is appropriate, then, to ask you now whether or
not you can propose a diagnosis on the basis of Littre's text.
Gnnek: This case contains an excellent description of trismus, a tonic spasm
of the muscles of mastication. 6 We can exclude a local ailment of the
mouth and the temporo-mandibular joint because of the patient's gener-
alized illness. This particular symptom can occur during an epileptic fit, in
the last stage of rabies, during strychnine poisoning (for instance, with
nux-vomica) or, and above all, as the basic sign of infection with Clostrid-
ium tet:ani. In this case, the clinical picture is certainly not that of a grand
mal epileptic fit. Rabies and strychnine poisoning can also be eliminated
on the basis of the absence of agitated movement and clonic spasms. The
patient is calm, immobilized, with consciousness intact, not shaken by
342 DISEASES IN THE ANCIENT GREEK~WORLD
convulsions, agitated, or delirious. Her return to health would also tell

against a diagnosis of rabies or massive poisoning.
A single possibility remains: tetanus. This disease definitely existed in
classical Greece. Three other case histories in this same book relate to it
(Epidemics)vn, 36-38). 7 In all three, Littre proposed it as the diagnosis.
Why does he not do so for the one that interests us here? Probably because
of the rapid and apparently complete cure. In Littre's time, attenuated
and chronic forms of tetanus were poorly understood. Besides, Littre did
not even know that it was an infection and so defined "tetanus" only in
clinical terms. His diagnosis of "tetanus" always corresponds to acute
tetanus in modern terms, usually a fatal form of the disease unless there is
a particular medical intervention. I mention in passing that the Hippo-
cratic physicians used the term tetanos to designate a pathological concept
that in some contexts overlaps the disease we call by that name and in
others does not. 8
This particular case is a chronic tetanus. Painful contractions along the
nape of the neck and the back are very characteristic of it, and this writer
describes it in precise, lapidary style. Three days after a brief and harsh
attack that is preceded by a fever and features spasms, immobilization of
the whole body, and voice deformation, the symptoms begin to recede.
The convalescence in this disease is long and marked by frequent remis-
sions. So we may well ask ourselves about the actual duration of this
particular observation. Was the physician really able to follow the patient's
progress after the stated period? Was he an itinerant physician who never
stayed very long in the places where his patients resided? Tetanus must be
preceded by a wound, but none is mentioned in this account. However,
in these relatively benign forms of tetanus, the wound can be minimal and
the incubation period very long (from ten to thirty days). So it's not
surprising that the Hippocratic physician did not see a relation between
the physical trauma and the paroxystic state.
R.: In book 7, there can be two reasons for an unfinished observation: you
know that books 7 and 5 seem 1to me to have been copied at different dates
by two medical teachers or even students from a file in which each took
the observations or the parts of observations that interested him and left
the rest. 9 That is the first reason why an observation can be unfinished,
but your study here shows that missing links, like the wound and its
incubation, or the convalescence, can be part of the original observation
itself. It is not that the writer (that is the term we must reserve for the
person who prepared the file, calling the persons to whom we owe book 7
in its present state an editor) was a poor observer (the whole book is proof
to the contrary). In the conditions he worked under, there were limits to
the duration of his observation, either, as you suggest, because he was an
itinerant physician, or simply because the patients at the time were not as
thin-skinned as they are now and so would not have had recourse to a
physician except when they suffered a great deal. There are other cases in
which the physician lets us know that he himself was not present at the
outset of a disease. 10
Let's get into the details. For the word trakhe/,os)I understand that you
do not always accept the translation "neck." Here, is it "neck" or "nape
of the neck"?
G.: For medical reasons, it is clearer to translate this term here as "nape
of the neck,'' since probably we can consider the nape as part of the neck,
and in this case we risk misleading the reader if we do not, because the
word "neck" nowadays designates especially the anterior and lateral por-
tions of it. For another example, see Epidemics)vn, s, where the discussion
is of the sign of meningitis,\ so it must be "nape of the neck."
R.: Besides, from the standpoint of the language, I can offer no objec-
tion. From the Liddell-Scott-Jones dictionary,
,,
I learn that in Plato trakhe/,os
is probably distinguished from aukhen 'nape of the neck,' but that
aukh~n) used in I+ognostic)23, is glossed as trakhe/,osby Galen. 11 So the
medical interpretation must rule the day. For the peace of mind of Helle-
nists, however, it would be best to explain this in a note once and for all
at the first instance of trakhe/,ostranslated "nape of the neck," then to
cross-refer to it explicitly every time we translate it as such.
G.: Littre, who was a physician and a philologist, was certainly correct
in translating the word rakhis here as "back" and not "spine."
R.: Again, it's the physician who must decide how to render ou panu
engkratestouton: since ou panu can mean "not at all" or "not completely,"
with the first meaning explicable as a litotic use of the second, as when we
say that someone is not very nice, meaning that the person is not nice at
all. So should we say here that the patient "has absolutely no control over
these areas" or that he "does not have complete control" over them?
G.: That he "does not have complete control" over them: the patient
is not suffering from a true paralysis, and the physician seems to me to be
carefully distinguishing these symptoms from the absolute impossibility of
voluntary movement.
R.: Now we are faced with the biggest problem. To be sure, we accept
Littre's correction epi tous. But right after pareinai) M adds the two words
ouk en to the text of the recentioresthat Littre used. The effect is to change
both the grammatical construction and the meaning. Were it not for these
two words, we would translate "clenched jaws, and squeezing the teeth
together too hard to let a probe pass through." "Squeezing" is a transla-
tion of the word ep[ that gives it the value of a second verb coordinated
with "clenched," even if we can say that xunegmenai is to be understood
with kaf; and the construction "too hard to" governs the infinitive parei-
nai) an epexegetic infinitive. But if we add ouk en) the sense will be "and
it was not possible to pass anything over the teeth larger than a probe."
The infinitive paretnai is then a complement to ouk en.Clearly it is no
longer possible to render epi by "squeezing," and a real weakness of this
solution is that the words "pass anything over" suit only the lower jaw!
But the main problem is that the teeth are no longer as tightly clenched as
they were in the first translation, since· now the probe can pass. It is
incumbent on us to pay careful attention to all the ancient readings that
were unknown to Littre and to prefer them whenever possible. Is there a
medical way to accept the version in M? And what exactly is this probe?
G.: It's a metallic probe with the shape of a knife .whose blade is dull
and whose handle ends in a small ball, whence its name (mile) which recalls
the Greek word for apple). Such probes have been found in excavations. 12
As for the jaws, in this disease they are very tightly clenched, and Littre's
text and translation are satisfactory. If it were absolutely necessary to accept
the reading in M, I would have to imagine that this practitioner-I'm sure
he tried to insert his probe, I can see him doing it-finally succeeded
because the woman was old and in Greece dentition was often poor;
plausible, but still!
R.: Let's not go too far. It happens often enough that the oldest manu-
scripts provide an absurd reading and that the later ones restore the text,
whether because a good corrector had the ingenuity to discover the right
reading or because it occurred in an old manuscript otherwise unknown
to us. 13 So we're perfectly entitled not to idolize the M manuscript, the
more so since this error is of a known type. The construction without ouk
enis perfectly good Greek, if a bit weak, and someone who did not
understand it thought he was making the text clearer by adding to it. So
we can stick with Littre here.
As for the patient's voice, the change that the disease produces and that
is designated by the word pselli cannot be stuttering, since Aristotle draws
a distinction between psellosand traulos 'stuttering. ' 14
G.: And in tetanus the voice does not stutter. It is distorted and hoarse
instead. Littre is wrong here. Stuttering is a symptom of psychological or
neurological origins, but what is affected here are muscles in the larynx
and the glottis. The voice is intelligible but very distorted.
R.: It remains for us to interpret the word paralelumenon;I really hesitate
to speak, as Littre does, of paralysis, although that is the French (as well
as the English) word derived from it, since it is clear that the writer is
searching for the proper term. As little as purely stylistic issues concern
him (which, by the way, does not prevent him from being unconsciously
artistic, given his powers of observation), he would not have started a list
of three words with the harshest and most precise one. And you have just
brought out how careful he is to avoid confusion with paralysis in the strict
sense of the term; it is significant that his effort in this regard is particularly
plain just before he uses a word that, if transcribed as such into French or
English, would introduce the very notion he is at pains to avoid.
G.: Yes, I do not like the word "paralysis" here at all, since it suggests
hen1iplegia or something similar. The body in such cases of tetanus is
immobile but not really paralyzed.
R.: Having in mind the definition of paraluo in Bailly' s school dictionary
as "relacher les muscles d'un cote du corps" [relax the muscles on one
side of the body], I thought of translating it "without muscles, without
movement, without strength."
G.: Uh-oh! Above all, don't speak of muscles, because the Hippocratics
never gave them a role in movement, which to them was produced only
by bones and ligaments.
R.: So much for that ineptitude. I had in mind an expression that was
also as untechnical as possible, even a little familiar. Looking at it another
way, what is the difference for modern medicine between paralysis and the
state described here?
G.: Paralysis is the impossibility of voluntary movement, but here there
is immobility without it necessarily being impossible to move when de-
sired. This is something not situated on the level of the central nervous
system. It's a difficulty in the transmission of the nervous impulse, or a
disposition to avoid pain.
R.: How about translating it "inert"?
G.: I'd have no objection.
R.: So when I ask you, as I just did, to define for me a difference between
two modern notions, the concern is to understand and make understood a
difference between them that the ancient observer also perceived, just as
when we were speaking of the difference between psellosand stuttering.
Someone may object to our rejection of the word "paralysis" here because
the ancient observer did not have the notion "nervous impulse." But we
are not imputing to him a concept that was unknown to him. It can happen
that the Hippocratic writer uses a word from our vocabulary in a sense that
is technical for him but not the same as its modern technical sense, or, again,
it can happen that we believe that a term designates for him just what it
designates for us, but that our belief is mistaken. In both instances, it is
likely without being absolutely inevitable that we would use the modern
term and add a note to avoid any confusion in the reader's mind. A note is
also obligatory when we do not translate it with the modern term, to explain
why not. Such a note would be necessary here to explain ,vhy we do not
translate the word paraluo as "paralysis" in our sense, on the grounds that
the author is struggling to express something that he knows to be different
from paralysis, as his successive approximations demonstrate. In practice, a
note is always necessary except when the old and new senses of the word in
question coincide perfectly.
DISEASES IN THE ANCIENT GREEIC WORLD
A Case of TyphoidFever
From the lengthy observation of Hermoptolemos's wife, which takes
up two and a half pages in Littre, I append a translation that the two of us
have agreed upon and whose detail we will not take up as we did for VII,
8. 15 Instead the discussion is in tended to show the way we are able to work
together on the commentary to such texts.
In Hermoptolemos's wife, wintertime, fever, headache; when she drank, as if it

was hard to swallow; she got up and said that her heart was being crippled; her
tongue was livid from the outset. First manifestation taken for the result of a chill
after bathing; awake night and day; after the first days, when asked, not only her
head, as before, but her whole body was suffering. Thirst, sometimes intense,
sometimes moderate. From the fifth and sixth days to the ninth, she was almost
raving, then she began talking to herself in incomplete phrases; drowsiness with it;
sometimes, she tried to grasp the plaster wall with her hand, or she would apply a
cool pillow from beside her head to her chest; from time to time she kicked off
her bedclothes. In her right eye was the characteristic spot of blood, and a tear;
urine, with the appearance that is always bad in children. Stools at first yellowish,
then watery and the same color. On the eleventh, the heat seemed to diminish
and at times the thirst ceased, to the extent that if nothing was given her she did
not ask. Moments of rest, after the initial stage, occurred satisfactorily by day, but
at night she did not sleep, and she suffered at night especially. On the ninth, her
belly was disturbed, watery substances, on the tenth as well; the days previous, fits
of rage, whining like a small child, outcry, frights, looks in every direction every
time the drowsiness broke. On the fourteenth, it was a job to take and hold her
just for a moment, she was jumping and shrieking with all her might as though
smitten with horrible pain and terror; afterward, she was again calm, drowsy and
sluggish without seeing and at times without even hearing; she switched frequently
from one state to the other, from tumult to calm, almost the whole day. The
following night, she excreted some bloody substances, somewhat viscous, then
again somewhat muddy, and after that, of a very pronounced leek green, or black.
On the fifteenth, brusque movements, and some frights; and her shouting grew
fainter, but the exasperation, the anger, the weeping continued if one did not rush
to do what she wished. As soon as the first few days passed, she again began to
recognize persons and objects, and the spot on her eye was gone. But her folly,
her incoherence, her shrieking, followed by the slipping into drowsiness I spoke
of above, kept up. She could hear only irregularly, at times perfectly even if one
whispered, at other times one had to speak loudly; feet ~ways as warm as the rest
of her body until her last days. The sixteenth, less severe. The seventeenth, though
she was in a more moderate state than the other days, she curled herself up in a
ball like someone shivering, and her fever went up. Great thirst; other things as
before; there was shaking in her hands, and she shook her head; bruises around
the eyes and hostile looks; the thirst was violent. As soon as she drank, she asked
for more; she would grab the container, and drink all at once; you couldn't take
it away from her again; tongue dry, bright red, and the whole mouth and lips were
dry, ulcerous; possessed by the shakes, she would put her hands to her mouth and
nibble them, and whatever you offered her to chew or swallow, she would gulp
down and swallow greedily and furiously; her look was angry. Three or four days
before the end, shivering would come over her at times, so that she curled up,
covered herself up, and gasped for breath; contractions, the legs; chills in the feet;
thirst as ever; getting up to relieve herself either to no purpose or for the sake of a
little thin stuff, exerting herself for a short time. The last day, the twenty-third,
eyes wide open in the morning, quick glances all around, at times she was calm,
and without covering herself or dropping off. Toward evening, the right eye
moved as though it saw or wanted something, from an exterior angle toward the
nose; she recognized and answered a question posed to her; she began to stutter,
and because of the shrieking, her voice was broken and hoarse.
R.: Surely this is one of the most beautiful texts in Greek, and its
movement is remarkable, as well as the quality of its observed details; a
work of art, even if no artistic intention is apparent in it. It may be that I
have thrust one upon it by translating the patient's own words at the
beginning of the text ''her heart was being crippled''; there are not enough
attestations of the verb giuJ5ofor us to know with certainty if the overuse
of this word, whose proper meaning is "amputate" but which ended up
meaning "weaken," had reached the point where we should here speak of
a simple "weakness" in the heart. The physician has taken the trouble to
note that it was the patient's own word, and I think that is a sign in favor
of an expressive translation of a word that he thought striking. 16 Every-
where else the prevailing aesthetic is simplicity.
There are two older manuscripts at our disposal here, M and V, at least
from the word "first manifestation" (pr6phasis)on, in the second sentence.
That is where the long lacuna in V ends. The differences from Littre's text
that can be established on the basis of their readings are relatively insignif-
icant: an otiose mention of the eleventh day after the tenth in the phrase
concerning the ninth day and those following it, and an absurdity (visitors
who recognize the patient, rather than the other way around) in the
sentence after the fifteenth day-again a case of a reading in the newer
manuscripts that is superior to the older ones, though we should not come
to think of that as a general circumstance.
The word pr6phasisis used in the beginning, remarkably, in its strictly
etymological sense, to mean the first phenomenon to appear in a devel-
oping process. Its causal sense developed from this one in all thinkers,
especially medical ones, who did not waste words and who wished to
scrutinize a reality that was visible and tangible. 17 We particularly refrain
from searching for a causal connotation in this term as the result of our
regrettable tendency always to translate it "pretext" or "immediate cause,"
as though there were another, more profound one. Actually, for this
writer, there is no cause at all. The issue is to discover what is the first
phenomenon that is definitely part of the process he is trying to describe.
Neither the bath nor the chill is such a phenomenon, which must be the
first attack of fever that follows them. As for this phenomenon, he only
knows of it by doxa: edokei'was taken for.' The physician was summoned
once the fever had already lasted for some time. It can also be considered
DISEASES IN THE ANCIENT GREEI< WORLD
likely that, since the first observation properly speaking (it is the first one
dated in relation to the onset of the disease) was on the fifth day, the dates
are recorded from the pr6phasison. Considering the importance of these
dates in the description and in Hippocratic doctrine, that is an essential
item to determine.
In response to your constant urging, I have, needless to say, avoided
translating the Greek word koma as "coma" as Littre does, since in almost
every instance it designates a state in which the subject remains con-
scious.18 I proposed "torpor" and we finally agreed on "drowsiness." It
should really be made a rule that whenever a Greek word has a precise
meaning in the translation that is radically different from the sense in
which it was used by the Greek medical writers, the word must not be
translated as its cognate in the language of the translation. But we are now
in your territory-I think I have gone as far as I can without a medical
interpretation, which is ... ?
G.: Two aspects of the narrative-the remarkable description of move-
ments that Galen calls "carphology" (gathering bits of straw), 19and the
states of confusion, the "mild delirium," which, in this patient, alternates
with manze 'rage, wildness, folly' 20-suggest, at the outset, three different
infectious diseases: typhoid fever, a common type of septicemia, or a
meningococcemia. But the color of the stools, the yellow ochre that is
characteristic of typhoid and is still used as a diagnostic sign nowadays,
points us in the right direction. In all its aspects, this passage is a beautiful
description of typhoid fever. 21 Or rather we should call it salmonellosis,
since without bacteriological analysis there is no way to distinguish be-
tween typhoid fever in the strict sense, caused by Salmonella "t)tphi)and
some other diseases caused by the germs of the same genus (paratyphoid
fevers). 22
Intermittent deafness is common in this context. It is due to a bacterial
injury of the brain and not to ear lesions. The observation about the
patient's feet is noteworthy: it is there because the physician is trying to
establish a differential diagnosis. In exanthematic typhus-its clinical man-
ifestations were known to the Hippocratic physicians, who thought of
them as particularly deadly forms of acute fever-there are often complica-
tions in the extremities. Moreover, by his silence with respect to a certain
symptom, the author gives away the differential diagnosis in question: he
makes no mention of an exanthema. Actually, it is a symptom of both
typhoid and typhus, but in the former it is very unobtrusive (lenticular
spots that are often hardly visible, and then only through a careful search
of covered parts of the body), while in the latter it cries out for attention
(purple spots that turn brownish). So the typhic state in Hermoptolemos's
wife is not a rickettsiosis. The purple eruptions it produces would not have
escaped the attention of this particular clinician, while the light rash ac-
companying typhoid fever could have no special significance to him, nor
would it arouse his interest enough for him to mention it.
There is another interesting omission: enlargement of the spleen. The
physician's silence with regard to this symptom, which he considers mi-
nor, does not always denote its absence in a patient. If the issue of spleen
size never comes up in this case, that is of no consequence for the diagnosis
I have proposed, since its enlargement in typhoid is soft, painless, and not
necessarily palpable. Modern medicine knows it from percussion, a proce-
dure unknown to the ancient physicians.
Without treatment, a typhoid victim's fate is usually decided around
the end of the third week of the disease. At that point, the fever falls, the
delirium lessens, and slow ,recovery can begin. That is what happened to
Hermoptolemos's wife, but her organism was not really able to recover,
and she fell prey to disturbances of her hydric and electrolytic metabolism
(note the red tongue, the split lips, and the appearance of her face). The
chills stem from septicemic attacks. An intestinal hemorrhage, which is a
very common complication in salmonelloses, signals the end.
In this whole description, there is but one surprising remark: the move-
ment of her right eye on the day of death. Usually, one eye does not move
without the other. If the observation is to be taken literally, there was
paralysis of the left eyeball, a fairly rare occurrence. In any case, it is the
first known description of nystagmus. To my mind, the words used are
simply an elliptical way of describing the direction of the nystagmus in
both eyes, since the rhythmic jerk in only one eyeball is directed toward
the nose. 23
R.: That's a likely possibility, in view of the willfully elliptical style of
this writer. He may well have thought it pointless to specify the movement
of the other eye. Would you let me quibble with you a little over the
expression "differential diagnosis"? It's just a matter of which words to
use. What delights the modern reader of Epidemics)vn, is to construct a
diagnosis from the elements that are so remarkably well described by the
ancient observer. For him, on the other hand, there is no talk at all of
diagnosis, and if, rarely, he mentions the name of a disease, he never once
asks himself if a patient is suffering from one disease instead of another. In
this respect, he remains strictly within the tradition of the treatise R-ognos-
tic) whose last chapter says, in essence, "If you complain that in my work
there are not enough names of diseases, you are mistaken, since they are
all there"-this despite the fact that he actually provides very few. Even
so, it is also true that this physician does do what you say he does, that is,
he carefully notices one symptom or another because experience has taught
him about different cases in which the foreseeable processes were distinct.
In all this, it seems clear that his way of thinking was the best one possible
before the existence of microbes and viruses became known.
G.: As you say, our disagreement is only an apparent one. And since we
are discussing ways of thinking that our writer forbids himself to engage
in, I am also struck by the consistency with which he avoids speaking of
causes. For once, however, in the text in question, he has allowed himself
to mention one, since it seemed so obvious to him: I mean where he says
that toward the end, the patient's voice became hoarse as the result of
having cried out. But, alas, he is mistaken! Her voice was hoarse not from
crying out, but from a specific infection of the larynx caused by the bacil-
lus. It's an utterly typical symptom.
R.: The patient's cries constitute a structuring motif in the account and
left a lasting impression on the physician's mind. Or rather, her cry, since
the word is always in the singular in the text. I wished to draw attention
to it by translating the word boi with the word "clameur" [outcry]. To
stress the vivid impressions engraved in this observer's memory, it was also
important to render the meaning of the Greek definite article when he
speaks of "the" spot of blood and not "a" spot, even though he had not
spoken of it previously. There is no reason to hesitate to add, as I have,
the word "characteristic" to reproduce what is in fact a consistent sense
of the Greek article-not that it was characteristic of the disease (which
would be to reason like a diagnostician and contradict what we have just
said), but characteristic of this patient. I would not be so bold as to
translate the article's force here with "this spot of blood that struck me
so" or "that bothered me so much," but that is really what's being said.
Translating it "her" spot of blood would be inadequate, not clear enough.
lvur ConciseCaseHistories
R.: A text can be much shorter than this account of typhoid and still be
complete. I know how much you admire chapter n6 of this same seventh
book for the concision with which it describes a case of breast cancer in a
few lines:
In a woman, at ,Abdera, a carcinoma developed on the breast; its appearance was
so: a bloody ikhor flowed from the nipple; once the flow 'stopped, she died. 24
G.: Modern medicine confirms the diagnosis of karkfnomaproposed by

the Greek physician. 25 The case in question was probably an epithelioma.
Breast cancer, with typical lymphatic spread and cachexia, is well described
in another text from the Hippocratic corpus. 26 Indeed, it looks as though
then, as now, it was a relatively common disease in women of a certain
age. A famous patient at the end of the sixth century B.c., Queen Atossa,
daughter of Cyrus and wife of Darius, is thought to have suffered from it.
Herodotus tells us that she '' had a tum or (phuma)on her breast that burst
open and was slowly spreading.' ' 27 However, the complete cure that was
effected without mutilation of the breast by the physician Democedes of
Craton leads one' to suppose that Atossa's disease was a mastitis, not a
malignant tumor. In fact Herodotus, who is always very precise in his
choice of technical terms, avoids speaking of cancer in this instance and
instead uses the term phuma) which can designate an abscess as well as a
neoplasm. 28 As for the ikh?Jr)the word signifies the sero-purulent discharge
from a wound, the exudation from a sore, the juice of animal flesh, and so
forth, that is, what flows when fleshy parts of the body are liquefied (to
be sure, this last part of the definition is not a description of reality but
the way the ancient Greeks represented it to themselves); recent philolog-
ical research has shown that even in Homer the word ikh?Jrrefers to such
a liquid and not to "the Qlood of the gods," as is commonly thought. 29
R.: Here are three other case histories from the seventh book of the
Epidemics. They appear to me to pose no problem whatever from the
standpoint of modern medical interpretation:
In the commander of the large ship, whose right forefinger, bone and all, had been
crushed by the anchor, an inflammation appeared, a dry gangrene, and some fever.
He was mildly purged on the fifth day. The bouts of fever subsided then as well as
the pain. A part of the finger fell off. After the seventh day, an ikhlJr came out
properly. Afterward, he said that with his tongue he wasn't succeeding in explain-
ing everything. Prediction: the backward spasm is on its way. The jaws began
clenching, then it went to the nape of his neck; on the third day, he was totally
convulsed backward, with sweating; on the sixth day after prediction, he died. 30
The prorrhisis (prediction) seems,, clearly to mean what we usually call the
prognosis, but the word prognostikonhas a broader sense in Greek and in
the Hippocratics in particular. In my opinion, we should avoid translating
prorrhisis "prognosis," as Littre does. By inserting the words "is on its
way,'' I am restoring in this text an expression that occurs in its parallel
passage in Epidemics)v. Its presence here is, to my mind, guaranteed by
the existence, in M and the vulgate, of an unintelligible addition to the
text adopted by Littre (his text ,vas based on V).
At Cardia, in Metrodoros's little boy, after a toothache, dry gangrene of the jaw,
horrible overgrowth of flesh on the gums; there was moderate suppuration; the
molars and the jawbone became detached. 31
The man wounded in the head with a stone thrown by a Macedonian had a cut
over his temple equivalent to a scratch; when struck he had seen stars and fallen.
On the third day, he stopped talking; a fever, not very high; beating temples, as
from mild heat; he heard nothing, was not conscious, would not stay still; perspi-
ration on the forehead and from below the nose to the chin. He died on the fifth
day.32
G.: The first of these three cases is a perfect description of tetanus. 33 The
expression "backward spasm" is unusual, but it corresponds to the root
meaning of the word in the text, opisthot6nos.This word made its way
unchanged from Greek into modern medical parlance to designate the

phenomenon under discussion in this case, namely, the tonic contraction
of all posterior muscles along the spine. To keep it in a modern translation
of the text would imply that the original used highly technical language,
which is untrue. Needless to say, the physician's prediction of opisthoto-
nos and death is a triumph of the Hippocratic method of clinical
observation.
The expression '' dry gangrene'' gives a good sense of the pathological
reality referred to by the words sphakewsand sphakelismos.In the first his-
tory, it is an osseous gangrene of traumatic origin, and in the second, a
necrosis of the facial bones caused by noma. Noma is an ulcerative and
gangrenous stomatitis occurring in undernourished children. It is actually
a complication of infectious diseases that cannot occur without gross weak-
ening of the immune system. There was a time when noma was widespread
in Europe, but nowadays it is never seen in the industrialized world. Three
other cases of the same type are described in Epidemics)IV, 19, and v, 4.
They should be considered an indication of the low level of infant hygiene,
particularly in times of crisis (wars, food shortages).
In fact the third history, a remarkable analysis of the stages of dying in a
case of intracranial hemorrhage, does introduce war. Information like this
about combat with a Macedonian and, in the case that follows this one in
book 7, on the death of a warrior wounded at Delos by a javelin-is it of
any use in dating these texts?
R.: Books 5 and 7 were certainly composed during times of war; but war
is hardly unusual in the history of Greece. Still, these two books are
unusual in the Hippocratic corpus and among the Epidemicsfor reporting
actual war wounds; and there is talk of battles on Delos and Datos. Several
details seem to fit the historical situation that may have obtained around
and about 357 B.c. 34
Casesof EpilepsxPneumonia) and Amebic Dysentery

R.: The case histories we have looked at up to this point appear com-
plete, but here is proof that other texts have been cut short:
For Anechetos's boy, the same thing: wintertime, giving himself a rubdown, he
had gotten warm at the fireplace in the bathhouse and suddenly got epileptic
convulsions. Once the convulsions subsided, he started looking around, not yet
conscious, but the next morning, after coming to, he was again seized: a convulsive
state, but not much foaming at the mouth; the third day, only some bits of speech;
and the fourth, he could only express himself by movements of the tongue, his
voice faltered, he was unable to speak and stopped at the beginning of words. On
the fifth day, his speech was very disturbed; the convulsions came on, and he was
not in possession of his senses; then, when all that ceased, his speech only barely
returned to a normal state. 35
,f'
On the sixth day, the account comes to an end in words that Littre
translates as follows: "Le sixieme, ayant ete mis a !'abstinence de tout,
decoction d'orge et boisson, il n'eut rien et ne fut pas repris" [the sixth
day, put on abstinence from everything, gruel and drink, he had nothing
and was not taken again]. The Greek text Littre translated is as follows:
hektaroi aposkhomenoipanton kai rhophematon kai potott) oudin) kai ouk eti
eldmbanen. Here is the text as we now have it from M and V: hektafoi
aposkhomenoipanton) kai rhuphi matos kai potou ouketi eldmbanen (a text that
Littre knew in almost exactly this form by way of C, a sixteenth-century
manuscript that is almost always an exact copy of V and what he called the
vulgate). This text from the oldest manuscripts has to be translated, "the
sixth day, put on abstinsnce from everything, he did not even touch gruel
or drink.'' And we can be sure that this text is authentic. Why? Since we
can see so clearly how one text became the other. The text in M and V
could in no way conclude the account. So the physicians who were using
it around the end of the Middle Ages urged the copyists to tease a conclu-
sion out of it somehow, rather than abandon the reader on some bland
detail of the patient's treatment and in ignorance of the final outcome. In
C attempts are still being made to clean it up. The first step was to add the
word "nothing," "none of," rather than understand the construction of
the verb "take" with the genitive (as in "take of the drink" = "take
some/any drink"); once the word "nothing" was added, it acquired the
sense of a complete sentence without a verb and was taken to mean ''noth-
ing more," that is, that the patient had no more attacks; finally, the verb
"take" received a new subject, an understood subject that was no longer
the patient but instead the disease itself: "(the disease) took him no more."
These shorthand phrases are not at all improbable given our author's
elliptical style, but it is also perfectly natural to think that the phrase
originally had the form it still does in the old manuscripts. Or is it that the
conclusion of the account was stated clearly enough in its first words: ''For
Anechetos's boy, the same thing"? But what is "the same thing"? One
would be hard put to discover any specific resemblance between this ac-
count and the one that precedes it. Nor am I reluctant to believe that the
expression is further proof that we do not have the whole text, or even
that the account that preceded our chapter 46 in the original collection
was not necessarily the one in our chapter 45.
G.: Epilepsy was of great interest to the physicians of the classical age. 36
The Hippocratic treatise the SacredDisease)which was certainly earlier than
this text, is specifically about epilepsy and tries to desacralize it. The case
you have just cited shows how the clinical study of epileptic states was still
intriguing physicians. The gaps that philological inquiry discovers in chap-
ter 46 coincide with silences in the description from a medical standpoint
and with the problems it poses. In other chapters as well, one gets the
impression that the narrative has been truncated or abridged; for instance,
in chapter 13.
R.: Here's a translation of it:
The shopkeeper from outside the walls, the one with pneumonia, 37 his belly let
loose right away. On the fourth day, lots of sweat; the small fever seemed to have
stopped; small cough, almost nothing. The fifth, sixth, seventh days, the fever
took over. Sweat on the eighth. The ninth, his vomit was yellow. The tenth, his
stool was compact, infrequent. Around the eleventh, relieved. Around the four-
teenth, cured. 38
G.: This history begins on the fourth day, without providing any indi-
cation of what took place previously: the editor (he was probably not in
fact the one, you call the writer, who made the original reports) cut
everything that had come before by using, flatly, the name of a disease,
giving a diagnosis, which is against his custom, and calling the patient "the
one with pneumonia." This editor is eager to get to what interests him:
diarrhea in a case of pneumonia.
Likewise, in observation number 55, the usual clinical report on the
beginning of the disease is lacking.
R.: So it is. Here is this particularly concise chapter:
In Cleotimos 's shoemaker, whose bowels had been loose for a long time, a fever
came over, and a swelling in the form of an abscess on the liver descended to the
lower abdomen; bowels remained loose, then another abscess on the liver rising
along the hypochondrium. He died. 39
G.: No information on the patient's initial state; nothing about his

urine; and likewise, no precise chronological information. Even the way
of saying that a tumor develops "on the liver" is inconsistent with the
usual form of such descriptions, in which the writer describes what he sees
and not what he imagines within the body. How does he know that this
tumor develops on the liver rising along the hypochondrium? Are we to
imagine a dissection, or even an operation? It is assuredly not our writer's
custom to make us worry about the conditions under which he knows
what he is observing-he who is so careful to distinguish clearly between
what he has learned by doxa) by hearsay, as against what he reports as his
own observation. It is hard to imagine him resorting to exceptional inves-
tigative methods without telling us about them; it is much easier to imag-
ine that the editor meddled with the account after it was written, giving a
summary of his interpretation of the case in order to cut out a part of the
original that didn't interest him.
Despite the case's abbreviated presentation, a retrospective diagnosis
does not seem to me impossible. Chronic diarrhea and abscess of the liver
suggest one thing: amebic dysentery. The ameba in question, Entamoeba
hystolytica)is a parasite common to all primates. It is ubiquitous and found
often in Greece to this day. Sensitized by this example, I think I recognize
it in several other passages in the Hippocratic corpus (particularly in Epi-
demics) III, and tne Coan Prenotions).It even seems to me that the disease
of Theodoros's wife (or daughter) in Epidemics)VII, 24, is nothing other
than a suppurating hepatitis of amebic origin.
R.: So the signs of cuts in the descriptions in book 7 confirm our
suspicion that this whole book is only the vestige of a collection of obser-
vations that was once much richer. Long after the collection was first
made, perhaps even during the second century B.c., 40 some student or
teacher of medicine chose the pieces that have come down to us and threw
out those he thought useless for his studies or his teaching. But he still left
us with a good deal of it, so that today it interests us more than many
other texts in the corpu~, not only because the quality of the observations
is remarkable, but also because here, the philologist is almost on virgin
ground. Despite Littre's perspicacious, positive judgment, nineteenth-
century scholars and our contemporaries as well have generally continued
to leave book 7 outside the purview of their research. In doing so they
have made themselves abiding victims of the distrust that Galen showed it
long ago.
G.: My interest in this text is enlivened by the relatively pure state of
the observations it contains-they are as free as possible from the general
concepts (on the correlation between "meteorology" and diseases) that
are telling in Epidemics)I and III, but muted here. At times these ancient
facts seem to lie before me like a patient on a hospital bed.
Notes
Introduction
I. This discussion restates some ideas presented at the First Hippocratic Collo-
quium held in Strasburg, 1972. See M. D. Grmek, "La realite nosologique au
temps d'Hippocrate," in La collectionhippocratiqueet son role dans /Jhistoirede la
medecine(Leiden, 1975),pp. 239-55.
2. For the history of the concept of disease, see E. Berghoff, Entwicklungsges-
chichte des I(rankheitsbegriffes)2d ed. (Vienna, 1947); W. Riese, The Conceptionof
Disease)Its History,Its Versions)and Its Nature (New York, 1953);G. Canguilhem, Le
normal et lepathologique)3d ed. (Paris, 1966); M. D. Grmek, s.v. Bo/est [Disease] in
Medicinska Encicklopedi_ja) 2d ed. (Zagreb, 1967), 1: 490-529; P. Diepgen, G. P.
Gruber, and H. Schadewaldt, "Der Krankheitsbegriff, seine Geschichte und Prob-
lematik," in Handbuch der allgemeinen Pathologie (Berlin, 1969), 1: 1-50; D. D.
Copeland, "Concepts of Disease and Diagnosis," Persp. Biol. Med. 20 (1977):528-
38; and A. L. Caplan, H. T. Engelhardt, and J. J. McCartney, eds., Conceptsof
Health and Disease(Reading, Mass., 1981).
3. For example, we can cite the works of Danielle and Michel Gourevitch on
the representation of diseases on clay statuettes of the Hellenistic era. See especially
"Terres cuites hellenistiques d'inspiration medicale au Musee du Louvre," ltesse
Mid. (1963), pp. 2751ff, and "Temoins d'argile," Abbotempo 1 (1965): 16-21. The
authors of these remarkable studies have succeeded in identifying the following
pathological states: acromegaly, achondroplasia, senile cachexia, leprosy, hydro-
cephalus, ascites, umbilical hernia, hemorrhoids, facial paralysis, trismus, eyelid
ectropion, rhinophyma, elephantiasis of the leg, torticollis, gibbosity, and the
Klippel-Feil syndrome (fusion of the cervical vertebrae). Although the material in
question is later than mine, I stress the value of this research from the standpoint
of method. In classical and archaic Greek art, bodily deformation appears much
less frequently than in Hellenistic statuettes, but that could be due to aesthetic
sensibility rather than the absence of the diseases. Nevertheless, the fact that gaiters
are never represented in Greek pictorial or plastic art is revealing and confirms
other signs of the absence of endemic hypothyroid cretinism.
4. Such an approach is taken by C. G. Gruner in his Morborum Antiquitates
(Breslau, 1774), and followed by A. Hirsch, Handbuch der historisch-geographischen
Pathologie)2d ed., 3 vols. (Stuttgart, 1881-86);J. D. Rolleston, The History of Acute
Exanthemata (London, 1937);A. Pazzini and A. Baffoni, Storiadellemalattie (Rome,
1950);W. R. Bett, ed., The Historyand Conquestof Common Diseases(Norman, 1954);
and many others. Even the most recent authors do not stray from the beaten path:
E. H. Ackerknecht, Geschichteund Geographieder wichtigstenI(rankheiten (Stuttgart,
1963);F. Henschen, The Historyand Geographyof Diseases(New York, 1966); and M.
Sendrall et al., Histoire culture/lede la maladie (Paris, 1980).
5. M. D. Grmek, "Preliminaires d'une etude historique des maladies," Annales
357
358 NOTES TO PAGES 4-9
E.S. C. 24 (1969): 1437-83; an updated German translation of this article is to be

found in A. E. Imhof, ed., Biologiedes Menschen in der Geschichte(Stuttgart, 1978),
pp. 79-96. The first public use of this term was in my contribution to a discussion
at a medical colloquium in London in 1966. See F.N.L. Poynter, ed., Medicine and
Culture (London, 1969), pp. II9ff.
6. See G. Herdan, "The Mathematical Relation between the Number of Dis-
eases and the Number of Patients in a Community," J. Roy. Statist. Soc._,A, 120
(1957);320-30; C. B. Williams, Patterns in the Balance of Nature and Related Problems
in Quantitative Ecology (London and New York, 1964); and N. T. J. Bailey, The
Mathematical Theoryof Infectious Diseasesand Its Applications) 2d ed. (London, 1975);
see also J. Atchison and J.A.C. Brown, The Log Normal Distribution (London,
1957).
7. Hesiod, Works and Days) 11.109-15, trans. H. G. Evelyn-White (Cambridge
and London, 1914; rept. 1977), p. II.
8. Worksand Days) 11.174-82, pp. 15-17.
9. Since the Renaissance, we speak of her box, not jar.
10. Hesiod, Worksand Days) 11.90-100, p. 9.
II. In this regard, see R. M. Frazer, "Pandora's Diseases, B1l]a 102-104," Gr.
Rom. & Byz. Stud. 13(1972):235-38,and A. Casanova, La famiglia di Pandora. Analisi
fi!ologicadei miti di Pandora e Prometeonella wadizione esiodea(Florence, 1979).
12. Plato, &public) 405c-406c.
13. Celsus, Prooemium)4-5, trans. W. G. Spencer (London and Cambridge, 1935;
rev. and rept. 1948), p. 5.
14. For constants in diagnostic procedures and in the description of symptoms,
see C. Habrich, F. Marguth, and J. H. Wolf, eds., Medizinische Diagnostik in
Geschichteund Gegenwart (Munich, 1978), especially the contributions of G. Freiser,
"Diagnosis und diagignoskein. Zurn Krankheitserkennen im Corpus Hippocrati-
cum," pp. 91-99, and R. Wittern, "Zurn Krankheitserkennung in der knidischen
Schrift De internis affectionibus)'' pp. 109-19.
15. We consider a retrospective diagnosis adequate if it takes account of all the
symptoms mentioned, explains the chief ones, and contradicts none of them; it
should also accord with epidemiological conditions brought to light by medical
historians. But such a diagnosis is not necessarily the only one possible. Most
ancient clinical accounts are lacking from the viewpoint of modern medicine and
could wear several labels from modern pathology. Modern diagnosis proceeds from
the symptom to the fundamental lesion and its etiology, a procedure that intro-
duces a high risk of error. On this subject, see M. Burger, Klinische Fehldiagnosen)
2d ed. (Stuttgart, 1954), and J. A. Jaquez, ed., The Diagnostic Process(Ann Arbor,
1964). The margin of error in diagnosis in a modern, well-equipped hospital ranges
from 20 to 40 percent, as confirmed by autopsy. See V. Munck, "Autopsy Finding
and Clinical Diagnosis," Acta Med. Scand._,suppl., 266 (1955); P. Gallo and F.
Nardi, "Sul divenire della patologia umana," Becenti progressiin medicina 61 (1976):
413-49; and L. Goldman et al., "The Value of the Autopsy in Three Medical
Eras," New Eng. ]our. Med. 308 (1983):1000-1005.
16. As I have shown for several diseases of the spine: M. D. Grmek, "Die
Wirbelsaule im Zeitgeschehen," Med. Welt 25 (1974): 70-76. For a broad overview
of this important feature of the natural history of diseases, one can consult R.M.J.
Harper, Evolutionary Origins of Disease (Barnstaple, 1975).Those modern biologists
who offer such an explanation of anatomical and physiological imperfections and
the origin of some diseases consider it a product of Darwinism. However, it has
been used since antiquity in the context of creationism. In the third century B.C.
the Stoic philosopher Chrysippus posed the question ''whether the diseases of
man happen in accord with nature," and his answer was that "it was not part of
NOTES TO PAGES 10-14 359
nature's initial design to subject man to disease, since that was totally incompatible
with the author of, nature, who engendered all that is good. But since he was
creating and placing in the world many great things that were both useful and
fitting, some drawbacks arose that were inseparable from the creation process.
Such drawbacks were made by nature because of inevitable linkages." For instance,
says Chrysippus, "when nature was fashioning the body of men, fineness of struc-
ture and the thing's own advantage dictated that the head be formed of very thin
and minute bones. But as a result of thus heeding the advantage of the chief part,
there was an exterior drawback. The head was poorly protected and easily broken
by slight blows. Likewise diseases and pains came into the world along with health''
(Aul us Gellius, NoctesAtticae) VII, 1).
17. V. Capecchi, ''Quelques reflexions sur la prehistoire du pied bot congeni-
tal," Actesdu CongresAss. Europ. Paliopath. (Caen, 1980).
18. Art:, 62, and Genit:, 10. For ancient ideas on the etiology of this deformity,
see M. Michler, Die l(lumpfusslehre der Hippokratiker (Wiesbaden, 1963),pp. 44-52.
19. See especially Aer:, 14; De morbo sacro) 2; Itorrh:, II, 5; Aristotle notes that
"deformed children are born from deformed parents: the lame sire the lame, the
blind the blind," Hist. anim'.,)vu, 5856.
20. See S. Ghinopoulo, Piidiatrie in Hellas und &m (Jena, 1930). For the elimi-
nation of crippled children, see M. Moissides, "Le malthusianisme clans l' Anti-
quite grecque," Rev. d)Hist. de Droit 47 (1969): 177-97.
21. Such is the view of N. Rothschild, Ueber das Alter der Hiimophilie (Berlin,
1882).See M. Schachter, "L'histoire de l'hemophilie: Quelques precisions," Bull.
Soc. Franr;.d)Hist. Med. 27 (1933):101-5; H. Hafliger, Zur Geschichteder Hiimophilie)
unter besondererBerucksichtigungder Schweiz (Basel and Stuttgart, 1969); and G.I.C.
Ingram, "History of Haemophilia," four. Clin. Path. 29 (1976): 469-79.
22. Hippocrates, Hemorrhoids (Littre, VI, 436-44), text and translation by R.
Joly (Paris, 1978). See S. Glaser, "Hippocrates and Proctology," Proc. &y Acad.
Med. 62 (1969): 38off.
23. See E. D. Baumann, "De Diabete Antiquo," Janus 37 (1933):257-70; H.
Schadewaldt, Die GeschichtedesDiabetes (Frankfurt, 1971);and C. L. Gemmill, "The
Greek Concept of Diabetes," Bull. New York Acad. Med. 48 (1972):1033-36.
24. F. Merke, "Die Eiszeit als primordiale Ursache des endemischen Kropfes,"
Schweiz. Med. Wschr. 95 (1965):1183-92.
25. For example, see W. Doerr, K. Kohn, and H. H. Jansen, Gestaltwandel
klassischerl(rankheitsbilder (Berlin, 1957);M. Moser, ed., Diseasesof Medical Itqgress
(Springfield, 1964); and M. Howe, World Geqgraphyof Human Disease (New York,
1977).
26. C. Nicolle, Destin des 1naladiesinfectieuses(Paris, 1933). See also R. Dubos,
"The Evolution of Infectious Diseases in the Course of History," Canad. Med.
Assoc. ]our. 79 (1958):445-51, and A. Cockburn, The Evolution and Eradication of
Infectious Diseases(Baltimore, 1963).
27. See J. du Boulay, Por-traitofa GreekMountain Village (Oxford, 1974).
28. R. and E. Blum, Health and Healing in 'Rural Greece(Stanford, 1965).
29. Acut'.,)5, I (Littre, II, 226; Joly, 37).
30. This word is omitted in most editions, notably in Joly's. It is the reading of
the oldest manuscript of this text (Marcianusgr.) 269).
31. Acut'.,)5, 1-2 (Littre, II, 226), after the translation by R. Joly (Paris, 1972), pp.
37-38. The translation is the same as his except in the names of the diseases, where
I have preferred to transliterate rather than translate. Although his rendering of
pleur£tisas "pleurisy" is defensible, that of phren£tis as "encephalitis" I consider an
anachronism.
32. Plato, Alcibiades) II, 139e.
NOTES TO PAGES 14-19
33. Epid°JI and 111. See below, chapter 12.

34. Aer.J 3-4 (Littre, n, 14-22), from HippocratesJvol. 1, trans. W.H.S. Jones
(Cambridge and London, 1923;rept. 1948), pp. 75-78.
35. Hierocles and Philagrios, PhilogelosJno. 94. See the now standard edition of
A. Thierfelder (Munich, 1968).
36. M. D. Grmek, "Histoire des recherches sur les relations entre le genie et la
maladie," Rev. Hist. Sci. 15 (1962): 51-68. For 'bibliography, see J. B. Gilbert and
G. E. Mestler, Diseaseand Destiny: A Bibliographyof Medical Referencesto the Famous
(London, 1962).
37. To be sure, there are critical situations when the health of a military leader
or a political figure can have important historical consequences. As for the princes
and the great political figures of the past, one cannot be too careful in forming a
judgment on the bearing their diseases may have had on history. Usually the
literature on this subject is too anecdotal. In France, it has found a wordy spokes-
man in Dr. Auguste Cabanes, who tried to justify his method (L)histoireeclairiepar
la clinique) [Paris, 1921]). But it was impossible for him to go beyond the stage of
research into "curiosities" and the "medical indiscretions of history."
38. Emile Littre had defined historical pathology as "a still embryonic science
whose development will be one of the intellectual duties of the twentieth century."
39. The bibliography on epidemics and history is immense. There are more than
100 titles on the "Great Plague."
Chapter One: LiteraryReflectionsof PathologicalReality

1. In particular, E. Rosner, "Terminologische Hinweise auf die Herkunft der
friihen griechischen Medizin," in Medizingeschichtein unsererZeit (Artelt Festschrift)
(Stuttgart, 1971), pp. 1-22. See also R. Stromberg, GriechischeWortstudien: Unter-
suchungen zur Benennung von TierenJ PflanzenJ I&rperteilen und I(rankheiten (Gote-
borg, 1944); J. Pokorny, Indogermanisches etymologischesWorterbuch (Bern and
Munich, 1948-1969); H. Frisk, GriechischesetymologischesWorterbuch (Heidelberg,
1954-72); P. Chantraine, Dictionnaire itymologiquede la languegrecque: Histoire des
mots (Paris, 1968-80).
2. An important list of nomina morborum is given by Galen, Meth. med.J n, 2
(Kuhn, x, 81-85). For a more complete thesaurus of these terms, see J. E. Heben-
streit, Exegesisnominum Graecorumquae morbosdefiniunt (Leipzig, 1751).
3. To be sure, the origin of several terms in pathology is unknown. Rosner,
"Hinweise" (above, n. 1), pp. 14ff., mentions among doubtful cases hub6s 'hunch-
back' and dothiin (abscess on the body surface; perhaps a Mediterranean term
derived from viticulture, like Latin furunculus); according to him, the word mfl-
phosis 'loss of eyelashes' is non-Inda-European. Moreover, the terms anthrax and
eskhara have been thought to be of Egyptian origin, but even if the theory is
correct, originally neither term had anything to do with pathology (anthrax is first
used only in the sense "burning coal"and eskharain the sense "hearth, fireplace").
See P. V. Ernstedt, Egipetskiezaimstvovaniyav greceskomyazyke (Moscow and Len-
ingrad, 1953),pp. 21-24, 26ff.
4. Rosner,
D. ''H. 1nwe1se
. '' (ab ove, n. 1) , p. 14.
5. For the overall picture, see J. Chadwick, The Deciphermentof Linear B (Cam-
bridge, 1958),and S. Hiller and 0. Panagl, Die fruhgriec.hischenTexteaus mykenischer
Zeit (Darmstadt, 1976). Of fundamental importance is the work of M. Ventris and
J. Chadwick, Documents in Mycenaean Greek) 2d ed. (Cambridge, 1973); see also
L. R. Palmer, The Interpretation of Mycenaean Greek Texts (Oxford, 1963).
6. See W. H. Hein, "Die Bedeutung der Entzifferung des Linear B fiir die
Arzneimittelgeschichte," Phann. Zeitung 106 (1961):1145-48, and C.P.W. Warren,

'' Son1e Aspects of Medicine in the Greek Bronze Age,'' Medical History 14 (1970):
364-77.
7. Very probably the word i-ja-te) which occurs on PY Eq 146 in a nonmedical
context, represents the Mycenaean form of the Homeric term iater 'doctor.' See
Ventris and Chadwick, Mycenaean Greek (above, n. 5), p. 123,and C. Gallavotti and
A. Sacconi, Inscriptiones Pyliae ad Mycenaeam aetatem pertinentes (Rome, 1961),p. 66.
Palmer, Mycenaean Greek Texts (above, no. 5), p. 422, accepts the identification as
well.
8. Landau, Mykenisch-griechische Personennamen (Goteborg, 1958),esp. the chap-
ter "Korperliche Eigenschaften," pp. 190-93.
9. Tablet reference: KN Dx 6059. See ibid., p. 192.
10. Unless stated otherwise, the translations cited and the references given are
from Richmond Lattimore, trans., The Iliad of Homer (Chicago, 1951)and The
Odysseyof Homer (New York', 1965).
n. For an overall perspective on the Homeric epics, see F. Robert, Homere (Paris,
1950); F. Buffiere, Les mythes d)Homere et la penseegrecque (Paris, 1956);W. Schade-
waldt, Von Homers Welt und Werk) 3d ed. (Stuttgart, 1959);A.J.B. Wace and F. H.
Stubbings, A Companion to Homer (London, 1963); Ch. Mugler, Les origines de la
sciencegrecque chez Homere: L)homme et Funivers physique (Paris, 1963).
12. For medical themes in the Homeric poems, see especially J. F. Malgaigne,
Etudes sur !)anatomic et la physiologic d)Homere (Paris, 1842); Ch. Daremberg, La
medecine clans Homere (Paris, 1865);0. Korner, Die iirtzlichen I<enntnisse in Ilias und
Odyssee (Munich, 1929); A. Botto-Micca, Omero medico: Medici) ferite e medicina in
Omero (Viterbo, 1930); H. Buess, "Mediko-chirurgisches in Ilias und Odyssee,"
Dtsch. Med. Wschr. 81 (1956):1818-22;H. Sigerist, History of Medicine) vol. 2, Early
Greek) Hindu) and Persian Medicine (New York, 1961);K. Mitropoulos, "Homerou
Iatrika," Platon 14 (1962): 145-76; A. Albarracin Teul6n, Romero y la medicina
(Madrid, 1970); and A.H.M. Kerkhoff, "La medicine clans Homere: Une biblio-
graphie," Janus 62 (1975):43-49.
13. Korner, Die iirtzlichen I<enntnisse (above, n. 12), pp. 3ff. See also T. D. Sey-
mour, Life in the Homeric Age (New York, 1963), and M. I. Finley, The World of
Odysseus (New York, 1954; rev. ed. 1965).
14. In addition to the publications cited above, n. 12, see also 0. Braumiiller,
I(rankheit und Tod bei Homer (Berlin, 1879); W. H. Friedrich, Verwundung und Tod
in der Ilias (Gottingen, 1956); F. M. Pontani, Lamorte degli eroi (Florence, 1975);
and B. Moreux, "La nuit, l'ombre et la mort chez Homere," Phoenix 25 (1971):
95-105.
15. It is difficult, even impossible, to translate the Homeric terms psukhi) thu-
m6s, and phrenes correctly into a modern language. Much ink has been spilled in
quest of their meaning. From the viewpoint of the history of ideas, it is useful to
read what Daremberg, La medecine (above, n. 12), pp. 53-59, thought in the last
century and what B. Snell writes nowadays; see Die Entdeckung des Geistes: Studien
zur Entstehung des europaischen Denkens bei den Griechen (Hamburg, 1948; rev. ed.
Gottingen, 1975);B. Simon, Mind and Madness in Ancient Greece (London, 1978);
and especially M. McDonald, "Terms for Life in Homer," Trans. St. Coll. Phys.
Philadelphia (1982),pp. 26-58.
16. For example, see the description of the death of Patroclus (Iliad 16.805-63),
of Lykaon (21.49-85, 95-127), and of Hector (22.319-66).
17. Iliad 16.480-507.
18. Iliad 5.584-89.
19. Iliad 22.365-67.
20. Odyssey 11.218-22. This is our translation, which differs from the standard
French translation by V. Berard (Paris, 1955-56)and from that of Lattim?re in the

translation of thumos (Berard: "ame"; Lattimore: "spirit") and psukhe (Berard:
"ombre"). It should also be stressed that the word ines designates fibrous tissues,
chiefly tendons but also nerves and ligaments. Berard's translation of"fury of fire"
as "energie de la brulante flamme" is unacceptable to me for its anachronistic
connotations.
21. Iliad 9.408-9.
22. See Daremberg, La medecine (above, n. 12), pp. 57ff., and A. Nehring,
"Homer's Descriptions of Syncopes," CP 42 (1947): 106-21.
23. Iliad 5.696, 5.309, and 14.418.
24. Iliad 23.466; Odyssey24.345.
25. Odyssey10.557-60.
26. Odyssey3.276-83.
27. Cf. Odyssey17.251.
28. Iliad 24.612. For Artemis's other interventions, see Iliad 6.428 and 21.483, as
well as Odyssey20:61.
29. The term loimos)usually translated "plague," occurs in Iliad 1.61. The epi-
demiological content of this Homeric narrative is obscured by the mythologizing
of reality. See Heraclitus, QuaestionesHomericae)7; F. Buffiere, Les mythes d)Homere
et la penseegrecque(Paris, 1956), pp. 130 and 195-99; F. Berheim and A. A. Zenner,
"The Sminthian Apollo and the Epidemics among the Achaeans at Troy," Trans.
Amer. Philo!. Assoc. 108 (1978): 11-14.
30. Odyssey11.198-203. Again, our translation, since we differ from Lattimore
(and, for that matter, Berard) in the passage concerning the consumptive disease.
31. Iliad 6.211.
32. Odyssey4.611.
33. In the Iliad and the Odysseythere is some evidence about bodily needs, food,
and so on, but despite what is often alleged, no true preoccupation with hygiene.
See L. Moule, "L'hygiene clans les poemes homeriques," Bull. Soc. Frant;. Hist.
Med. 17 (1923): 350-77, and R. Lorenz, Bei-trtigezur Hygiene bei Homer (Munich,
1976).
34. Hesiod, Worksand Days) 11.695-705, in Hesiod)HomericHymns) and Homerica)
trans. H. G. Evelyn-White (Cambridge and London, 1914; rept. 1977). Eugenic
exhortations abound in Theognis, the sixth-century poet from Megara, who in-
vokes the wisdom of animal husbandry to justify his preference, in a wife, of race
(genos)over wealth: see especially 11.183-92. Not all the poetry ascribed to Theognis
is agreed to be as old as the sixth century, however.
35. Odyssey7.112-21, 8.245-49.
36. Odyssey15.405-11.
37. Herodotus, The Histories) trans. A. de Selincourt (Baltimore, 1954; rept.
1968), IX, 122, p. 599.
38. In particular, Airs) Waters) and Places. For an excellent analysis of the birth
of this medical-geographic notion, see W. D. Smith, "Physiology in the Homeric
Poems," TAPA 97 (1966): 547-56.
39. Iliad 1.247-48.
40. Iliad 2.216-20.
41. See below, chapter 2, n. 88.
42. Iliad 2.266.
43. Odyssey19.246-48; Odyssey2.15-16; Iliad 10.316; Iliad 1.594 and 18.394-405;
Odyssey8.307. Cf. Korner, Die arztlichen Iv:nntnisse (above, n. 12), p. 17.
44. Iliad 2.594-600. The blinding of the lyre-player Thamyris is figured on an
Attic hydria of the fifth century B.C. (Oxford, Ashmolean Museum no. 530). Greek
vase-painting also illustrates the blinding of other legendary personages: the kings
Phineas and Oedipus (see, for instance, British Museum E291, E302, and G105).
45. Odyssey10.492ff.
46. Odyssey8.63-64. Other blind singers are mentioned by Hesiod and Euripides.
47. Cf. Maximus of Tyre, Diss.) 38, 1, and Porphyry, Quaest. Hom. ad Od? ed.
H. Schrader (Leipzig, 1882),II, 72.
48. Hymn to Apollo) 172. Cf. Thucydides, III, 104.
49. Proclus, Vita Homeri) 16-18, apud A. Severyns, &cherchessur la Chrestomathie
de ltoclos (Paris, 1963),4:68. According to modern studies, ho~os is an institutional
term signifying any member of a group characterized by solidarity. Thence it came
to designate a hostage responsible for others, a guide for a blind person, and even,
perhaps, a singer interpreting an author's works. See L. Deroy, "Le nom
d'Homere," Antiquite Classique41 (1972):427-39.
50. According to [Plutarch,] De vita Homeri) 1, 2 (C. Muller, Frag. Hist. Graec.J
I, 227).
51. Plato, Phaedrus) 243a.
52. Hellenistic and Roman copies of the bust of Homer are kept in the archaeo-
logical museumsin Rome, Naples, Florence, Modena, and Bonn. The Louvre has
a Hellenistic terra cotta replica of it (D1549). See R. and E. Boehringer, Homer:
Bildnisse und Nachweise (Breslau, 1939), and G.M.A. Richter, Portraits of the Greeks
(London, 1965), pp. 45-56. On the iconography of Stesichorus, the old blind
hunchback, see R. Heidenreich, "Eine Dresdener Mantelstatue," Arch. Anz.
(1972), pp. 570-83.
53. Cicero, Tusculan Disputations) V, 39, n4.
54. Proclus, Vita Homeri) 47-49, apud Severyns, &cherches (above, n. 49), 4: 72.
55. Lucian, Veraehistoriae) II, 20.
56. Dio Chrysostom, Orat.J 36. ,,
57. See in particular Suda s. v. Homeros.
58. Pausanias, Descriptionof Greece)IV, 33, 7, and II, 33, 3.
59. Heraclides Ponticus, De rebuspublicis) 32, in C. Muller, Frag. Hist. Graec? II,
222.
60. [Herodotus], Vita Homeri) 7.
61. Among modern publications on this subject, there are an anonymous article
entitled "La cecite d'Homere," Chron. Med. 12 (1909): 680, and the studies of A.
Esser, "Augenkrankheiten beriihmter Personlichkeiten in der Antike," Med. Welt
14 (1940): 894, and Das Antlitz der Blindheit in der Antike) 2d ed. (Leiden, 1961),
p. 10.
62. For the identification of the Egyptian term nehat in Papyrus Ebers with tra-
choma, see B. Ebbell, Alt-tigyptische Bezeichnungen fur J(rankheiten und Symptome
(Oslo, 1938),p. 26. It is generally accepted that this form of granular conjunctivitis
was a scourge of Pharaonic Egypt. See G. Lefebvre, Essai sur la medecineegyptienne
de Pepoquepharaonique (Paris, 1956),pp. 74ff. The presence of trachoma in Greece is
attested from the first appearance of texts in Greek on eye disease. The Hippocratic
treatise Vision recommends some reasonable and relatively simple measures for the
treatment of granulations on the eyelids (De visu) 4; Littre, IX, 156). The most
recent editor of the text, Robert Joly, is surely correct to state, in his commentary
on this chapter, that it "undoubtedly deals with the granulations of trachoma"
and "is very judicious in its information and advice" (Hippocrates, Des lieux dans
Phomme [Paris, 1978], p. 170). This work probably dates from the beginning of the
fourth century B.c., and the disease in question ..Nas not considered a novelty in
Greece at that time. See also M. Meyerhof, The History of Trachoma Treatment in
Antiqui-ty and during the Arabic Middle Age (Cairo, 1936).
63. See the famous analysis in Snell, Die Entdeckung (above, n. 14), pp. 13-18.
64. H. J. von Schumann, "Phanomenologische und psychoanalytische Unter-
suchung der Homerischen Traume: Ein Beitrag zur Klarung der umstrittenen
Blindheit des Dichters," Acta Psychother. 3 (1955): 205-19.
65. According to the famous British politician William Ewart Gladstone, whose
argument was taken up and developed by the ophthalmologist Hugo Magnus and
the philologist Lazarus Geiger, the Homeric Greeks had not attained the perfect
color perception characteristic of modern man. This strange idea was the result of
an erroneous application of Darwinism and some ill-founded philological reason-
ing. See W. E. Gladstone, Studies on Homer and the Homeric Age (Oxford, 1858),
3:457-96. With much conviction but little success, W. Schultz defended the hy-
pothesis that the Greeks were color-blind, that is, hereditarily blind in the blue
and green portion of the spectrum (Das Farbenempfindungs-system der Hellenen
[Leipzig, 1904]). For the refutation, see Korner, Die iirtzlichen I<enntisse(above, n.
12), pp. 52ff., and especially the study by M. H. Marganne, "Le systeme chroma-
tique clans le corpus aristotelicien," Etudes classiques46 (1978): 185-203.
66. He listed the following localizations: dome of the skull 6, forehead 7, temple
3, auricular region 8, orbital region 1, nose 1, mouth 1, jaws 2, throat 6, nape of
neck 10, decapitation 2, clavicular region 4, sides of the chest 1, front of the chest
9, upper chest 1, mammary region 10, heart 1, hypochondrium at the level of the
diaphragm 1, middle of belly 5, flank and lower belly 10, umbilical region 2, groin
1, liver 4, back 9, rear shoulder 3, front shoulder 9, shoulder removal 1, arm removal
1, arm 2, forearm 5, wrist 2, buttock 3, hip 1, thigh 3, knee 2, back of the knee 1,
calf 1, ankle 1 (Daremberg, La medecine [above, n. 12], pp. 76-77).
67. Instead of Daremberg's 140 wounds whose placement is clearly specified,
Frolich counted 147.
68. H. Frolich, Die Militiirmedizin Homer)s (Stuttgart, 1879), pp. 56-60.
69. See especially Korner, Die iirtzlichen I<enntnisse (above, n. 12), pp. 86ff.;
Botto-Micca, Omero medico (above, n. 12), pp. 59-63; T. Mildner, ChirU1;gieund
Wundbehandlung vor Troja (Traunstein, 1962). For the most recent and most com-
plete list of wounds, classed topographically from head to toe, see A. Albarracin
Teul6n, "La cirugia homerica," Episteme 5 (1971): 83-97.
70. Iliad 5.297-317, 431-70, 512-18.
71. Iliad 16.738-42.
72. Iliad 5.584.
73. See Seymour, Life in the Homeric Age (above, n. 13), pp. 62off., and Albarracin
Teul6n, "La cirugia" (above, n. 69), pp. 95ff.
74. Iliad 16.345-50.
75. Iliad 17.293-98.
76. Iliad 16.4n-14 and 569-80.
77. Iliad II. 349-60.
78. Iliad 5.309-10.
79. Probably this is why Frolich, wrongly, did not include it in his table. If we
also include the case of Euryalos, discussed below, and Eumelos's accident during
a chariot race (Iliad 23.391-97), the Iliad contains three descriptions of head wounds
without fatal consequences, only one of which is due to warfare.
80. Daremberg (La medecine [above, n. 12], pp. 60-62) made this diagnosis long
ago.
Sr. Iliad 23.687-98.
82. See Daremberg, La medicine (above, n. 12), pp. 62-64; Botto-Micca, Omero
medico (above, n. 12), p. 60; and above all Albarracin Teul6n, "La cirugia" (above,
n. 69), pp. 89ff., and E. Prim, "Hals, Nase und Ohr in Ilias und Odysee," Zeitschr.
Laryng. Rhin. Oto!. 43 (1964): 330-43. In Frolich 's statistical table, decapitations and
some wounds in which the weapon reaches the skull by way of the neck are classed
as head wounds. This is why the number of neck wounds in Frolich is relatively
smaller than in statistics compiled by more recent historians.
83. Iliad 22.324-29. In connection with Hector's mortal wound, it is noteworthy
that previously, in single combat with Ajax, he had a narrow escape: of all the epic
heroes, he is the only certain survivor of a neck wound, a superficial cut that made
his black blood ooze up (Iliad 7.260-63).
84. Carotid hen1orrhage is surely the cause of sudden death when the weapon
reaches the neck "by jaw and ear" (Iliad 16.606ff.; cf. also 13.177-81and 671ff.).
Antilochus kills Thoon straightaway by cutting clean through "the entire blood
vessel I which runs all the way up the back till it reaches the neck" (Iliad 13.545-
49). Here the Homeric anatomy is caught in error, since there is no important
vessel running along the back of the trunk. This can only be a lesion of the carotid
(or, possibly, the brachio-cephalic trunk artery or the internal jugular vein), with
the "dorsal vessel" being in that case, clearly, the aorta. An imaginary dorsal vessel
is also mentioned in the Hippocratic corpus. Korner and, more recently, Kudlien
defend the empirical basis of this vascular lesion in Homer, against Friedrich who
speaks of "Scheinrealismus" in regard to it. See F. Kudlien, "Zurn Thema Homer
und die Medizin," Rh. Mus. 108 (1965):293-99.
85. Iliad 20.480-83. See also 14.49 and 11.145-47, 259-61.
86. A particularly well-aimed and forceful spearcast could skirt the shield and
pierce the bronze or iron mail, as in the case of the Trojan Alcathoos (Iliad 13.438-
44). Stabbed in the middle of his chest by a spear, "he cried out then, a great cry,
broken, the spear in him, / and fell, thunderously, and the spear in his heart was
stuck fast / but the heart was panting still and beating to shake the butt end / of
the spear." Although the penetration of the spear's point through the coat of mail
is plausible, the enormous power of the heart to transmit vibrations through the
heavy spear haft belongs to the poet's imagination.
87. Cf. Iliad 11.577-79;13.411ff.;17.347-49; 20.460-72.
88. See L. Schonbauer, "Die Chirurgie bei Homer," Neubur;gersFestschrift
(Vienna, 1948), pp. 436-39.
89. Daremberg, La midecine (above, n. 12), pp. 69ff.; Albarracin Teul6n, "La
cirugia," (above, n. 69), pp. 91ff.
90. Iliad 11.380-82.
91. Iliad 5.537-40; see also 5.615-17,16.463-65, and 17.516-24.
92. Iliad 13.567-75.
93. Iliad 4.527-31.
94. Iliad 16.81ff.
95. Iliad 5.65-68.
96. Such was the opinion, long ago, of Malgaigne, Etudes (above, n. 12), p. 17.
The Iliad contains the description of a wound with exactly the same trajectory for
a different weapon, an arrow (13.650-54). One of the descriptions could be an
imitation of the other.
97. See Daremberg, La midecine (above, n. 12), pp. 70-72, and Albarracin Teu-
16n, "La cirugia," (above, n. 69), pp. 86ff. and 92ff. For wounds of upper limbs,
see especially H. Lipschutz, "Hand, Arm, and Shoulder Trauma in the Iliad)"
Sur;gery51(1963):833-36.
98. Iliad 8.324-29. Our translation, which differs from that of Lattimore in the
interpretation of par) 3mon) kairi6n) and neur'in (1. 328). In book 22 Homer uses an
analogous, but not identical, expression to describe the place where Achilles
wounds Hector mortally, where the collarbone separates the shoulder from the
throat. The wound of Teucer is more lateral than Hector's, a fact that agrees
completely with the pathological results of both wounds.
99. Iliad 5.73-75.

100. Iliad n.267-68.
IOI. Iliad II. 396-98.
102. Already in A. Brendel, De Romero Medico (Wittenberg, 1700); then, for
example, the works of Frolich (above, n. 68) and A. Floquet, Homere medecinJ
thesis (Paris, 1912).
103. Korner, Die itrtzlichen ICenntnisse(above, n. 12), pp. 88-89.
104. Korner's arguments were refuted by E. Fuld, "Quelques remarques sur les
sciences naturelles et medicales clans Homere, '' Rev. etud. hom. 2 (1932):10-17.
105. See Frolich, Die Milititrmedizin (above, n. 68), pp. 61-68; Seymour, Life in
the Homeric Age (above, n. 13), pp. 623ff.; and especially 0. Schmiedeberg, Ueber
die Pharmaka in der Ilias und Odysee (Strasburg, 1918), as well as G. Majno, The
Healing Hand (Cambridge, Mass., 1975),pp. 141-52.
106. It is true that the Iliad speaks of the ulcerated sore on the leg of Philoctetes,
which arose as a chronic complication from snakebite, but the treatment of it is
described in a summary by Proclus of the Little Iliad) not in the great epics. In any
case, this wound is of a magical character.
107. For instance, Iliad 4.190, where Agamemnon tells Menelaus that a doctor
will palpate and medicate his cut, a superficial, belt-high wound of the trunk.
Strangely, this word never occurs in the Odyssey.
108. See chapter 4, p. 125.
109. The Greek word "trauma is not attested before Herodotus. It designated a
bodily wound but also a material loss.
no. I use this term in a restricted sense, of the permanent alteration of a function
in the absence of an ongoing pathological process.
ni. To be consulted in this connection area A. Semelaigne, Etudes historiquessur
Falienation mentale dans FAntiquite (Paris, 1869); R. Ganter, "Psychopathologisches
aus Homer," Munch. Med. Wschr. 72 (1925):1924-26; J. L. Heiberg, Geisteskrank-
heiten im klassischenAltertum (Berlin and Leipzig, _1927);J. Mattes, Der Wahnsinn
im griechischen Mythos und in der Dichtung (Heidelberg, 1970); and especially
Simon, Mind and Madness (above, n. 15). We leave aside a long list of writers from
S. Freud to 0. Rank and G. Devereux who took the ancient literary texts as points
of departure for psychoanalytical interpretations.
n2. Cf. B. Lincoln, "Homeric lussa: Wolfish Rage," IF 80-81 (1975-76): 98-105.
For the antiquity of rabies, see Aristotle, Hist. anim.J 604a, and a vase in Boston,
Museum of Fine Arts no. 00.346.
n3. Iliad 17.695. See also Odyssey4.704, where the stupefaction of Penelope is the
subject.
n4. Iliad 5.10, 241.
n5. See E. D. Baumann, "De asthmate antiquo," Janus 38 (1934): 139-62, and
H. Schadewaldt, Geschichte der Alle1lJieJvol. 2, Heufieber und Asthma bronchiale
(Munich, 1980), pp. 188-94.
n6. Iliad 8.181-83and 9.243.
n7. Odyssey14.480-502 and 17.22-25.
n8. Odyssey5.453-57 (slightly modified from Lattimore).
n9. Odyssey9.410-12. According to one recent commentator, the passage alludes
to cholera (Seymour, Life in the Homeric Age [above, n. 13], p. 626). There is no
real basis for this diagnosis.
120. Celsus, De medicinaJpraefatio, 3. For the distinction in archaic Greek thought
between external wounds amenable to medical treatment and internal diseases of
divine origin, see F. Kudlien, Der Beginn des medizinischen Denkens bei den Griechen
(Zurich and Stuttgart, 1967), pp. 89-91.
I2I. Iliad LIO.
122. See P. Lain Entralgo, "Die Krankheit in den Dichtungen Homers," in

Heilkunde in geschichtlicherEntscheidung (Salzburg, 1950), pp. 35-47; Sigerist, History
of Medicine (above, n. 12), 2: 20-24; and G. Freiser, Allgemeine I(rankheitsbe-
zeichnungen im CorpusHippocraticum (Berlin and New York, 1976), pp. 89-91.
123. See H. Oldelehr, Seekrankheitin der Antike (Diisseldorf, 1977), pp. 14ff.
124. Odyssey14.252-56.
125. Thesmophoriazusae)882.
126. Semonides, Women) 54. See Oldelehr, Seekrankheit (above, n. 123), pp. 1-3.
The modern word "nausea" is derived, by the intermediation of classical Latin,
from the word nautfa (root naus 'ship'). ·
127. Daremberg, La medecine(above, n. 12), p. 86.
128. Odyssey5.394-97.
129. Iliad 13.666-70.
130. Odysseyn.201.
131. Hesychius identifies the Homeric teked1Jnwith phthfsis in later authors, and
Suda as well as Pollux puts the two terms in the same class. See B. Meinecke,
"Consumption (Tuberculosis) in Classical Antiquity," Ann. Med. Hist. 9 (1927):
379-402.
132. Sophocles, Antigone) 819.
133. Euripides, Alcestis) 204 and 236.
134. Iliad 22.31.
135. Brendel, De Homero Medico (above, n. 102).
136. Daremberg, La medicine (above, n. 12), pp. 91ff.
137. See W.H.S. Jones, Malaria: A NeglectedFactor in the History of Greeceand
&me (Cambridge, 1907), pp. 23ff.
138. F. Robert, Homere (Paris, 1950), p. 208.
139. Hesiod, Works and Days) 496-99, trans. H. G. Evelyn-White (Cambridge,
1914; rept. 1977). See also the critical edition and commentary on this text by
M. L. West, Hesiod)sWorksand Days (Oxford, 1978).
140. See W. Schiller, "Das Hungerodem bei Hesiod," Janus 25 (1921):37-44.
141. Worksand Days 240-45, 533-36.
142. Plutarch, Banquet of the SevenSages)eh. 14.
143. With the sole exception of certain words occurring in the Pseudo-Hesiodic
fragments on the Proetides.
144. Jones, Malaria (above, n. 137), p. 25.
145. Hesiod, Worksand Days) 100-104; see also West ad foe.
146. It is a process that apparently begins in the framework of a magico-religious
conception of the world and is enabled by the integration of magical causality into
the natural order of things. That much was well shown by H. Frankel, Dichtung
und Philosophiedesfruhen Griechentums) 2d ed. (Munich, 1962), p. 130, and devel-
oped by G. Freiser, I(rankheitsbezeichnungen (above, n. 122), pp. 91ff. I cannot
subscribe to the opinion of U. von Wilamowitz-Moellendorf, who thought that
Hesiod "rejects all attempts to reduce disease to natural causes" in the lines cited
(Ewa [Berlin, 1928], p. 53), or of G. Lanata, according to whom Hesiod's diseases
are "silent figures utterly immune to natural causality" (Medicina magicae religiosa
popolarein Greciafino al/Yetadi Ippocrate[Rome, 1967], p. 30). The distinction be-
tween natural and magical causality should not be retrojected by our analysis into
a system of thought in which precisely the confusion between the two was of
fundamental importance.
147. Alcmaeon, fr. 4 (Diels-Kranz); text transmitted by Aetios Doxographos,
Placita) V, 30, r. See H. Diels and W. Kranz, Fragmenteder Vorsokratiker)6th ed.
(Berlin, 1951),1:215ff.This central passage is actually a reconstn1ction by Diels from
several variants. The difference between the short and the long versions is far from
NOTESTOPAGES40-44
negligible, since in the former monarkhfa signifies the domination of a single qual-
ity, while in the latter it signifies the domination of one quality over its opposite.
Contrary to accepted opinion, I believe that isonomfa signifies equilibrium among
all the qualities and not equality within each pair of opposed qualities.
148. In this context, the word hier6s should not be understood in the sense of a
link with the cult of a divinity but in that of participation in supranatural and
mysterious phenomena.
149. See Preiser, I(rankheitsbezeichnungen (above, n. 122), pp. 47, 51, 82-85, and
87-89.
150. De morbo sacro) I, 1-3 and 10 (Littre, VI, 352-53).This translation differs from
those ofLittre, Daremberg, and Joly. It is based on the critical edition of the Greek
text by H. Grensemann, Die hippokratische Schrift ((Ueber die heilige I(rankheit))
(Berlin, 1968), p. 60.
151. See Lana ta, Medicina magica (above, n. 146), and above all the masterly
monograph of 0. Temkin, The Falling Sickness) 2d ed. (Baltimore, 1971).
152. De morbo sacro) II, 4 (Littre, VI, 364; Grensemann, 68).
153. The Hippocratic writer militating against this name hasn't yet at his disposal
another technical term for the disease. See Grensemann, Die hippokratische Schrift
(above, n. 150),..pp. 5ff.
154. Herodotus, III, 33, trans. A. de Selincourt (Baltimore, 1954; rept. 1968). Cf.
T. S. Brown, "Herodotus' Portrait ofCambyses," Historia 31(1982):387-403.
155. Heraclitus, fr. 46 D-K (Diels and Kranz, Vorsokratiker [above, n. 147], 1: 161).
156. See M. Marcovich, Heraclitus (Merida, 1967), p. 575.
157. Diogenes Laertius, Vitae philosophorum) IX, 3, after the translation by R.
Genaille (Paris, 1965), 2: 163.
158. Ibid., IX, 5.
159. I cite him from Archilochus, Fragments) ed. F. Lasserre and A. Bonnard
(Paris, 1958)(L-B). I have also used E. Diehl, Anthologia Lyrica Graeca) fasc. III, 3d.
ed. rev. E. Beutler (Leipzig, 1952)(D); M. Treu, Archilochos) griechisch und deutsch
(Munich, 1959); and J. Tarditi, Archilochus (Rome, 1968). See also the article by
J. N. Dambassis, Epet. Etair. I(yklad. Melet. (1968), pp. 672-81.
160. Archilochus, fr. 93 L-B (60 D).
161. Archilochus, fr. 236 L-B (from Philoxenos, Anecd. Oxyr., I, 164).
162. C. Daremberg, Etat de la medecine entre Homere et Hippocrate (Paris, 1869),
p. 5.
163. See F. Lasserre, Les epodesd)Archiloque (Paris, 1950).
164. Schol. Theocr., II, 48.
165. Pap. Oxy., 2313,frs. 13and 14.
166. Archilochus, fr. 27 L-B (42 D).
167. A. Bonnard in Archilochus, Fragments (Paris, 1958),pp. 9ff.
168. The poet says (fr. 105L-B) that he lived off "those figs" and food "gotten
from the sea."
169. Daremberg, Etat de la medicine (above, n. 162), pp. 5ff.
170. Archilochus, fr. 198L-B.
171. Archilochus, fr. 200 L-B.
172. Archilochus, fr. 85 L-B, and see Jones, Malaria (above, n. 137),p. 26.
173. Archilochus, frs. 245, 249, and 266 L-B.
174. T. Reinach and A. Puech, Alcee)· Sapho (Paris, 1938). See especially the fa-
mous poem of Sappho (fr. 2) that was translated and commented on by Boileau in
his Traite du sublime.
175. See, for example, Anacreon, fr. 428 Page; Theocritus, 30, 30; Prodikos, fr.
84 B 7 D-K. Among the medical writers who thought of love as a disease, the most
significant are Soranus, Galen, and Oribasius. For secondary sources on this sub-
ject, see H. Crohns, "Zur,Geschichte der Liebe als Krankheit," Arch. I(ult. Gesch.
3 (1905): 66-86; M. M. Mesulam, "The Diagnosis of Love-sickness: Experimental
Psychopathology without a Polygraph," Psychophysiology9 (1972): 546-51; M.
Clavolella, Malattia d)amoredalFAntichita al Medioevo (Rome, 1976); and A. Giedke,
Die Liebeskrankheit in der Geschichteder Medizin) diss. (Diisseldorf, 1983).
176. Even in Homer, wine "clouds one's reason." For the signs of drunkenness,
see Odyssey14.466-69 and 21.293-98. The effect of wine on the soul is expounded
on at great length and with remarkable literary erudition by Galen in his treatise
Quod animi mores. See J. 0. Leibowitz, "Acute Alcoholism in Ancient Greek and
Roman Medicine," Brit. four. Addiction 62 (1967): 83-86, and G. Preiser, "Wein im
Urteil der griechischen Antike," in G. Volger, ed., Rausch und Realittit (Cologne,
1981),pp. 296-303.
177. Theognis, Poemeseligiaques)ed. and trans. J. Carriere (Paris, 1975),II. 2IIff.
178. Bulletin epigraphique)no. 385 (1978). See C. Meillier, "Un cas medical clans
une inscription funeraire," Zschr. Pap. Epig. 38 (1980): 98.
179. Diogenes Laertius, r, 81. See Reinach and Puech, Akee)· Sapho (above, n.
174), 190.
180. 0. Masson, Les fragments du poete Hipponax (Paris, 1962). For example, see
frs. 32 and 33.
181. Hipponax, fr. 36 Masson (29 D). Later, Aristophanes exploits the theme of
the blindness of the god of wealth.
182. Hipponax, frs. 10, 34, 59, 12, and 26 Masson (II, 25, 55, 15, and 39 D).
183. Hipponax, fr. 118Masson (X + XII D). The text is in Pap. Oxyr. 2176. See
Masson, Hipponax (above, n. 180), pp. 84-86 and 162-66.
184. Daremberg, Etat de la medecine (above, n. 162), p. 6.
185. Pap. Oxyr. 2174, fr. 4.
186. Hipponax, fr. 73 Masson (1 + 71 D).
187. Masson, Hipponax (above, n. 180), p. 143.
188. "A slave's head is never straight, it is always bent and sits on a slanting
nape. Rose or hyacinth never sprang from squill" (Theognis, Poemeseligiaques)r,
11.535-37).
189. Theognis, Poemes (above, n. 177), II. 173ff.The term epialosdesignates fever
in general or, according to Hesychius, in particular the shudder that precedes its
onset. It is also the name of an evil spirit that causes nightmares. According to
Jones, Malaria (above, n. 137), pp. 27ff., the word is associated with attacks of
malaria.
190. Theognis, Poemes (above, n. 177), r, 11.II22ff.
191. E. Benveniste, "La doctrine medicale des Indo-Europeens," RHR 130(1945):
5-12.
192. Pindar, Pythian) 3, 47-53, from The Odes of Pindar) trans. R. Lattimore
(Chicago, 1947; rept. 1959), pp. 53ff.
193. See. D. Brandenburg, Medizinisches bei Herodot (Berlin, 1973); A. Corlieu,
Etude medicatesur la retraite des Dix-Mille (Paris, 1878);and H. Cosson, Etude sur les
allusions medicateschez Xenophon) thesis (Paris, 1966).
Chapter Two: Paleopathology

1. See M. A. Ruffer, Studies in the Palaeopathologyof Egypt) ed. R. L. Moodie
(Chicago, 1921).This Anglo-French doctor gave a new meaning to the neologism
proposed in 1892 by R. W. Shufeldt to denote the study of pathological clues in
animal fossil remains (cf. "Notes on Palaeopathology," Popular ScienceMonthly 42
(1892); 679-84). A pioneer in paleopathological research on Egyptian bones and
mummies, Ruffer was professor of bacteriology in Cairo and a world-famous epi-

demiologist. He stayed in Greece during World War I, where he participated in
the organization of the health services of the Allied armies. He was interested in
Greek paleopathology but published nothing on the subject because his life was
abruptly cut short: returning by ship from Salonika, he died on the open sea, the
victim of a torpedo attack. For the story of his life, see W. E. Swinton, "Sir Marc
Armand Ruffer, One of the First Palaeopathologists,'' Canadian Medical Association
Journal 124 (1981):1388-92.
2. J. F. Esper, Ausfuhrliche Nachricht von neuentdeckten Zoolithen unbekannter
vieifussiger Thiere (Nuremberg, 1774). For the beginnings of paleopathology, see
H. E. Sigerist, History of Medicine) vol. 1, Itimitzve and Archaic Aiedicine (New York,
1951),pp. 38-42.
3. R. L. Moodie, Palaeopathology:An Introduction to the Study of Ancient Evidences
of Disease (Urbana, 1923);L. Pales, Paleopathologieet pathologie comparative (Paris, 1930).
This work, a doctoral thesis in medicine, is of exceptional scope and significance.
4. Pales Paleopathologie) p. 4. Indeed, Pales uses paleopathological evidence to
draw conclusions about the etiology and pathogenesis of certain diseases (notably
tooth decay, spondylitis, temporo-maxillary osteoarthritis, and cranial
osteoporosis).
5. See the update and bibliography published by Sigerist, History of Medicine
(above, n. 2), pp. 45-101 and 532-39. For European paleopathology, see especially
G. Wilkie, Die Heilkunst in der europaischen Thrzeit (Leipzig, 1936).
6. E. H. Ackerknecht, "Palaeopathology," in A. L. Kroeber, Anthropology
Today(Chicago, 1953),p. 120.
7. For an overview, see M. D. Grmek, "Metodi nuovi nello studio delle mal-
attie antiche,'' in Annuario della Enciclopedia della Scienza (Scienza e Tecnica 1975)
(Milan, 1975), pp. 71-84. See also L. von Karolyi, "Palaeopathologie (Aufgaben,
Objekte, Methodik, gegenwartiger Stand)," Sudhoffs) Archiv 54 (1970): 398-422.
Among recent handbooks, worth recommending are D. Brothwell and A. T.
Sandison, eds., Diseases in Antiquity (Springfield, 1967); R. T. Steinbock,
Paleopathological Diagnosis and Interpretation (Springfield, 1976); and D. J. Ortner
and W. G. Putschar, Identification of Pathological Conditions in Human Skeletal Re-
mains (Washington, D.C., 1981).
8. H. A. Harris, Bone Growth in Health and Disease (London, 1933),and C. Wells,
"Les lignes de Harris et les maladies anciennes," Scalpel 177 (1964): 665-71.
9. See P. Stastny, "HL-A Antigens in Mummified pre-Columbian Tissues,"
Science 183(1974): 864. For HLA associations with disease, see J. Dausset and A.
Svejgaard, HLA and Disease (Copenhagen, 1980), and D. C. James, "HLA-B 27 in
Clinical Medicine: Historical Reflection on the Discovery of the Disease Associa-
tion," Brit. ]our. of Rheumatology 22, no. 4, suppl. 2 (1983):20-24.
10. See A. Savicki et al., "Presence of Salmonella Antigens in Feces from a
Peruvian Mummy," Bulletin of the New York Academy of Medicine 52 (1976): 805-13.
11. See J. Jansen and H. J. Over, "Observation on Helminth Infections in a
Roman Army Camp," Itoc. First Intern. Congress Parasit. (Rome, 1966), 2: 791;
A. W. Pike, "Recovery of Helminth Eggs from Archaeological Excavations,"
Nature (London) 219 (1968): 303ff.; P. J. Wilke and H.J. Hall, Analysis of Ancient
Feces:A Discussion and Annotated Bibliography (Berkeley, 1975).
12. See A. MacKie et al., "Lead Concentration in Bones from Roman York,"
]our. Archaeol. Soc. 2 (1975):235-37; J. E. Ericson et al., "Skeletal Concentration of
Lead in Ancient Peruvians," New England Journal of Medicine 300 (1979): 946-51.
13. See J. L. Angel, "The Bases of Paleodemography" American Journal of Phys-
ical Anthropology 30 (1969): 427-37, and "Paleodemography and Evolution," ibid.
31(1969): 343-54·
14. V. M~ller-Christensen,' 'Osteo-archaeology as a Medico-historical Auxiliary
Science," Med. History 17 (1973):411-18. I prefer the term "osteoarchaeology" to
"paleo-osteology," which has also been proposed for this field.
15. H. Schliemann, Mycenes) trans. J. Girardin (Paris, 1879), p. 68.
16. See E. H. Acknerknecht, Rudolf Virchow (Stuttgart, 1957),pp. 187-91.
17. See H. Schliemann, Ilios) ville et pays des Troyens(Paris, 1885).The book's
preface is by Virchow. In several places, it testifies to the active participation of
Virchow in Schliemann's work and to the friendship that united them. An appen-
dix (pp. 964-70) contains a medical report by Virchow, "Pratique medicale en
Troade (1879).'' The discussion revolves around malaria, tuberculosis, and injured
workers. Two of Virchow's observations deserve comparison with clinical descrip-
tions in the Hippocratic corpus: malarial spleen enlargement in children and ge-
ophagy in an anemic child. For Virchow's Mediterranean trip, see also his own
accounts in Zschr.fur Ethno/,ogie11(1879):179ff., 204-17, 254-81.
18. R.Virchow, Alttrojanische Graberund Schadel(Berlin, 1882).
19. R. Virchow, "Ueber alte Schadel von Assos und Cypern," Abhandl. I(.
lteuss. Akad. Wiss. zu Berlin: Phys.-math. Cl. 24 (1884),II, pp. 1-55.
20. R. Virchow, "Schliemann's letzte Ausgrabung," Sitz.-Ber. I(. lteuss. Akad.
Wiss. zu Berlin (1891),pp. 819-28; R. Virchow, "Ueber griechische Schadel aus alter
und neuer Zeit, und iiber einen Schadel von Menidi der fiir den des Sophokles
gehalten wird," Sitz.-Ber. I(. lteuss. Akad. Wiss. zu Berlin (1893),pp. 677-700.
21. C. Stephanos, "Grece: Geographie medicale," in Dictionnaire encycl,opedique
dessciencesmedicates)ed. A. Dechambre (Paris, 1884), 4th series, 10: 363-581.
22. Ibid., pp. 432-33.
23. Ibid., pp. 433ff. In modern skulls cranial capacity varies greatly, with the
average at about 1500 ( +250) cc. It should be stressed that despite numerous
dedicated attempts, no correlation has ever been established between the volume
or weight of the brain (anatomical data that are easily measured and tied to cranial
capacity) and its intellectual potential or individual intelligence or the level of
cultural development (themselves theoretical values that are difficult if not impos-
sible to determine and quantify). The brain pan of fossil Homo sapiensis essentially
unchanged in modern man. If Kant had an enormous cranial capacity (1740 cc.), it
is also true that Leibniz 's was not more than 1422 cc., which is at the average
Virchow established for male skulls in Greece. According to recent anthropological
measurements, the cranial capacity of the ancient Greeks is about the same as that
of other European peoples. G. de Morsier (Essaisur lagenesede la civilisationscienti-
fique actuelle [Geneva, 1965]) has tried hard to explain the cultural progress of
humanity by brain mutation. According to him, in the course of 30,000 years,
Homo sapienshas undergone "two evolutionary cerebral mutations," the first tak-
ing place "around 6,000 years ago in a well-defined region of the Middle East"
(the "crescent of Neolithic civilization") and the second "500 years ago at the end
of the 15th and the beginning of the 16th century in a well-defined territory of
central and western Europe" (this is the "circle of scientific mutation"). G. de
Morsier has had the incredible effrontery to devise a "biological" history of the
progress of human thought in which the "Greek miracle" is erased and short-
circuited by a pseudo-scientific trick.
24. I. G. Koumaris, "Sur quelques variations des os des cranes grecs anciens,"
Anthropol,ogie29 (1918):29-36.
25. H. E. Sigerist, History of Medicine) vol. 1 (above, n. 2).
26. C. M. Furst, Zur Anthropo/,ogieder priihistcrichen Griechen in A1lJolis(Lund,
1930) (Lunds Univ. Arsskrift) n.s., sec. 2, 26, no. 8 [1930]), and Zur ICenntnisder
4nthropol,ogiederpriihistorischenBevolkerungder Insel Cypern (Lund, 1933)(Lunds Univ.
Arsskrift) n.s., sec. 2, 29, no. 6 [1933]).
27. E. Breitinger, "Die Skelette aus den submykenischen Grabern," in W.

Kraiter and K. Kubler, ICerameikosJ vol. 1, Die Nekropolendes 12. bis ro. Jahrhunderts
(Berlin, 1939), pp. 223-61.
28. See especially J. L. Angel, "Skeletal Changes in Ancient Greece," American
Journal of Physical Anthropowgy, n.s., 4 (1946): 69-97, and "Some Problems in
Interpretation of Greek Skeletal Material: Disease, Posture, and Microevolution,"
ibid., n.s., 12 (1954):284ff.
29. Cf. J. L. Angel, "A Racial Analysis of the Ancient Greeks," American Jour-
nal ofPhysicalAnthropowgy, n.s., 2, no. 4 (1944). See also his work on the popula-
tion of Lerna (n. 31, below), p.6.
30. Angel, "Skeletal Changes" (above, no. 28), p. 69.
31. The principal publications of John L. Angel containing osteoarchaeological
analysis, listed by site, are "Classical Olynthians," in D. M. Robinson, Necrolyn-
thia: Excavationsat Olynthus (Baltimore, 1942), pt. 11,appendix, pp. 211-40; "Skel-
etal Material from Attica," Hesperia 14 (1945): 279-363; "Troy: The Human
Remains,"in C. W. Blegen, Troy (Princeton, 1951);"The Human Remains from
Khirokitia," in P. Dikaios, J(hirokitia (London, 1953),pp. 416-30; "Human Bio-
logical Changes in Ancient Greece, with Special Reference to Lerna," Yearbook
Amer. Phil. Soc. (1957),pp. 266-70; "Appendix on the Early Helladic Skulls from
Aghios Kosmas," in G. E. Mylonas, Aghios J(osmas (Princeton, 1959), p. 169-79;
''Human Skeletal Remains at Kara ta~,'' American Journal of Archaeowgy70 (1966):
255-57, and 72 (1968):260-64, and 74 (1970): 253-59 (henceforth Angel, "Karata~"),
"Human Skeletal Material from Franchthi Cave," Hesperia 38 (1969): 380-81; The
Peopleof Lerna: Analysis of a ltehistoric Aegean Population (Princeton and Washing-
ton, D.C., 1971);"Early Neolithic Skeletons from Catal Hiiyiik: Demography and
Pathology," Anatolian Studies 21 (1971):77-98; "Late Bronze Age Cypriotes from
Bamboula," in J. L. Benson, Bamboula (Philadelphia, 1972); "Human Skeletons
from Grave Circles at Mycenae," in G. E. Mylonas, 0 TafikosI(ykws B tau Mykenou
(Athens, 1973), pp. 379-97; "Early Neolithic People of Nea Nikomedea," in E.
Schwidetzky, Die Anflinge des Neolithikums vom Orient bis Nordeuropa (Cologne,
1973), pp. 103-12; "Early Bronze Age Karata~, People and Their Cemeteries,"
American Journal ofArchaeowgySo (1976): 385-91 (henceforth Angel, "Karata~ Peo-
ple"); "Ancient Skeletons from Asine," in S. Dietz, Asine uJ ExcavationsEast of
the Acropolis1970-1974 (Copenhagen, 1982), fasc. 1, pp. 105-38. This scholar's bibli-
ography was published in the American Journal of Physical Anthropowgy 51 (1979):
509-16.
32. R. P. Charles, "Etude anthropologique des necropoles d' Argos, contribu-
tion a l'etude des populations de la Grece antique," Bull. Corresp.Helfin. 82 (1958):
258-313(henceforth Charles, "Argos r"); Etude anthropowgiquedes necropoles d)A1JJOSJ
contributiona Petudedespopulationsde la Greceantique (Paris, 1963)(henceforth Charles,
A1JJOSn); Anthropologie archeologique de la Crete (Paris, 1965); J. Dastugue,
"Grotte de Kitsos (Lavrion). rv. Les ossements humains pathologiques," Bull.
Corresp. Helfin. 98 (1974): 749-54; J. Dastugue and H. Duday, "Les ossements
humains pathologiques,'' in N. Lambert, ed., La grotte prihistoriquede J(itsos(Paris,
1981). The other publications will be cited as they come up in the subsequent
discussion.
33. J. L. Angel, "Patterns of Fractures from Neolithic to Modern Times,"
Anthrop. J(ozlemenyek(Budapest) 18 (1974): 9-18.
34. See T. W. Jacobsen, "17,000 Years of Greek Prehistory," ScientificAmerican
234 (1976): 76-87, esp. pp. 81ff., and Angel, "Franchthi Cave" (above, n. 31), pp.
38off. In the Petralona Cave in Chalcidice the skull of a fossil hon1inid was discov-
ered that is at least 200,000 years old. It is well preserved, thanks to a thin layer of
stalagmite around it, and it hasn't a trace of disease or wounds. It does not concern
us at the moment, though, since the specimen in question is not Homo sapiens.
35. Angel, "Karata§, People" (above, n. 31), p. 385.
36. Angel, "Mycenae" (above, n. 31), pp. 380-82.
37. Angel, Peopleof Lema (above, n. 31), pp. 42 and 91, pl. 2.
38. Ibid., pp. 58ff. and pls. 13and 22.
39. See the whole treatise De capitisvulneribus (Littre, III, 182-260) and in partic-
ular chs. 8 and 16. It is instructive to consult the commentaries ofJ. E. Petrequin,
Chiru:1;gied)Hippocrate (Paris, 1877), 1: 412-558, and the modern medico-historical
analyses of E. Iversen, "Wounds in the Head in Egyptian and Hippocratic Medi-
cine," in the festschrift volume Studia orientaliaJohannis Pedersendicata (Copenha-
gen, 1953),pp. 163-71,and of E. S. Gurdjian, Head Injuryfrom Antiqui-ty to the ltesent
(Springfield, 1973).
40. Charles, A1;gosII ~above, n. 32), pp. 35and 66, pl. 10, fig. 4.
41. Angel, "Skeletal Material from Attica" (above, n. 31), pp. 301and 303, fig. 6.
42. Virchow, "Assos und Cypern" (above, n. 19), pp. 25, 26.
43. See the works of Angel on Lerna, <.;atal Huyuk, and Karata§. On the skull
from Asine, see Furst, A1;golis)(above, n.-26), p. 123, pl. 8, and Angel, "Ancient
Skeletons from Asine" (above, n. 31), p. 109. P. M. Fischer noted a fracture of the
lower jaw that healed remarkably well on a Bronze Age Cypriot skull from Trypes
near Dromolaxia (OpusculaAtheniensia 13[1980]: 140).
44. Angel, People of Lerna) p. 61, and "Human Biological Changes," p. 269
(both above, n. 31).
45. Angel, "<_;atal Huyuk," p. 91 and pl. 4, and "Ancient Skeletons from
Asine," p. 109, fig. 15, (both above, n. 31).
46. See A.J.B. Wace, Mycenae: An ArchaeowgicalHistory and Guide (Princeton,
1949), p. 117,and Angel, "Mycenae" (above, n. 31), p. 381.
47. Fract., 1 (Littre, III, 414), trans. E.T. Withington, Hippocrates(London and
Cambridge, 1928;rept. 1948), 3: 95.
48. For classical knowledge of the treatment of fractures, see E. J. Gurlt, Ge-
schichteder Chiru1;gieund ihrer Ausubung (Berlin, 1898);A. Benedetti, Traumatowgia
al tempo di Ippocrate (Rome, 1969); A. Roselli, La chiru1;giaippocratica(Florence,
1975).
49. Angel, "<_;atalHuyuk (above, n. 31), pp. 91, 94; pl. 4.
50. Angel, Peopleof Lema (above, n. 31), pp. 44, 91; pl. 26.
51. Breitinger, "Die Skelette" (above, n. 27), p. 236.
52. Angel, Peopleof Lema (above, n. 31), p. 42. In Corinth, there was a local hero
in the protogeometric period (ninth century B.c.), a man of about 45, tall, broad-
shouldered, with a fractured right hand, bad teeth, and marked degenerative joint
disease (E. V ermeule, Aspectsof Death in Early GreekArt and Poetry[Berkeley, 1977],
pp. 206ff.).
53. Angel, "<_;atalHuyiik" (above, n. 31), pp. 91, 94; pl. 4.
54. Furst, A1;golis(above, n. 26), p. 123,fig. 52.
55. E. Vermeule, Greecein the-Bronze Age (Chicago and London, 1954), p. 9,
pl. I.
56. Angel, "Mycenae" (above, n. 31), p. 384.
57. Angel, Peopleof Lema (above, n. 31), p. 44.
58. Angel, "<_;atalHuyiik" (above, n. 31), p. 94 and pl. 4.
59. Angel, Peopleof Lerna (above, n. 31), pp. 50, 92; pl. 26.
60. Ibid., p. 58. I will take up the paleopathology of purulent inflammations in
chapter 4, which is devoted to the historical reality concerning pyogenic microbes
and the sequelae of nonspecific infections.
6r. See especially E. Guiard, La -irepanation cranienne chez les Neolithiques et chez
les Rimitifs modernes (Paris, 1930); Sigerist, History of Medicine (above, n. 2), 1: 101-
13;J. Dastugue, "Le diagnostic des trepanations empiriques du crane," in Travaux
et documents du Gen-irede palioanthropologie et de paliopathologie (Lyon, 1975),t. 2, vol.
2, pp. 71-79.
62. Angel, "~atal Hiiyiik" (above, n. 31), p. 9r.
63. A. D. Tsouros, "Ereuna se proi:storika pathologika osta: t6 kranfo ton Ark-
hamon," Anthropos (Athens) 1, no. 1 (1974): 55-60.
64. Y. Sakellarakis and E. Sapouna-Sakellaraki, "Drama of Death in a Minoan
Temple," National Geographic Magazine 159(1981):205-22.
65. Charles, A13os II (above, n. 32), pp. 67-69.
66. De capitisvulneribus (Littre, III, (182-260), especially chs. n, 30, and 3r. See J.
E. Petrequin, "Recherches historiques sur !'operation du trepan chez les anciens
et en particular sur la trepanation clans la contusion du crane d'apres Hippocrate,"
Bull. Soc. Chir. de Paris (1867),pp. 155ff.
67. Angel, "Karata~," p. 256, and "Karata~ People," p. 385(both above, n. 31).
68. Angel, People of Lema (above, n. 31), pp. 43ff. and pl. 5.
69. Angel, "Mycenae" (above, n. 31), pp. 380, 391;pl. 248.
70. Angel, "Ancient Skeletons from Asine" (above, n. 1), p. 109.
7r. Charles, "Argos r" (above, n. 32), pp. 31off.
72. L. Munter, Das Grab des Sophokles (Athens, 1893).
73. The spokesman for Miinter's opponents was P. Wolters. For a resume of the
first polemics, see R. Virchow, "Ueber den vermeintlichen Sophokles-Schadel und
iiber die Grenze zwischen Anthropologie und Archaologie," Zschr. fur Ethnol.
(Verh. Berl. Ges. Anthrop.) 26 (1894): n7-25. Miinter's hypothesis has not found its
way into the handbooks and reference books.
74. Virchow, "Ueber griechische Schadel" (above, n. 20), pp. 687-95.
75. The anecdote of the gluttonous high liver who is killed by a fatal grape,
which is too symbolic to be true, is also told of Anacreon and Pietro Aretino. As
for Sophocles, the Vita tells two other versions of his demise: he perished while
reading from one of his tragedies-a long tirade made him lose his breath-or he
died from joy on learning that he had won a literary contest. None of the three
versions stands up to historical criticism. The only thing known from a reliable
source is that he "died well, without having to suffer indignity" (Phrynichus, fr.
31). On this subject see J. Labarbe, "La mort tragique de Sophode," Bull. Classe
Lettr. Acad. Rny. Belgique 55 (1969): 265-92.
76. See M. S. Gazzaniga, The Bisected Brain (New York, 1970); R. W. Sperry,
"Lateral Specialization in the Surgically Separated Hemispheres," in F. 0. Schmitt
and F. G. Worden, eds., The Neurosciences: Third Study B-ogram (Cambridge, Mass.,
1974), pp. 5-19.
77. Plutarch, Pericles) 3, 3-4, in The Rise and Fall of Athens: Nine Greek Lives by
Plutarch) trans. lan Scott-Kilvert (Baltimore, 1960; rept. 1966), p. 167.
78. T. Fay, "The Head: A Neurosurgeon's Analysis of a Great Stone Portrait,"
Expedition (Philadelphia) 1, no. 4 (1958-59):12-18.
79. L. A. Schneider, Asymme-irie griechischer I(opfe vom 5. ]h. bis zum Hellenismus
(Wiesbaden, 1973).
80. The main arguments in favor of it are in F. Studniczka, "Das Bildnis Men-
anders," Neue Jahrb. klass. Altertumsgesch. 21 (1918):1-31. For a more recent and
more complete study, see M. Bieber, The History of the Greek and Rnman Theater
(London, 1961),pp. 82-92. Other scholars have thought they recognized the head
as Virgil's, but it is hardly likely that the statue is of Roman origin.
8r. Suda) s.v. Menandros; Phaedrus, Fabulae Aesopiae) v, 1, 12; Alciphron, Epistu-
lae) IV, 18, 4. See S. Charitonidis, L. Kahil, and R. Ginouves, Les mosaiques de la
maison de Menandre a Mytilene (Berne, 1970) (Antike J(unst) Beiheft 6), pp. 27-28
and pl. 1; J. I. Bungarten, Menanders und GlykerasBrief bei Alkiphron) thesis (Bonn,
1967).
82. Angel, "Skeletal Changes" (above, n. 28), pp. 75 and 81-82; Peopleof Lema
(above, n. 31), pp. 87ff.
83. Angel, "Mycenae" (above, n. 31), pp. 38off.
84. N. G. Gejvall and F. Henschen, "Two Late Skeletons with Malformations
and Close Family Relationship from Ancient Corinth," Opuscu/a Atheniensia 8
(1968): 179-93.
85. See G. Brauer and R. Fricke, "Zur Phanomenologie osteoporotischer Ver-
anderungen bei Bestehen systemischen hamatologischer Affektionen,'' Homo 31
(1980): 198-2rr.
86. Louvre Museum D rr78. See D. and M. Gourevitch, "Terres cuites helle-
nistiques d'inspiration medicale au Musee du Louvre," PresseMed. (1963),p. 2751,
fig. 3, and M. D. Grmek, La medecinegrecque c/assique(Memento therapeutique
Latema 1974) (Paris, 1974), pl. 9. The syndrome was described for the first time in
Nouvelle iconographiede la Salpe-triere25 (1912):280-85. For the typical appearance of
those so handicapped and accompanying disorders, see P. C. Windle-Taylor et al.,
Ann. Oto!. Rhin. Lar. 90 (1981):210-16.
87. For the archaeological description of this figurine, see A. N. Stillwell, Cor-
inth) vol. 15, pt. 2, The Potter)s Quarter: The Terracottas(Princeton, 1952), p. 143.
For medical commentary, see T. Skoog, "A Head from Ancient Corinth," Bull.
Hist. Dent. 19 (1971):50-54.
88. For the osteoarchaeological diagnosis of the two cases, see C. S. Bartsocas,
"Stature of Greeks of the Pylos Area during the Second Millennium B.C.," Hip-
pocratesMagazine 2, no. 2 (1977): 157-60, and "An Introduction to Ancient Greek
Genetics and Skeletal Dysplasias," Progr.Clin. Biol. Res. 104 (1982): 3-13. The rele-
vant text of the Iliad concerning Thersites is cited in chapter 1, above, p. 24. For
its medical interpretation, see C. S. Bartsocas, "Kleidokraniaki dysostosis par'
Omiro" Arch. Hellin. Paid. Hetair. 36 (1973):107-9.
89. Charles, "Argos r" (above, n. 32), pp. 280 and 3rr.
90. Angel, "Skeletal Changes" (above, n. 28), p. 82.
91. Angel, Peopleof Lema (above, n. 31), pp. 52, 55, and 92, pl. 24.
92. Hippocrates, Art., 51-59 and 62. See P. Bade, "Hippokrates und die ange-
borene Hiiftverrenkung," Acta chir. Scand. 67 (1930): 34-42, and M. Michler, Die
I(lumpfusslehre der Hippokratiker: Eine Untersuchung von De articulis (Wiesbaden,
1963) (Sudhoffs Arch. Gesch. Med.) suppl. 2). Dislocation of the hip is well depicted
on an Etruscan or Hellenistic bronze statuette (Klejman Collection, New York).
Beginning in the classical period, clubfoot appears on Greek vases in representa-
tions of Hephaestus's return to Olympus (Vienna 3577;Athens NM 664; Rhodes
10,7rr; Florence 4209; etc.).
93. For detailed documentation on this problem, see M. D. Grmek, "La paleo-
pathologie des tumeurs osseuses malignes,'' Histoire des sciencesmedicates9, no. 1
(1975-76): 21-50.
94. Specimen no. 1474 in the National Archaeological Museum in Florence.
The case was presented by L. Capasso to the Third European Congress of the
Paleo pathological Association at Caen (1980).
95. Several scholars have incorrectly supposed that they recognized cancer on
certain fragmentary reliefs and marble busts and on terra cotta figurines (e.g.,
Louvre D 586 and D rr46; Metropolitan Museum C.S. 1434). Some took for a
tumor what is just a fruit being held close to the chest. The absence of a breast on
an ancient bust is insufficient evidence for the diagnosis of an amputation per-
formed as treatment for cancer, since it could well be a mythological representation
(an Amazon) or mammary aplasia. The only cases in which a cancer diagnosis
should be taken seriously remain a Hellenistic marble bust with a damaged left
breast and a terra cotta statuette from Smyrna with deep ulceration of the mam-
mary region. Both cases were brought to light by Theodor Meyer-Steineg, who
tells us that the second was kept at the Museum of the Evangelical School of
Smyrna, The photographs Meyer-Steineg took of these two objects have often
been reproduced, but I do not know where the objects themselves are now. The
same is true for the clay head of a boy with a globular tumor on his right eye.
Obtained by Meyer-Steineg on the island of Cos, it has not been seen since the
last world war. See T. Meyer-Steineg, Darstellungen nonnaler und krankhaft veriin-
derter I(orperteile an antiken Weihgaben (Jena, 1912), and E. Hollander, Plastik und
Medizin (Stuttgart, 1912). According to G. Oschmann (1970), what remains of
Meyer-Steineg's collection has been given to the Institute of Pathology at Jena.
The Museum of Taren tum has a figurine that, judging by photographs, resembles
the one published by Meyer-Steineg and suggests a diagnosis of sarcoma of the eye
socket or of retinoblastoma. See A. Galeone, "Stati patologici nell'arte greca,"
Atti Mem. Accad. Stor. Arte San. 4 (1938):332, fig. 3.
96. Angel, Peopleof Lerna (above, n. 31), pp. 51, 89, and 92; pl. 24.
97. Gout is mentioned twenty-two times in the Hippocratic treatises and ap-
pears in them as an everyday disease whose nature and symptoms need no account.
The oldest case of gout known to history is that of Hieron, tyrant of Syracuse
(died ea. 466 B.c.), as mentioned in Pindar Pyth) I, 90.
98. A lcibiades) 2, 140a.
99. Elliott Smith and F. Wood Jones, "The Pathological Report," Bull. Arch.
Survey of Nubia 2 (1910).
100. C. Wells, "A Palaeopathological Rarity in a Skeleton of Roman Date,"
Medical History 17 (1973):399ff.
101. Although Pliny the Elder (Natural HistO'lj'yXXVI, 100) considers gout in Italy
"an ailment of foreign origins," i.e., a poisonous gift from the Greeks, for Lucian
(Pro mere. cond., 31and 39) its spread is due to the Romanization of the world, and
for Seneca (Epist.) 95) it is the result of moral decay, especially gluttony.
102. C. Wells, "Prehistoric and Historical Changes in Nutritional Diseases and
Associated Conditions," Progressin rood and Nutrition Science no. n (1975):729-79,
esp. pp. 758-60.
103. See M. M. Boross and J. Nemeskeri, "Ein bronzezeitlicher Nierenstein aus
U ngarn," Homo 14 (1963):149ff. Paolo Orsi, former director of the Archaeological
Museum in Syracuse, has said in a private letter that he has seen, during one of
the many excavations in Sicily, a large urinary stone. This statement, disclosed by
W. Ebstein (Janus 5 [1900]: 333), is extremely vague, lacking any indication of the
location or the date of the tomb in question.
105. J. L. Angel, "Skeletal Material from Attica" (above, n. 31), pp. 308ff.
106. Herodotus, IX, 83.
107. Plutarch, Pyrrhus) 3; Pliny, Natural HistO'lj'yVII, 69; Tzetzes, Chi!., III, 950;
and Valerius Maximus, 1, 8, 12. See C. S. Bartsocas, "Complete Absence of the
Permanent Dentition: An Autosomal Recessive Disorder," American Journal of
Medical Genetics 6 (1980): 333ff.A similar malformation exists in a skull from Monte
Cassino. On a mandible from Mycenaean Pylos there is a smooth bony surface
where the teeth should be, except for one canine tooth and the socket of the
other. Cf. S. Marinatos, "Pylos Excavations," Praktika tis en Athenais Archaiol.
Etair. (1966), pp. 195-209.
108. Herodotus, VII, n7. The same author (I, 68) relates the discovery in Tegea
of the bones of a man who measured 7 cubits (10 feet!). Considered the remains of
Orestes, they must have been fossilized bones of a giant animal from the Pleisto-
cene. See G. Huxler, "Bones for Orestes," Greek)&man) and Byzantine Studies 20
(1979): 145-48.
109. A typical example is the statuette in the Louvre numbered D 1176. The
overall appearance of the face is characteristic of the disease. See, on this subject,
M. and D. Gourevitch, "Terres cuites" (above, n. 86), p. 2751, fig. 2, and
Grmek, La medecinegrecque (above, n. 86), pl. 8. I am not convinced of the diag-
nosis of acromegaly for most of the terra cotta figurines cited as examples of its
artistic representation. Likewise, I do not agree with Gerald D. Hart, who be-
lieves he has identified this disease on coins struck with the image of Ptolemy I
Soter ("The Diagnosis of Diseases from Ancient Coins," Archaeology26 [1973]:127).
no. See the publications of F. Regnault, P. Richer, H. Meige, A. Galeone,
D. and M. Gourevitch, C. Wells, and C. S. Bartsocas. In collaboration with
Danielle Gourevitch, I am preparing a critical reexamination of Greco-Roman
pathological iconography.,
111. H. Grimm, "U eber Rachitis und Rachitis-V erdachtsfille im ur- und friihge-
schich tlichen Material," Zschr.ges. Hyg. 18 (1972): 451-55, and Wells, "Prehistoric
and Historical Changes" (above, n. 102), pp. 756ff.
112. W. Ebstein, "Ueber das Vorkommen der Rachitis im Alterthum," Janus 5
(1900): 332-37.
113. G. Regoly-Merei, "Paleopathological Examination of Skeletal Finds in the
Roman Period and Description of Diseases in Greek and Roman Medical Texts,"
Medical History in Hungary (1970), pp. 58ff.
114. In a study of pathological traits in the work of statuette sculptors from Asia
Minor, Jean-Martin Charcot and Paul Richer have drawn attention to heads with
"the most diverse kinds of skull deformations," which are "placed on the de-
formed torsos of rickets-sufferers" (Lesdifformeset lcsmaladesdans tart [Paris, 1889],
p. 9). Specifically, they mention a statuette of a young man "endowed with the
thinnest legs" and a "chest rounded like a hunchback." The terracotta in ques-
tion is Louvre D 573. It was found in Pergamum, and Simone Besques's catalogue
(3:96) dates it to the second century B.C. I have examined it and believe that the
diagnosis of rickets is only one among many possibilities. The "rickety" appear-
ance of a Hellenistic terra cotta in the collection of P. M. de la Charlonie is more
convincing (Musee Archeologique Municipal de Laon, inv. no. 37.376). The diag-
nosis of rickets for an Etruscan votive torso from Lucera is not compelling, despite
E. Greco, "La patologia nella antichita classica attraverso lo studio degli ex-voto
anatomici," Il Policlinico67, sez. prat. (1960): p. 1248.
115. For instance, see Soranos, Gyn.J I, 112, and II, 43-45, and Galen, De san.
tuendaJ 7.
116. Herodotus, III, 12, trans. A de Selincourt (Baltimore, 1954; rept. 1968), p.
178. The battle of Papremis took place in 460 B.C. at the time of the Egyptian
uprising against the Persian satrap Achaimenes.
117. Actually, following Coray, Littre, and Hirsch, some philologists and histo-
rians associate certain passages iIJ.the Hippocratic corpus with scurvy (especially
Int.J 46, and Epid.J VII, 47), but that interpretation must be forcefully rejected.
First of all, what nineteenth-century scholars understand by "scurvy" does not
completely overlap with current notions of the vitamin C deficiency disease from
a clinical or etiological point of view. An ulcerous, gangrenous affliction on the
mouth is not necessarily scurvy, and when it is linked with chronic spleen enlarge-
ment (as often in classical accounts) a diagnosis of malarial cachexia is much more
likely than vitamin deficiency. The "bloody intestinal obstruction" (eilcoshaima-
tftes) in Int., 46, is certainly not scurvy (paceLS] s.v., which is inspired by Littre's
translation of the passage [VII, 282, in his edition]) but a toxic purpura. In the case
of Kleokhos (Epid..,vn, 47), Littre's interpretation is untenable from the mere fact
that the patient took excessive amounts of honey. He suffered from either a blood
disorder or an inflammation of the lymphatic vessels accompanied by an infectious
ulcerous stomatitis.
n8. Strabo and Pliny tell how the Roman armies were struck by an epidemic
that made their mouths stink, was caused by water of poor quality, and could be
treated by a certain plant (Pliny, Natural Histo'Y)';xxv, 3; Strabo, Geogr.)16). The
information about symptoms given by these authors is so meager that it supports
either a diagnosis of scurvy or one of infectious stomatitis. For the details, see
M. D. Grmek, "Les origines d'une maladie d'autrefois: Le scorbut des marins,"
Bull. Inst. Oceanogr.Monaco) special no. 2 (1968), pp. 505-23, esp. p. 508.
n9. Wells, "Prehistoric and Historical Changes" (above, n. 102), pp. 756ff.
120. See M. Baudouin, "La prehistoire du rhumatisme chronique: La plus vieille
maladie du monde," Mid. Internat., no. 2 (1923),pp. 43-48; Pales, Paliopathologie
(above, n. 3); H. Grimm, Vm;geschichtliches) frugeschichtliches und mittelalterliches
Fundmaterial zur Pathologieder Wirbelsitule (Leipzig, 1959) (Nava Acta Leopoldina 21,
no. 142); H. Buess and H. Koelbing, I(urze Geschichteder ankylosierendenSpondylitis
und Spondylose(Basel, 1964), (Documenta Geigx Acta Rheumatologica)no. 22).
121. J. B. Bourke, '' A Review of the Paleo pathology of the Arthritic Diseases,''
in Brothwell and Sandison, Diseases (above, n. 7), pp. 352-70; C. M. Cassidy,
"Arthritis in Dry Bones: Diagnostic Problems," Henry RJrdHosp. Med. Journal 27
(1979): 68ff; A. Cockburn et al., "Arthritis, Ancient and Modern: Guidelines for
Fieldworkers," ibid., pp. 74-79.
122. The disease is known in French as arthrose. Certain French writers use osteo-
arthrite hypertrophiquedegenerativeas a synonym, but this expression is considered
inappropriate by eminent rheumatologists. In fact, the term arthrite should be
reserved for inflammatory processes involving the synovial membrane. For up-to-
date knowledge on rheumatic ailments, see S. de Seze et al., Le diagnostic en
rhumatologie (Paris, 1978), and W. N. Kelley et al., Textbookof Rheumatology (Phila-
delphia, 1981).
123. For example, see R. D. Jurmain, "Stress and the Etiology of Osteoarthri-
tis," Amer. ]our. of Phys. Anthro. 46 (1977): 353-66.
124. W. M. Krogman, "The Skeletal and Dental Pathology of an Early Iranian
Site," Bull. Hist. Med. 8 (1940): 28-48.
125. Angel, "<;atal Hiiyiik" (above, n. 31), p. 91 and pl. 2.
126. B. Alpagut, "Some Paleopathological Cases of the Ancient Anatolian Man-
dibles," J. Human Evolution 8 (1979): 571-74.
127. Angel, PeopleofLerna (above, n. 31), p. 52.
128. Dastugue and Duday, "Les ossements humains" (above, n. 32).
129. Furst, A13olis (above, n. 26), p. 12 L.
130. Angel, Peopleof Lema (above, n. 31), p. 89 and pl. 24,.
131. Ibid., p. 89. For a description of an occupational disease of the arm in the
patients of Hippocrates, see below, chapter 12, the section "Some Individual Case
Histories From Perinthus."
132. Charles, "Argos r," p. 3n, andA13osrr) p. 67 (both above, n. 32).
133. French authorities call this disease spondylarthrose;the majority of Anglo-
American medical writers use the term "spondylitis."
134. See J. Forestier, F. Jacqueline, and J. Rotes-Querol, La spondylarthriteanky-
losante (Paris, 1951),and J. Forestier and R. Lagier, "Ankylosing Hyperostosis of
the Spine," Clin. Orthoped. 74 (1971):65-83. The definition ofForestier's syndrome
is recent. It calls for the revision of some earlier osteoarchaeological diagnoses,
especially cases thought to be ankylosing spondylitis. In many cases of Forestier's
syndrome, one can see not just vertebral hyperostosis but also diffuse extraspinal
osteophyte growth. ,Cf. D. Resnick et al., "Diffuse Idiopathic Skeletal Hyperos-
tosis (DISH): Forestier's Disease with Extraspinal Manifestations," Radiology 115
(1975):513-24. It appears that DISH is tied to metabolic factors such as obesity and
diabetes.
135. See M. D. Grmek, "Die Wirbelsaule im Zeitgeschehen," Med. Welt 25
(1974): 70-76.
136. Furst, A1lJolis (above, n. 26), pp. 14 and 122, fig. 49.
137. See p. 74.
138. Angel, "Mycenae" (above, n. 31), pp. 381ff.,pl. 249.
139. Angel, People of Lema (above, n. 31), pp. 58ff., pl. 25.
140. Aff) 29 (Littre, VI, 240-42).
141. Furst, A1lJolis (above, n. 26), p. 122.
142. Angel, PeopleofLerna (above, n. 31), pp. 88ff., pl. 25.
143. Angel, "Some Probldns" (above, n. 28), p. 284.
144. Angel, "{_;atal Huyiik" pp. 85 and 90. See also Angel, "Karata~" (above,
n. 31), p. 253, where it is stated that this disease affects 66 percent of the men's
vertebrae and 57 percent of the women's in the necropolis.
145. Angel, People of Lema (above, n. 31), pp. 86ff.
146. Angel, "Osteoarthritis in Prehistoric Turkey and Medieval Byzantium,"
Henry Ibrd Hosp. Med. ]. 27 (1979): 38-43.
147. See Angel, "Skeletal Changes" (above, n. 28), p. 82; People of Lema (above,
n. 31), p. 87; "Mycenae" (above, n. 31), p. 379. Dastugue and Duday ("Les osse-
ments humains" [above, n. 32]) found Schmorl's nodule in a Neolithic thoracic
vertebra from a cave near Laurion (1726I(itsos).
148. See G. Bergmark, "Jehles och Scheuermanns ryggradforandringar bid antika
skulpturer" ("Jehle's and Scheuermann's Deformations of the Spinal Column on
Sculptures from Antiquity"), Nordisk Medicin 33 (1947): 325.
149. I use this term in its modern sense, to refer to acute or chronic joint diseases
whose initial and principal lesion is an inflammation of the synovial membrane.
150. Aff) 30 (Littre, VI 242). For rheumatism in ancient Greek medical literature,
see especially A. Delpeuch, Histoire des maladies: La goutte et le rhumatisme (Paris,
1900), and A. Ruiz Moreno, Las afecciones reumaticas en el Corpus Hippocraticum
(Buenos Aires, 1941). For the commonness of sore throats in antiquity, see below,
chapters 4 and 12.
151. See J. M. Riddle, "Rheumatoid Arthritis," Henry Ibrd Hosp. Med. J. 27
(1979): 18-23, and Kelley et al., Textbook of Rheumatology (above, n. 122).
152. See Cockburn et al., "Arthritis, Ancient and Modern" (above, n. 121),pp.
77-79; T. Appelboom et al., "Rubens and the Question of the Antiquity of
Rheumatoid Arthritis,"]. Amer. Med. Ass. (1981),pp. 483-86; and D. E. Caughy,
"The Arthritis of Constantine IX," Ann. Rheum. Dis. 33 (1974): 77-80.
153. C. L. Short, "The Antiquity of Rheumatoid Arthritis," Arthr. Rheum. 17
(1974): 193-205. His opinion is not shared by Ronald E. Domen, who cites medieval
European and American cases of the disease and stresses the diagnosis in the
Egyptian mummy mentioned above. See his attempt to set the record straight,
"Paleopathological Evidence of Rheumatoid Arthritis,"] . .i:lmer. Med. Assoc. 246
(1981):1899.
154. See P. Stastny, "Immunogenetic Factors in Rheumatoid Arthritis," Clin.
Rheum. Dis. 3 (1977):315-32,and V. Lemaire, ''Polyarthrite rhumato"ide et systeme
HLA," Concours Med. 102 (1980): 1967-69.
155. See W. W. Buchanan, ''The Contribution of History to the Study of the
Aetiology of Rheumatoid Arthritis,'' Med. Hist. 23 (1979): 229ff.
156. L. L. Klepinger, "Paleopathological Evidence for the Evolution of Rheu-

n1atoid Arthritis," Amer.]. Phys. Anthro. 50 (1978):n9-22.
157. See PaleopathologicalNewsletter) no. 25 (March 1979), pp. 1off., and no. 26
(June 1979), pp. 6ff.
158. L. Sokoloff et al., "Spinal Ankylosis in Old Rhesus Monkeys," Clin. Or-
thop. 61 (1968): 285-93.
159. See Forestier, Jacqueline, and Rotes-Querol, La spondylarthrite (above, n.
134); S. de Seze and M. Phankim-Chapuis, "Naissance de la pelvi-spondylite rhu-
matismale,'' Histoire de la Medecine 10 (1960): 44-53; Buess and Koelbing, I(urze
Geschichte (above, n. 120); D. A. Brewerton et al., "Ankylosing Spondylitis and
HLA-B 27," Lancet 1 (1973): 904-7; C. V. Jimenez, "Ankylosing Spondylitis,"
Henry lord Hosp. Med. J. 27 (1979): 10-13; D. G. Spencer, R. G. Sturrock, and
W. W. Buchanan, "Ankylosing Spondylitis: Yesterday and Today," Med. Hist.
24 (1980): 60-69.
160. It was described and diagnosed by Leon Pales. See Buess and Koelbing,
I(urze Geschichte (above, n. 120), p. 26.
161. See the review of their research by P. A. Zorab, "The Historical and Prehis-
torical Background of Ankylosing Spondylitis," ltoc. Roy.Soc. Med. 54 (1961):415-20.
162. See Bourke, "Paleopathology of the Arthritic Diseases" (above, n. 121),
pp. 357-60.
163. E. Benassi, "I raggi x al servizio dell'archeologia," llivista Biellese 10 (1955):
n-14.
164. Art.) 47 (Petrequin, II, 4n).
165. See M.A. Ruffer, "Note on the Presence ofBilharzia haematobia in Egyptian
Mummies of the Twentieth Dynasty (1250-1000B.c.)," Brit. Med.]. 1 (1910):16.
166. Angel, Peopleof Lerna (above, n. 31), pp. 48 and 93.
167. See G. Regoly-Merei, "Palaopathologische und epigraphische Angaben zur
Frage der Packen in Altagypten," Sudhoffi Arch. Gesch. Med. 50 (1966): 4n-17.
168. The specimen was presented to the Colloquium in Paleopathology at
Toronto, 1978.
169. J. Dastugue, "Pathologie des hommes fossiles de l'abri de Cro-Magnon,"
Anthropologie 71 (1967): 470-92.
170. P. L. Thillaud, "L'histiocytose x au Paleolithique (sujet n° 1 du Cro-Mag-
non)," Anthropologie 85 (1982):219-39.
Chapter Three: Paleodemography

r. Aside from the publication of Angel cited in the previous chapter, see
D. R. Broth well, "Palaeodemography," in W. Brass, Biological Aspects of Demog-
raphy (London, 1971), pp. 1n-29; C. Masset, "La demographie des populations
inhumees: Essai de paleodemographie," L)Homme 13 (1973):95-161;and J. D. Wil-
ligan and K. A. Lynch, Sources and Methods of Historical Demography (New York,
1982).
2. The conclusions that follow are my own. They are inspired, in part, by
the excellent article of J. L. Angel, "Health and the Course of Civilization as
Seen in Ancient Greece," The Interne (1948), pp. 15-48. For an overall orientation,
I have consulted chiefly F. Heichelheim, An Ancient Economic History from the
Palaeolithic Age to the Migrations of the Germanic) Slavic) and Arabic Nations 3 vols.
(Leiden, 1958-70); P. Leveque, L)aventure grecque) 2d ed. (Paris, 1968); and K. W.
Butzer, Environment and Archaeology: An EcologicalApproach to ltehistory (Chicago,
1971).
3. See especially R. J. Rodden, "An Early Neolithic Village in Greece," Sci-
entific American 212 (1965): 82-90 (on Nea Nikomedia); J. Mellaart, <;;atalHuyuk: A
Neolithic Town in A'l}atolia (New York, 1967); F. L. Dunn, "Epidemiological Fac-
tors: Health and Disease in Hunter-Gatherers," in R. B. Lee and I. de Vore, Man
the Hunter (Chicago, 1968); R. Tringham, Hunters) Fishers) and Farmers of Eastern
Europe) 6000-3000 B. c. (London, 1971); and J. Cauvin, Les premiers villages de Syrie-
Palestine du IXe au VIP millenaire avant]. C. (Lyon, 1978). For some optimistic
assessments, see M. Sahlins, Stone Age Economics (Chicago, 1972), and for a discus-
sion of difficulties on the level of demography, see M. N. Cohen, The RJOdCrisis
in H-ehistory: Overpopulation and the Origins of Agriculture (New York and London,
1977). For the medical side, see especially T. A. Cockburn, ''Infectious Diseases in
Ancient Populations," Curr. Anthro. 12 (1971): 45-62. Cf. also S. Clarke, "Mortal-
ity Trends in Prehistoric Populations," Hum. Biol. 49 (1977): 181-86.
4. For the prehistoric periods in question, see E. Vermeule, Greece in the
Bronze Age (Chicago, 1964)~ G. E. Mylonas, Mycenae and the Mycenaean Age
(Princeton, 1966); and M. I. Finley, Early Greece: The Bronze and Archaic Ages
(London, 1970). Sanitary conditions are analyzed by C. P. W. Warren, "Some
Aspects of Medicine in the Greek Bronze Age,'' Med. Hist. 14 (1970): 364-77.
5. Warren, "Some Aspects" (above, n. 4), pp. 367-71; F. Schultze, "Rein-
lichkeit, Bad und Wasche bei den mykenischen Griechen," Munch. Med. Wschr.
84 (1937): 743-45; and H. Schmidt-Ries, Wasser fur Hellas: Das Wasser in alt-
griechischen Raum (Diisseldorf, 1956).
6. See Anthony M. Snodgrass, The Dark Age of Greece (Edinburgh, 1971), and
Finley, Early Greece (above, n. 4).
7. According to Camp, Greece, and especially Attica, underwent a particularly
harsh and long drought around 700 B.c. that would have caused famine, epidem-
ics, and an abrupt rise in the death rate. This scholar's opinion is based on the
relatively high number of abandoned wells in Athens at the time in question. See
J. McK. Camp, "A Drought in the Late Eighth Century B.c.," Hesperia 48 (1979):
397-4n. The arguments against this hypothesis are more convincing. Snodgrass
points out that the number of sites and tombs in Greece points to an enormous
upsurge of population, at least locally, in the two generations before 700 B.C.
Between 780 and 720 B.c., the population of Attica increased by a factor of seven.
See A. M. Snodgrass, Archaic Greece: The Age of Experiment (London, 1980).
8. Angel, "Health and Civilization" (above, n. 2), pp. 16-17. See F. Cha-
moux, La cir>ilisationgrecque a Fepoque archai'que et classique (Paris, 1963), and J.
Charbonneaux, R. Martin, and F. Villard, La Grecearchaique (Paris, 1968).
9. F. Chamoux, La cir>ilisation(above, n. 8); R. Flaceliere, La viequotidienne en
Grece au siecledePericles (Paris, 1959); G. Glotz, La citegrecque) rev. ed. (Paris, 1968);
M. Austin and P. Vidal-Naquet, Economie et societe en Grece ancienne: Periodes
archaique et classique (Paris, 1976) (Economic and Social History of Ancient Greece: An
Introduction) trans. and rev. M. M. Austin [Berkeley and Los Angeles, 1977]).
10. J. Burckhardt, Griechische I(ultu1'l]eschichte (1898-1902), ed. R. Marx (Stutt-
gart, 1941), 3: 136.
n. Angel, "Health and Civilization" (above, n. 2) p. 46.
12. Cf. W. H. McNeill, Plagues and Peoples (New York, 1976).
13. For the geography of ancient Greece, see the old but still commendable
works of C. Neumann and J. Partsch, Physikalische Geographie von Griechenland mit
besondererR.iicksicht auf das Alterthum (Breslau, 1885), and L. Lacroix, Iles de la Grece
(Paris, 1853). An exemplary summary statement in accord with modern geographical
concepts is to be found in N. G. L. Hammond, "The Physical Geography of
Greece and the Aegean," in A. J.B. Wace and F. H. Stubbings, A Companion to
Homer (Londo.:-1,1963).
14. The healthiness of the Greek climate and the importance of geographic
factors for the flowering of Greek civilization are well brought out by H. E.
Sigerist, HistoryofMedicine (New York, 1961),2: n-16 (''Archaic Medicine in Greece:
The Setting'').
15. Cf. Finley, Early Greece(above, n. 4), p. 6.
16. See E. Martini, Wege der Seuchen (Stuttgart, 1954), and McNeill, Plagues
(above, n. 12).
17. Aristotle, Meteor")I, 14. For the notion of "malarial complex" and the
influence of other pathogenic geographical factors, see M. Sorre, "Complexes
pathogenes et geographie medicale," Ann. Geograph.42 (1933):1-18, and Lesfonde-
ments biologiquesde lageographiehumaine (1951;rev. ed. Paris, 1971).
18. Plato, Laws) 707e.
19. See M. D. Grmek, "Geographie medicale et histoire des civilisations,"
Annales E. S. C. 18 (1963):1071-97 (esp. pp. 1073ff.).
20. R. Joly, Hippocrate(Paris, 1964), p. 75. See also R. Joly, Le niveau de la science
hippocratique(Paris, 1966), pp. 180-210.
21. See above, Introduction, the section, "The Greek Pathocoenosis in the
Classical Period.''
22. Airs) Waters) and Places)23, in Hippocratestrans. W. H. S. Jones (London
and Cambridge, 1946), 1: 131-33.
23. C. McEvedy and R. Jones, Atlas of World Population History (Harmonds-
worth, 1978), p. 19. See also E. S. Deevey, "The Human Population," Scientific
American 203 (1960): 195-205, and R. H. Ward and K. M. Weiss, eds., The Demo-
graphic Evolution of Human Populations(London, 1976).
24. C. Renfrew, "Patterns of Population Growth in the Prehistoric Aegean,"
in P. J. Ucko et al., Man) Settlement) and Urbanism (London, 1972), pp. 383-99.
See also W. A. McDonald and R. H. Simpson, "Further Explorations in South-
western Peloponnese,'' American Journal of Archaeology73 (1969): 123-77.
25. Renfrew, "Patterns of Population" (above, n. 24), pp. 394 and 397.
26. J. L. Angel, "Ecology and Population in the Eastern Mediterranean,"
WorldArchaeology4 (1972): 88-105 (esp. p. 95).
27. J. L. Angel, The Peopleof Lema (Princeton and Washington, D. C., 1971),
pp. 109ff.
28. See J. L. Angel, "The Bases of Paleodemography," American Journal of
PhysicalAnthropology30 (1969): 427-38. Unfortunately, statistics on fecundity among
prehistoric populations, which are based on the state of scars on public symphyses,
cannot be accepted as absolute values. It is preferable to consider them vague
indicators of some demographic tendencies. The enthusiasm of anthropologists
using this method must be set against the observations of forensic medicine, which
views it with suspicion. See especially B. Herrmann and T. Bergfelder, "Ueber den
diagnostischen Wert des sogenannten Geburtstrauma am Schambein bei der Iden-
tifikation," Zschr. Rechtsmedizin 81 (1978):73-78.
29. The fundamental work on this subject remains J. Beloch, Die Bevolkerung
dergriechisch-romischenWelt (Leipzig, 1886;rev. ed. Rome, 1968). Important further
considerations are in E. Cavaignac, Histoire de l)Antiquite vol. 2 (Paris, 1913),and
especially P. Salmon, ''La population de la Grece antique,'' Bulletin de l)Association
Guillaume Bude) 4th ser., 18 (1959):449-76. It is interesting to note that, by calcu-
lations of Jarde based on the production and consumption of wheat, this degree
of density represents a kind of biological limit. See below, at n. 107.
30. Salmon, "La population" (above, n. 29), pp. 457ff.
31. At about A.O. 310, Demetrius of Phalerum is said to have had a census done
of the population of Attica. According to Ctesicles (Athenaeus, VI, 272b), the
surprising and unlikely result was a count of 21,000 citizens, 10,000 resident aliens,
and 400,000 slaves. On problems and new perspectives in demographic research
on antiquity, see J. N. Corvisier, "La demographie historique est-elle applicable a

l'histoire grecque?" ~nn. Demog. Hist. (1980), pp. 161-84.
32. Thucydides, II, 13 and 31.
33. Other than the works of Beloch and Salmon cited above, n. 29, see espe-
cially A. W. Gomme, The Population of Athens in the Fifth and RJurth Centuries B.C.)
(Oxford, 1933); G. Glotz and R. Cohen, Histoire grecque) vol. 2, La Greceau VCsiecle
(Paris, 1938); and M. H. Hansen, "The Number of Athenian Hoplites in 431 B.C .. "
Symb. Osloens. 56 (1981): 19-32.
34. Thucydides, II, 14-16.
35. Salmon, "La population" (above, n. 29), p. 453.
36. See A. Landry, "Quelques apen;:us concernant la depopulation clans l' An-
tiquite greco-romaine," Rev. Hist. 177 (1936): 1-33, and J. Berard, "Problemes de-
mographiques clans l'histoire de la Grece antique," Population 2 (1947): 303-12. It
is instructive to compare the-reality of demographic problems with their image in
the minds of learned men. In this connection, see J. Moreau, "Les theories
demographiques clans l' Antiquite grecque," Population 4 (1949): 587-613.
37. Polybius, xxxvrr, 9.
38. Some specialists guess that measles and influenza, if they are to maintain
themselves, require an interacting community of approximately 500,000 members.
Cf. G. J. Armelagos and A. McCardle, "Population, Diseases, and Evolution,"
American Antiquity 40, no. 2, pt. 2 (1975): 1-10.
39. See F. L. Black, "Infectious Diseases in Primitive Societies," Science 187
(1975): 515-18. Macfarlane Burnet has shown that the most likely origin of the
measles virus is the germ resembling it that causes a certain disease in dogs. The
smallpox virus is linked genetically to the group of pox viruses of domesticated
animals, especially the cowpox virus. Since viruses are products of a degenerative
evolution of the genes of bacteria or protozoa that cannot maintain themselves in
free state, their continuing existence demands conditions favorable to immediate
contagion, in particular significant population density. But the prerequisite applies
only to the germs of acute diseases. Those viruses that live alongside humans
without causing much harm can be transmitted vertically and originate in phylo-
genetic ancestors, not domestic animals. Most likely that is the case for the herpes-
virus, and perhaps that of chickenpox and shingles. See M. Burnet, Viruses and
Man (Harmondsworth, 1955), and the same author's chapter in G. W. Leeper, The
Evolution of Living Organisms (Melbourne, 1962). See also G. de Beer, "Virus et
prehistoire," Archeocivilisation 7 (1968): 1-3.
40. The determination of age at death from these skeletal remains is based on
tooth eruption, formation of long bones, cranial suture closure, metamorphosis
of the public symphysis, radiographic translucency of proximal femur and hu-
merus, dental wear, and a variety of minor indicators. Substantial bias exists in the
present methods of analyzing cranial sutures and pubic symphysis metamorphosis.
41. J. L. Angel, "The Length of Life in Ancient Greece" J. Geront. 2 (1947):
18-24. See also S. Jarcho, "The Longevity of the Ancient Greeks," Bull. N. Y
Acad. Med. 43 (1967): 941-43.
42. Cf. J. L. Angel, "Skeletal Changes in Ancient Greece," American Journal
of Physical Anthropologx 4 (1946): 69-97 (esp. p. 72).
43. D. M. Robinson, Necrolynthia pt. 9 (Baltimore, 1942), 146-70.
44. Angel, "Length of Life" (above, n. 41), pp. 21ff.
45. See especially J. L. Angel, "The Bases of Paleodemography," American
Journal of Physical Anthropology n.s., 30 (1969): 427ff., and his "Ecology and Popu-
lation" (above, n. 26).
46. H. V. Vallois, "La duree de la vie chez l'homn1e fossile," Anthropologie 47
(1937): 499-532 (see also Comptes rendues de PAcadimie des Sciences204 [1937]: 60-62).
NOTES TO PAGES IOI-7
47. J. N. Biraben, "Duree de la vie clans la population de Columnata (Epi-

paleolithique oranais)," Population (1969), pp. 487-500.
48. Angel, "Ecology and Population" (above, n. 26), pp. 95ff.
49. Mellaart, <;atalHiiyiik (above, n. 3).
50. J. L. Angel, "Early Neolithic Skeletons from <::=atal Hiiyiik: Demography
and Pathology," Anatolian Studies 21 (1971): 77:-98.
51. Ibid. , p. 78.
52. Angel, "Ecology and Population" (above, no. 26), pp. 95ff.
53. J. L. Angel, in M. J. Mellink, "Excavations at Karata~-Semayiik and Elmali,
Lycia 1969," Amer.]. Archaeol. 74 (1970): 245-59.
54. J. L. Angel, "Early Bronze Age Karata~ People and Their Cemeteries,"
Amer.]. Archaeol. 80 (1976): 385-91.
55. See G. E. Mylonas, Aghios I(osmas: An Early Bronze Age Settlement and
Cemetery in Attica (Princeton, 1959), pp. 169-79.
56. Angel, People of Lerna (above, n. 27), pp. 109ff.
57. See J. L. Angel, "Human Skeletons from Grave Circles at Mycenae," in
G. E. Mylonas, 0 tafikos kyklos B tou Mykenou (Athens, 1973), pp. 379-97.
58. R. P. Charles, "Etude anthropologique des necropoles d' Argos," Bull.
Corresp. Hell. 82 (1958): 258-313, esp. pp. 307ff.
59. R. P. Charles, Anthropologie archeologiquede la Crete (Paris, 1965), pp. 208ff.
60. V. V. Caramalea et al., "Contributi paleodemografice la studiul unor com-
munitati tribale din epoca bronzului de pe teritoriul Rominiei," R--oblemeAntrop.
7 (1963): 243-61.
61. M. S. Senyiirek, "The Duration of Life in the Chalcolithic and Copper Age
Population of Anatolia," Anatolia 2 (1957): 95-110.
62. Angel, "Bases of Paleodemography" (above, n. 28), pp. 429-31.
63. See C. Wells, "Ancient Obstetric Hazards and Female Mortality," Bull.
r
N. Acad. Med. 51 (1975): 1235-49.
64. For these tables and their use, see L. Henry, Manuel de demographie histo-
rique (Geneva and Paris, 1967).
65. B. E. Richardson, Old Age among the Ancient Greeks (Baltimore, 1933).
66. Ibid., p. 234.
67. V. G. V allaoros, "E meoe diarkeia t~ zo~a efs t~n arkhafan Hellada,"
R--aktika ~s Aka&mias AthenlJn 13 (1938): 401-9.
68. J. C. Russell, "Late Ancient and Medieval Populations," Trans. Amer.
Phil. Soc.) n.s., 48, pt. 3 (1958): pp. 24-30.
69. M. Hombert and C. Preaux, "Note sur la duree de la vie clans l'Egypte
greco-romaine," Chronique d)Egypte 20 (1945): 139-46. For a more detailed analysis
by site, see B. Boyaval, "Remarques sur !'indication d'age de l'epigraphie funeraire
grecque d'Egypte," Zschr. Pap. Ep-ig. 21 (1976): 217-50.
70. For example, see A. R. Burn, "Hie Breve Vivitur:· A Study on the Expec-
tation of Life in the Roman Empire," Past and R--esent)no. 4 (1953), pp. 2-31; R.
Etienne, "Demographie et epigraphie," Atti III Co1'!!Jr.Intern. Ep-igr. Gr. Rnm.
(Rome, 1959), pp. 414-19; J. D. Durand, "Mortality Estimates from Roman Tomb-
stone Inscriptions," Amer. four. Socio/. 65 (1960): 365-73; and M. Clauss, "Probleme
der Lebensalterstatistiken aufgrund romischer Grabinschriften," Chiron 3 (1973):
395-417.
71. L. Henry, "La mortalite d'apres les inscriptions funeraires," Population 12
(1957): 149-52, and "L'age au deces d'apres les inscriptions funeraires," Population
14 (1959): 327-29. See also the excellent study by K. Hopkins, "On the Probable
Age Structure of the Roman Population,'' Popul. Stud. 20 (1966): 245-64, and the
conclusions of B. Boyaval, "Epigraphie antique et demographie: Problemes de
methodes," Rev. Nord 59 (1977): 163-91.
72. Aristotle, Politics) IV, 16, 1334b-1335a.

73. Thucydides, "II, 20.
74. See A. H. M. Jones, The Athenian Democracy (Oxford, 1957; rept. Balti-
more, 1986), pp. 82ff., and O. W. Reinmuth, The Ephebic Inscriptions of the EJurth
Century, Mnemosyne) suppl. (1971).
75. M. D. Grmek, On Ageing and Old Age (The Hague, 1958)(Monographiae
Biologicae, V, 2), and "Le vieillissement et la mort," in Encydopedie de la Pliiade:
Biowgie (Paris, 1965),pp. 789-99.
76. Iliad r.247ff (trans. R. Lattimore).
77. Iliad XXIII. 627ff.
78. Thucydides, II, 36.
79. Thucydides, VI, 12.
So. See above, p. 107. For the argument between Nicias and Alcibiades, its
relationship to transgenerational conflict, and the use of medical metaphors in
political discourse, see J. Jouanna, "Politique et medicine: La problematique du
changement clans le Regime des maladies aigues et chez Thucydide (livre VI)," in
Hippocratica (Paris, 1980), pp. 299-319.
81. See Grmek, On Ageing (above, n. 75), p. 30.
82. Psalms 90:10.
83. Herodotus, III, 22ff.
84. Censorinus, De die natali) 14, 12, and 15, 1-2. Plato himself died on the eve
of his eighty-first year.
85. For the dubious longevity of the ancient Greek philosophers, see L. Jerpha-
gon, "Les mille et une morts des philosophes antiques," Revue Be{!JePhiwl. Hist.
59 (1981):17-28.
86. See Grmek, On Ageing (above, n. 75), pp. 32-38, and A. Sauvy, Les limites
de la vie humaine (Paris, 1961).
87. Fragment of a stele from Lycia (Suppl. Epigr. Gr") II, 690). We decline to
take into account two late funerary inscriptions from Egypt, one of which, of
Panopolitan origin, gives the age of the deceased as n5, and the other, whose origin
and interpretation are both uncertain, as n7. There is also a late Greek epitaph in
Rome that specifies the deceased's age as no.
88. See G. Maspero, Les contespopulaires de FEgypte ancienne) 4th ed. (Paris, 19n),
pp. 32-34.
89. Censorinus, De die natali) 17, 4. See the edition by G. Rocca-Serra (Paris,
1980), p. 23.
90. These figures were drawn from various publications of J. L. Angel. For an
overall perspective and a comparison of the situation in Greece with that among
the populations of Egypt, see J. L. Angel, ''Biological Relations of Egyptians and
Eastern Mediterranean Populations during Pre-dynastic and Dynastic Times," J.
Hum. Evol. 1 (1972):307-13.
91. Angel, People of Lema (above, n. 27), p. 85; C. S. Bartsocas, "Stature of
Greeks of the Pylos Area during the Second Millennium B.C.," Hipp. Magazine 2
(1977): 157-60; and M. J. Becker, "Human Skeletal Remains from Kato Zakro,"
Amer.]. Archaeol. 79 (1975):271-76.
93. According to the anthropologist Robert P. Charles, "Greek artists had
conceived a plastic ideal either from pure aesthetic rules or under the spell of artistic
canons in vogue in other schools or according to models taken from outside the
ethnic type most common among themselves" (Charles, "Argos" (above, n. 58),
pp. 274ff.). For the anthropometric traits of Greek statues, see M. Clementelli and
A. Jalongo, "La formula corporea nella scultura dell'antica Grecia," Riv. Anat.
Artistica (Rome) 1, nos. 3-6 (1967): 43-54.
386 NOTES TO PAGES no-15
94. For an overview, see S. and J. Brommer, Die Erniihrung der Griecher und
Romer (Munich, 1943); D. R. and P. Brothwell, eds., HJodin Antiqui-ty (London,
1969); and P. J. Ucko and G. W. Dimbleby, The Domestication and Exploitation of
Plants and Animals (London, 1969). As for the paleopathological aspects of nutri-
tion, there is an excellent update and critique by C. Wells, "Prehistoric and
Historical Changes in Nutritional Diseases and Associated Conditions," Progressin
Iood and Nu-trition Science 1, no. 11(1975):729-79.
95. Sahlins, Stone Age Economics (above, n. 3), provides convincing arguments
in this regard.
96. Cf. Cohen, rood Crisis (above, n. 3).
97. At least that is the opinion of most of those who have recently discussed
the question. See Brothwell and Brothwell, Iood in Antiqui-ty (above, no. 94); the
section by J. Yudkin and N. A. Barnicot in Ucko and Dimbleby, Domestication and
Exploitation (above, n. 94); and W. A. Stine, "Evolutionary Implication of Chang-
ing Nutritional Patterns in Human Populations," American Anthrop. 73 (1971):
1019-30.
98. Iliad 19.160-66. See M. Primiero, "L'alimentazione nei poemi omerici,"
lliv. Stor. Med. 17 (1973):17-24.
99. Plato, Republic) 372a-c, trans. P. Shorey.
100. Republic) 372c-d. The Hippocratic corpus, in particular the treatise Diet)
contains precious information on the nutritional hygiene of Greeks during the
classical period.
101. Republic) 405c-d.
102. See P. T. Makler, "New Information on Nutrition in Ancient Greece,"
IClio 62 (1980): 317-19.
103. R. Van Reen, ed., Idiopathic Urinary Bladder Stone Disease (Washington,
D.C., 1977).
104. Cf. E. L. Prien, "The Riddle of Urinary Stone Disease," J. Amer. Med.
Assoc. 216 (1971):503-7; Van Reen, Idiopathic Bladder Stone Disease (above, n. 103),
and Makler, "New Information" (above, n. 102), p. 319.
105. Airs) 9; Morb") IV, 55;Progn.) 19; Epid.) VI, 3, 7; Aph.) III, 26. See E. Lesky,
"Zur Lithiasis-Beschreibung in Peri aer6n," Wien. Stud. 63 (1948): 69-83, and "Die
Lithiasis im Altertum, eine ausgesprochene Kinderkrankheit, '' Zschr. I(inderheilk.
(1948), pp. 250-59.
106. Morb") IV, 55 (Littre, II, 600).
107. A. Jarde, Les cerealesdans l)Antiquitegrecque (Paris, 1925),p. 142.
108. See H. Brabant, "La denture humaine a l'epoque neolithique," Bull. Soc.
Roy.Be!!J.Anthrop. 79 (1968): 105-41, and Wells, "Prehistoric Changes" (above, n.
94), pp. 741-44.
109. Becker, "Skeletal Remains" (above, n. 91), pp. 271-76.
no. Charles, "Argos" (above, n. 58), pp. 278-80. ·
111. Angel, People of Lema (above, n. 27), p. 90.
112. Angel, "(_;atal Hiiyiik" (above, n. 50), p. 90.
113. Angel, People of Lema (above, n. 27), p. 90.
114. Other factors can intervene (sexual differentiation in eating habits or in tasks
involving chewing) to invert the usual contrast: for instance, at (_;atalHiiyiik strong
dental abrasion occurs in 31.4 percent of the women and only 29 percent of the men.
115. Strong dental abrasion on teeth from the Greek necropolis of Pithekussai on
the island of Ischia in Italy actually begins in childhood (eighth and seventh cen-
turies B.c.). See F. R. Munz, "Die Zahnfunde aus der griechischen Nekropole
von Pithekussai auf Ischia," Archaeol. Anz. (1970), pp. 452-75. It is worth noting
that infant mortality there was especially high and average longevity low, judging
in particular from skeletons of this colony belonging to the eighth-century B.C.

stratum.
n6. A. J. Clement, "Caries in the South African Ape-man," Brit. Dent. ]our.
ror (1956): 4-7. On tooth decay in the prehistoric populations of Homo sapiens) sec
H. V. Vallois, "La carie dentaire et la chronologie des hommes prehistoriques,"
Anthropologie 46 (1936): 2orff.
n7. See H. Euler, Die Zahncaries im Lichte v01;geschichtlicherund geschichtlicher
Studien (Munich, 1939), and C. Wells, "Prehistoric Changes" (above, n. 94),
pp. 730-38.
n8. See J. L. Angel, "Greek Teeth, Ancient and Modern," Human Biol. r6
(1944): 283-97, and People of Lema (above, n. 27), p. 90; and C. Wells, "Prehistoric
Changes" (above, n. 94), pp. 732-34. According to Angel's initial research, the
frequency of decayed teeth i.n the Greek Neolithic reached 12 percent, but the
figure must be lowered as a result of subsequent studies of teeth at Nea Nikomedia
and <_;atalHiiyiik.
n9. Angel, "Mycenae" (above, n. 57), p. 307, and People of Lema (above, n. 27),
p. 90.
120. C. M. Furst, Zur Anthropologie der pri:ihistorischen Griechen in A1lJolis (Lund,
1930), p. 122, and Charles, "Argos" (above, n. 58), p. 309.
121. H. G. Carr, ''Some Dental Characteristics of the Middle Minoans,'' Man
60 (1960): n9-22.
122. Seen. n8.
123. E. Nyqvist, "Human Teeth from Kition," Opuscula Atheniensia 13 (1980):
185-88. See also P. M. Fischer, ibid., pp. 139-48.
124. E. Benassi and A. Toti, "Osservazioni sulle ossa rinvenute negli scavi della
necropoli di Spina," Atti Mem. Accad. Stor. Arte San. II, 24 (1958): 16-28; and
Munz, "Die Zahnfunde" (above, n. n5).
125. Pseudo-Aristotle, R-oblemata) XXII, 14, 93ra. Hippocrates blamed certain
kinds of nutritional debris for causing tooth decay; see G. Cootjans, "Le probleme
etiologique de la carie dentaire clans l'Antiquite," Revue Be{ge Stomat. 52 (1955):
677-88.
126. D. M. Hadjimarkos and C. W. Bonhorst, "Fluoride and Selenium Levels
in Contemporary and Ancient Greek Teeth in Relation to Dental Caries," Nature
(London) 193, no. 48n (r962):r77ff.
127. H. Duday, "La pathologie dentaire des hommes de Kitsos," in N. Lam-
bert, ed., La grotte prehistorique de I(itsos (Paris, 1981).
Chapter RJur: Common Purulent Inflammations

r. For an overview of the history of the concept of infection, see especially
T. Puschmann, Die Geschichte der Lehre von der Ansteckung (Vienna, 1895); 0.
Temkin, "An Historical Analysis of the Concept of Infection," in Studies in Intel-
lectual History (Baltimore, 1953), pp. 123-47; and M. D. Grmek, "Le concept d'in-
fection clans l' Antiquite et au Moyen Age," Rad. Jug.Akad. (Zagreb) 384 (1980):
9-55. For the history of ideas concerning the infection of wounds, see G. Majno,
The Healing Hand: Man and Wound in the Ancient World (Cambridge, Mass., 1974).
The writers of the Hippocratic corpus were not ignorant of the clinical manifesta-
tions of the suppurating infection of wounds and excelled in the description of
diseases that nowadays are considered infectious. There are medical-historical
publications on this aspect of the Hippocratic texts, for instance M. Soulangas,
Etude sur Hippocrate) son oeuvre) ses idies sur />infectionet les moyens antiseptiques (Paris,
1894); G. Sticker, "Fieber und Entziindung bei den Hippokratikern," Arch. Gesch.
Med. 22 (1929): 313-43, 361-81, and ibid. 23 (1930): 21-25, 92-100; and R. E. Siegel,
"Epidemics and Infectious Diseases at the Time of Hippocrates," Gesnerus 17
(1960): 77-98. However, the idea of the propagation of diseases by contagion is
missing from the Hippocratic texts, and clinical phenomena of this kind are ex-
plained therein by an appeal to endogenous factors or the special conditions in a
given environment.
2. The double staining method invented by Hans Gram makes it possible to
divide bacteria into two groups: gram-positive ones that retain the first (violet)
dye, and gram-negative ones that, when discolored by alcohol, retain the second
(red) dye.
3. See J. W. Schopf et al., "Electron Microscopy of Fossil Bacteria Two Billion
Years Old," Science149, (1965): 1365-67; E. S. Barghoorn, "The Oldest Fossils,"
Scient. Amer. 224 (1971):30-42; J. W. Schopf, "L'ere de la vie microscopique,"
Endeavour (French-language edition) 34, no. 122 (1975):51-58.
4. R. L. Moodie, Palaeopathology(Urbana, 1923); for the critique of his inter-
pretation, see Majno, The Healing Hand (above, n. 1), pp. 16 and 473.
5. Ju-kang Woo, "Mandible of Sinanthropus Lantianiensis," Curr. Anthrop. 5
(1964): 98-101.
6. See J. Enselme, Mort et maladies a Paube de Phumanite (Lyon, 1973),
pp. 144ff.
7. L. Pales, Paliopathologieet pathologie comparative (Paris, 1930), pp. 187-90;
H. E. Sigerist, Historyof Medicine) vol. 1, "frimitiveand Archaic Medicine (New York,
1951),p. 49; C. Wells, Bones)Bodiesand Disease(London, 1964), pp. 76ff.
8. See R. Hare, "The Antiquity of Diseases Caused by Bacteria and Viruses,"
in D. Broth well and A. T. Sandison, Diseases in Antiquity (Springfield, 1967),
pp. n5-31 (esp. p. 123).
9. M. A. Ruffer, "Remarks on the Histology and Pathological Anatomy of
Egyptian Mummies," Cairo Scient.]. 4 (1910):3-7.
10. J. L. Angel, The People of Lema (Princeton and Washington, D.C., 1971),
pp. 50 and 91-92, pl. 24.
n. R. P. Charles, "Etude anthropologique des necropoles d 'Argos," Bull Corr.
Hell. 82 (1958):310,and Etudes anthropologiques desnecropolesd)Argos (Paris, 1963),p. 66.
12. Cited from D. G. Rokhlin, Bolezni drevnih lyudei (Moscow and Leningrad,
1965),p. 93.
13. Angel, Peopleof Lema (above, n. 10), pp. 55and 84.
14. The archaeologist Alan J. B. Wace first pointed out the case of this Myce-
naean aristocrat in 1949, suggesting that he had "suffered from a septic wound of
the jaw" (A.J.B. Wace, Mycenae: An ArchaeologicalHistory and Guide [Princeton,
1949], p. n7). Purulent inflammations on Mycenaean jawbones were also reported
~y C. M. Furst, Zur AnthropologiederprahistorischenGriechenin ArgolisJLunds Univ.
ArsskriftJ n.s., sect. 2, 26, no. 8 (1930):122. A detailed description of the patholog-
ical state of the jayvbone of 59 Myc. was given by Angel in G. E. Mylonas, 0 Tafik6s
J(uklosB tou Mylienou (Athens, 1973),pp. 381ff.,pl. 245.
15. See J. L. Angel, "Appendix on the Early Helladic Skulls from Aghios
Kosmas," in G. E. Mylonas, Aghios I(osmas (Princeton, 1959),p. 169; Angel, People
of Lema (above, n. 10), pp. 39, 42, 50, 52, 84, and so forth, and pl. 22; J. Dastugue
and H. Du day, "Les ossemen ts h umains pathologiques," in N. Lambert, ed., La
grotte prehistorique de I(itsos (Paris, 1981); Charles, Etudes anthropologiques(above,
n. n), p. 66.
16. H. G. Carr, "Some Dental Characteristics of the Middle Minoans," Man
60 (1960): pp. n9-22.
17. J. L. Angel, "Skeletal Material from Attica," Hesperia14 (1945):279-363.
f\
18. For the problem of, the antiquity of gonorrhea, see below, chapter 5, the
section "Gonorrhea~ Spern1atorrhea, and Leukorrhea."
19. P. Chantraine, Dictionnaire ltymologiquede la languegrecque(Paris, 1974) 3: 952.
20. Aristotle, Degen. anim-; IV, 8 (777a7).
2r. Cf. Plato, Phaedo) 966, and Pseudo-Galen, Def. medicae)99 (Kuhn, x1x, 373).
22. J. Bollack, Empedocle (Paris, 1969), vol. 3, 2, p. 539. The false translation is
widespread, since it was recommended by H. Diels and W. Kranz, Die Fragmente
der Vorsokratiker)6th ed. (Berlin, 1951),1: 337, fr. 68. In order to sustain it, Kranz
had to emend the wordgala 'milk' in the manuscripts of Aristotle and replace it
with the word hazma 'blood,' an arbitrary change in the sense of Aristotle's
statements.
23. Aristotle, foe. cit. (above, n. 20).
24. For the "putrefaction" of the humors in an organism as a pathogenic factor,
see especially R. Wittern, Die hfppokratischeSchrift De Morbis I (Hildesheim, 1974),
pp. 18, 28, 32, 36, 38, 48, 52, 80, 82, 90, and so forth (for the Hippocratic text), and
p. 195 (for Wittern's remarks on the meaning of this family of words). See also
lton-h-; I, 99. For Hippocratic seps£sas the physiological process of digestion, cf.
Aff-; 84; Viet.) III, 80; Anat.) I. See A. Foesius, OeconomiaHippocratis (Frankfurt,
1588),p. 562. P. Chantraine, Dictionnaire ltymologique(above, n. 19), p. 592, is surely
correct in relating the substantives of the piton family to the verb pftthomai 'rot,'
but from the viewpoint of the history of ideas it is an error to translate, as he does,
the Hippocratic expressions sarkopuonas "infected flesh" and empuosas "infected."
The correct translation is "purulent," or even "putrid," but not "infected,"
since the Hippocratic concept (or even the Aristotelian one) of pus has no connec-
tion with the idea of infection either in the sense of defilement or in that of the
transmission of an agent.
25. For example, Morb-; 1, 19 and 29; Vet. med-; 18; Aph.) II, 40; Acut.; 38. See
Foesius, Oeconomia(above, n. 24), pp. 492 and 504.
26. Epid-; III, 4. I adopt the text of Jones (1, 242), not Littre.
27. De al£mento) 52 (Joly, 147).
28. See Foesius, Oeconomia (above, n. 24), p. 544, and M. P. Dumini, Le sang)
lesvaisseaux) le coeurdans la Collectionhippocratique)thesis (Paris, 1983).
29. De capitis vulneribus) 14 (Littre, III, 20; Petrequin, 1, 474-75).
30. The correctness of this Hippocratic observation is stressed by E. D. Chur-
chill, "Healing by First Intention and with Suppuration: Studies in Wound Heal-
ing," J. Hist. Med. 19 (1964): 193-214 (esp. p. 199).
3r. See, for instance, Epid.) VI, 5, 6: "In wounds, the blood flows in."
32. Ulc-;26 (Littre VI, 430; Petrequin, 1, 310-n).
33. Morb-; IV, 50, 4 (ed. R. Joly [Paris, 1970], p. 107).
34. Fract-; 31(Petrequin, 206-7).
35. See F. Kudlien, Der Beginn des medizinischen Denkens bei den Griechen (Zurich
and Stuttgart, 1967), pp. 49 and 72ff.
36. Majno, The Healing Hand (above, n. 1), p. 183.
37. For instance Aph-; VI, 20; Flat. (Littre, VI, 106 and elsewhere).
38. In the Hippocratic texts, the term spasmo[does not necessarily signify con-
tractions. It also designates extensions; the word refers to a state of tension. See
Wittern, Die hippokratischeSchrift (above, n. 24), p. 197.
39. Morb°) I, 14 (Littre, VI, 162; Wittern, 32) and 17 (Littre, VI, 170; Wittern, 44).
40. Morb.) 1, 15(Littre, v1, 164; Wittern, 36).
4r. Cf. ltogn-; 7, 17, 18, and 22; Aph.; VII, 44, 45; ltaen. Coacae)403 and 404.
42. Epid.) v1, 3, 4 (Littre, v, 294). My translation is perceptibly different fron1
the one Littre suggests.
43. Herodotus, Histories) III, 64-66, trans. A. de Selincourt (Baltimore, 1954;
rept. 1966), pp. 201-3 (replacing, as translations of esape)the words "gangrene" and
"mortification" by "decay" and "putrefaction").
44. Herdotus, Histories) VI, 136 (again with the words "gangrene" and "morti-
fication" replaced by words signifying putrefaction).
45. Herodotus, Histories) VI, 134, my translation. The verb spasthenaihere must
mean "break" or "fracture," not "twist" or "dislocate" as others translate it.
See L. W. Daly, "Miltiades, Aratus, and Compound Fractures," Amer.]. Philo!.
101 (1980): 59ff., who refers to the case of Aratus of Sicyon, a Greek general of the
third century B.C. who broke his leg, underwent several incisions, and for a long
time carried out his military duties from a litter (cf. Plutarch, Vita Arati) 33).
46. For the local and general symptoms of wounds, see for instance Morb°) rv,
48 (Joly, 103-4).
47. The Greek word is spasm6s'tetanus.'
48. Artie°) 63 (Petrequin, II, 472-75).
49. Moch!.) 30 (Petrequin, II, 594-97).
50. For instance Epid°)V, 26, 45, and 65 (Littre, V, 224, 234, and 244); Epid°)VII,
61 (Littre, v, 426). See Majno, The Healing Hand (above, n. 1), pp. 196-98.
51. Artie.) 55 (Petrequin, II, 446-47), trans. E. T. Withington, in Hippocrates)
vol. 3 (London 1928), p. 329. Cf. also Moch!°) 21.
52. See K. E. Muller, Die Geschichteder entziindlichen J(nochenerkrankungenund
ihrer chiru1lJischenBehandlung (Diisseldorf, 1938).
53. For a general description of this disease, see Morb.) II, 7 and 24 (Littre, VII,
14-16 and 38), and for concrete cases, for example, see Epid.) v, 97, and VII, 35
(Littre, v, 256 and 402). In the citation from the Hippocratic book Diseases)this
form of bone caries is designated by the word terediJn)whose primary application is
to a wood borer's drilling and piercing. Is this technical term based simply on a
morphological analogy between decayed bone and worm-eaten wood, or does it
represent the traces of a very old etiological hypothesis~
54. Epid°)I, 13, 9 (Littre, II, 704).
55. Epid.) v, 45 (Littre, v, 234).
56. Cf. Praen. Coacae) 196; Progn.) 23; Aph.) v, 23; Epid°) III, 3-4. See E. W.
Goodall, "Infectious Diseases and Epidemiology in the Hippocratic Collection,"
Proc.Royal Soc. Med. 27 (1934): 526.
57. Cases of gas gangrene are described in one of the katasttiseisof Thasos (Epid°)
III, 4 [Littre, v, 70-76]). See C. E. Kellett, "The Early History of Gas Gangrene,"
Ann. Med. Hist. 1 (1939): 452-59; and M. Sussmann, "A Description ofClostridium
Histolyticum: Gas Gangrene in the EpidemicsofHippocrates," Med. Hist. 2 (1958):
226ff.
58. See below, chapter 13, the section "Four Concise Case Histories," for a
discussion of tetanus in Epid°)VII, 36.
59. For instance Epid°)II, 1, 1, and perhaps Epid°)III, 7. See B. Ebbell, Beitriigezur
ttltestenGeschichteeinigerInfektions-krankheiten (Oslo, 1967), p. 44.
60. Epid°) v, 61 (Littre, v, 240-42). In a parallel description of the same case
(Epid°)VII, 33 [Littre, v, 402]), there is the additional detail that the patient's eyes
were greenish. For the historical circumstances of this wound, see F. Robert, "La
bataille de Delos (Hippocrate, Epideinies, v, 61 and VII, 33)," Bull. Corresp.Hell.
Suppl.) I, Etudes diliennes (1973), pp. 427-33. Other cases of peritonitis are described
in Epid.) v, 98-99, and Epid.) VII, 29-30. The presence of ileus (severe intestinal
blockage) is a noteworthy sign of traumatic peritonitis. For a medical commentary,
see Majno, The Healing Hand (above, n. 1), p. 193.
61. For the first form of pharyngitis (kunangkhe), see Morb.) II, n and 30 (Littre,
VII, 18, and 48); for the second, Morb.) II, 9 and 28 (Littre, VII, 16 and 46); and for
Ludwig's angina (hupogwssis) see especially Morb.) II, 31 (Littre, VII, 48). Specific
cases are described in Epid.) I, 13, case 5; and Epid? III, 1, 2, cases 6 and 7.
62. Cf. Morb.) III, 2 (Littre, VII, 120; Potter, 70-72). For other passages on this
disease, see A. Courtade, "L'otologie clans Hippocrate" (Paris, 1904) (from the
Arch. Intern. Laryng.), pp. 4-n.
63. Epid? VII, 5 (Littre, v, 373-76).
64. See especially Morb.) II and III. There are some useful remarks in A. Souques,
Etapes de la neurologiedans PAntiquitigrecque (Paris, 1936), pp. 72-79.
65. See E. Bazin, La pneumonie) la pleuresie et la phtisie chez les medecins de la piriode
greco-romaine (Paris, 1891),and A. Souques, "La douleur clans les livres hippocra-
tiques," Bull. Soc. Franq. Hist. Mid. 32 : (1938):178-86.
66. Morb? III, 15(Littre, VII, 136-38;Potter, 82-85). See the medical commentary
in P. Potter's edition (Berlin.5 1980), pp. n8-19. For other descriptions of this
disease, see for example Morb? II, 47 (Jouanna, 178-83), and Acut. (sp.), 31 (Littre,
II, 456-64; Joly, 83-84). The excellent study by G. Preiser should also be consulted:
ccPeripleumoniein den Schriften der knidischen Aerzteschule," in Medizingeschichte
in unserer Zeit (Stuttgart, 1971),pp. 31-35.
67. See for instance Morb? III, 16 (Littre, VII, 142-56; Potter, 86-96). A general
historical sketch is given by A. Baffoni, Storia delle pleuriti da Ippocrate a Laennec
(Rome, 1947).
68. Morb? II, 59 (Littre, VII, 92; Jouanna, 198ff.)
69. See Barbillon, "L'empyeme hippocratique," in Etudes critiques d)histoire de
la medicine (Paris, 1930), pp. 36-43; A. Souques, "La pleuresie et l'empyeme hip-
pocratiques," Itesse Mid. (1938),pp. 425-27; and R. E. Siegel, "Clinical Observa-
tion in Hippocrates," J. Mount Sinai Hosp. 31(1964): 295-97.
70. An excellent description in Aff int.) 14-17 (Littre, VII, 202-10). See
E. Desnos, Histoire de l)urologie (Paris, 1914); E. D. Baumann, "Ueber die Erkran-
kungen der Nieren und Harnblase im klassischen Altertum," Janus 37 (1933):34-
47, 65-83, n6-21, and 145-52; and M. D. Grmek and R. Wittern, "Die Krankeit
des attischen Strategen Nikias und die Nierenleiden im Corpus Hippocraticum,"
Arch. Intern. Hist. Sci. 26 (1977):3-32.
71. See H. Fasbinder, Entwicklungslehre) Geburtshulfe und Gyniikologie in den hip-
pokratischen Schriften (Stuttgart, 1897), and P. Diepgen, Die Frauenheilkunde der
Alten Welt (Munich, 1937),pp. 187ff.and 212-25.
72. Epid? I, cases 4, 5, and n; Epid.) III, 2d list, cases 2 and 14. See also Mu!.) I,
35-41.
Chapter Five: Syphilis

r. For a recent bibliography of the state of the question, see F. Guerra, "The
Dispute over Syphilis: Europe versus America," Clio Medica 13 (1978):39-61. This
author favors the autochthonous origins of European syphilis in the fifteenth
century. The arguments of the "Americanists" are well condensed by F. E. Rabello,
"Les origines de la syphilis," Nouvelle Itesse Medicate 2 (1973):1376-80. He stresses
the "sharp contrast between [anti-Americanist] hypotheses, which are at times
anecdotal and cute, and the impressive weight of facts that support the nonexis-
tence of syphilis [in Europe] before Columbus."
2. Especially important in this regard are J. Rosenbaum, Geschichte der Lust-
seuche im Alterthume (Halle, 1839),and J. K. Proksch, Die Geschichte der venerischen
I(rankheiten (Bonn, 1895).Both these monographs are remarkable for the historical
documentation assembled in them, but they must be used with the greatest care,
since their interpretation of those sources is not critical enough and no longer
corresponds to modern medical knowledge.
3. For a critique of the so-called classical descriptions of this disease, see I.
Bloch, Der Ursprung der Syphilis) 2 vols. (Jena, 1901-n), and E. Jeanselme, Histoire
de la syphilis (Paris, 1931).Even Socrates' snub nose on busts has been considered a
proof of the existence of syphilis in his day .. In this regard, see B. Springer, Die
genialen Syphilitiker) 2d ed. (Berlin, 1926), p. 61; G. Milian, "Socrate heredo-syphi-
litique," Paris med°)suppl., 14 (1913):597-603; and especially the critical review of
W. Fahlbusch, "War Sokrates mit angeborener Syphilis behaftet?" Denn. Woch-
enschr.107 (1938):1067-79. The snub nose can be a simple anatomical variation (that
is probably the case for Socrates' nose, which, though it lacked the beauty of
Cleopatra's, was not necessarily the sign of a disease), but it can also result from a
pathological collapse of the nose. An event of this sort is actually described in the
Hippocratic texts (Epid.) IV, 19, and Epid°)VI, 1, 3; Littre, v, 156and 266). It is true
that nowadays syphilis, both congenital and tertiary, is the most common cause of
this type of nasal deformity, but that is not sufficient reason for us to follow the
authors who interpret the central collapse of the nose as a pathognomonic sign of
syphilis and who cite the Hippocratic corpus as evidence for the antiquity of the
disease (for instance, J. Wright, The Nose and Throat in Medical History [Philadel-
phia, n.d.], p. 45, and R. Kapferer, Hippokrates-Fibel[Stuttgart, 1943], p. 127). The
syphilitic snub nose is chiefly caused by the destruction of the nasal septum by
vascular disturbances, while the Hippocratic text speaks of a consuming ulceration
that causes the collapse of the nose in a child by ''the exit of a bone from the roof
of the mouth.'' One can suppose in this case a perforation of the hard palate by
tertiary syphilitic inflammation, but the observations of the ancient Greek physi-
cian do not lend themselves to this diagnosis, since the ulceration in question is so
extensive that it makes the upper incisors fall out and even some lower teeth. This
is noma (gangrenous stomatitis), not syphilis (see chapter 13, below, the section
"Four Concise Case Histories," for a discussion of noma in Epid°) VII, n3). A
relatively recent paper suggests that the final illness of the Roman general Sulla was
syphilis (see T. F. Carney, "The Death of Sulla," Acta Classica4 [1961]:64-79);
the identification of this case of ''phthiriasis'' with syphilis rests on an argument
that is medically unacceptable.
4. See T. B. Turner and D. H. Hollander, Biologyof the Treponematoses (Geneva,
1957).
5. E. H. Hudson, Non-venerealSyphilis:A Sociological and Medical Study of Bejel
(Edinburgh and London, 1958); "Treponematosis and Man's Social Evolution,"
Amer. Anthrop. 67 (1965): 885-901; "Christopher Columbus and the History of
Syphilis," Acta Tropica25 (1968): 1-16.
6. C. J. Hackett, "On the Origin of the Human Treponematoses," Bulletin
WH.O. 29 (1963): 7-41; "The Human Treponematoses;" in D. Brothwell and
T. A. Sandison, Diseasesin Antiqui-ty (Springfield, 1967), pp. 152-69.
7. Hackett, "Human Treponematoses" (above, n. 6), pp. 161and 165.
8. D. R. Brothwell, "The Real History of Syphilis," ScienceJournal 6 (1970):
27-33.
9. T. A. Cockburn, "The Origin of the Treponematoses," Bulletin WH.O. 24
(1961): 221-28, and The Evolution and Eradication of Infectious Diseases(Baltimore,
1963), pp. 152-74.
10. J. D. Oriel and T. A. Cockburn, "Syphilis: Where Did It Come From?"
PaleopathologyNewsletter)no. 6 (1974): 9-12.
n. See H. C. Raven, The Anatomy of the Gorilla (New York, 1950); A. Fribourg-
Blanc and H. H. Mollaret, "Natural Treponematoses of the African Climate,"
Primates in Medicine 3 (1968): 110-18. The first author states the existence of a
syphiloid cutaneousddisease in apes, and the two others have detected anti-trepo-
nematic antibodies in the blood and treponemes living in the popliteal lyn1ph
nodes of African apes.
12. V. M9)ller-Christensen, "Evidence of Tuberculosis, Leprosy, and Syphilis in
Antiquity and the Middle Ages," J+oc.XIX Intern. Congr. Hist. Med. (Basel) 1964)
(Basel, 1966); "Venerische und nichtvenerische Syphilis," in Medizinische Diagnostik
in Geschichteund Gegenwart (Festschrift fiir H. Goerke) (Munich, 1978), pp. 226-34.
13. For the criteria that should be the basis of the paleopathological diagnosis of
treponematoses, see especially C. J. Hackett, "An Introduction to Diagnostic
Criteria of Syphilis, Treponarid, and Yaws (Treponematoses) in Dry Bones, and
Some Implications," Virchow)s Archiv fur Pathol. Anatomie und Histowgie 368 (1975):
229-41, and "Diagnostic Criteria of Syphilis, Yaws, and Treponarid (Treponema-
toses) and of Some Other Diseases in Dry Bones,'' Sitzungsbericht. Heidelberger
Akad. Wiss.) Med.-nat. I(l. (1976), pp. 339-470.
14. Hudson, "Treponen1atosis," (above, n. 5), p. 899.
15. See H. U. Williams, "The Origin and Antiquity of Syphilis," Archives of
Pathowgy 13 (1932): 779-814 and 931-83; S. Jarcho, "Some Observations on Disease
in Prehistoric North America," Bull. Hist. Med. 38 (1964): 1-19; C. W. Goff,
"Syphilis," in Brothwell and Sandison, Diseases (above, n. 6), pp. 279-94; N. G.
Gejvall and F. Henschen, '' Anatomical Evidence of pre-Columbian Syphilis in the
West Indian Islands," J. Occup. Therap. 25 (1971): 138-57; M. Y. El-Najjar, "Human
Treponematosis and Tuberculosis: Evidence from the New World," Amer.]. Phys.
Anthr. 51 (1979): 599-618. In several cases, carbon 14 dating makes it certain that the
bones in question are really pre-Columbian (twelfth to fourteenth century); cf. D.
Broth well and R. Burleigh, "Radiocarbon Dating and the History of Treponema-
toses in Man,"]. Archaeol. Sci. 2 (1975): 393-96.
16. T. D. Stewart and O. Spoehr, "Evidence on the Paleopathology of Yaws,"
Bull. Hist. Med. 26 (1952): 538-53.
17. See Hackett, "Diagnostic Criteria" (above, n. 13), pp. 230 and 238.
18. For a critical evaluation of the old research on this subject, see H. E. Sigerist,
History of Medicine (New York, 1951), 1: 54-56. The medieval cases from Siberia
described by D. G. Rokhlin and preserved in his museum in Leningrad have been
discredited by recent examinations (performed by V. M9)ller-Christensen and by
me). In the case from Spitalfields (London) published by D. Brothwell (Digging
Up Bones [London, 1972], p. 137, pl. 5), the diagnosis is certain but the date of the
specimen is not. There has been a diagnosis of "hereditary syphilis" on Neander-
thal remains deposited in the British Museum (cf. D. J. Wright, "Syphilis and
Neanderthal Man," Nature [London] 229 [1971]: 409); the diagnosis is a fantasy
based on dubious criteria. An extraordinary case was presented by Eliane Spitery
at the Third European Congress of the Paleopathological Association held at Caen
in 1980: in question is a male skull from the private collection of S. Gagniere
(Marseille) with the typical signs of gangosa (rhino-pharyngitis mutilans). The
diagnosis of tertiary yaws seems most likely, but the history of this museum speci-
men is troubling. It seems to have been found in a Roman cemetery at Ades in
1839. Precise documentation of the location and conditions of the find are lacking.
It might be the skull of a Roman soldier, as an old label has it, or it might be a
later intrusion. An African origin cannot be ruled out, despite the absence of
Negroid anthropometric traits. It would be very useful to submit this skull to
modern dating methods, particularly carbon 14 dating. The diagnosis of bejel
proposed by F. Kail and A. de Froe (Hantarzt 4 [1953]: 82) for a sixth-century
skeleton from Iraq seems to me uncertain and unlikely.
19. See in this connection the studies by L. Gluck and J. Fleger. A survey of
syphilitic endemics in modern times can be found in G. Solente, "Les principales
endemies de syphilis," ltesse Medicale 69 (1961): 2363-65.
20. E. H. Ackerknecht, Geschichte und Geographieder wichtigsten J(rankheiten
(Stuttgart, 1963), p. 107, and F. Henschen, The History and Geographyof Diseases,
trans. J. Tute (New York, 1966), p. 87.
21. Rosenbaum, Geschichteder Lustseuche (above, n. 2), translated into English
from the French version, Histoire de la syphilis dans FAntiquite (Brussels, 1847),
pp. 37-38. This work enjoyed great success before the public (there were seven
German editions and translations into several languages), and it was accepted with
enthusiasm by specialists. Karl Sudhoff, the leading historian of medicine at the
beginning of the twentieth century, wrote an encomiastic review of it and dubbed
it "unsterblich," immortal! See H. T. Koch, "Julius Rosenbaum (1807-1874) als
Medizinhistoriker," NTM (Leipzig) 18 (1981): 84-90.
22. Ibid., pp. 181-84, 196-98, and elsewhere.
23. Ackerknecht, Geschichteund Geographie(above, n. 20), p. 107; M. L. Brodny,
''The History of Gonorrhea among the Greeks and Romans,'' Trans. Amer. Neisser.
Med. Soc. 3 (1937): 92-106. In the cuneiform texts of Mesopotamia, there is talk of
venereal diseases. Though syphilis cannot be identified, gonorrhea seems probable.
On this subject, see R. Labat, "Geschlechtskrankheiten," in &a/lexicon der Assy-
riologie(Berlin, 1971), 3: 221-23.
24. Henschen, History and Geographyof Diseases(above, n. 20), p. 87.
25. Pazzini, "De amatorum morbis," Boll. Istit. Stor. Ital. Arte San. 10 (1930):
1-18 and 131-54, and "Ancora sulla blenorragia degli antichi," Atti Mem. Stor. Arte
San. (1983), pp. 85-90.
26. H. St. H. Vertue, "Enquiry into Venereal Disease in Greece and Rome,"
Guy)s Hosp. Rep. 102 (1953): 277-302, and J. D. Oriel, "Gonorrhea in the Ancient
World," Paleopath. Newsletter) no. 4 (1973).
27. The basic reference is Leviticus 22:4 in the Septuagint. The medical texts
usually cited for the ancient existence of the gonococcus are as follows: Hippo-
crates, Epid.) III, 7; Morb._,II, 51; Aph.) IV, 82; and Mul._, I, 24, and II, n6; Celsus,
De med.) VI, 18; Aretaeus, De diut. morb._,II, 5, and De cur. diut. morb._,II, 5; Galen,
De locisaffectis) VI.
28. Oriel, "Gonorrhea" (above, n. 26).
29. For instance, Epid")VI, 8, 29; Aff int.) 43; Morb._,II, 51; and Genit.) 1.
30. Morb") II, 51 (Littre, VII, 78-80; Jouanna, 188).
31. See M. D. Grmek and R. Wittern, "Die Krankheit des attischen Strategen
Nikias und die Nierenleiden im Corpus Hippocraticum," Arch. Intern. Hist. Sci.
26 (1977): 3-32.
32. Aph._, IV, 82. See also Aph.) VII, 57, and ltaen. Coacae)463.
33. See below, chapter n, "The Place and Date of the Disease of Philiscus."
34. So Jones for the Greek word polla) which could also be translated, as by
Littre, ''frequent, common.''
35. Epid._,III, 7 (Littre, III, 84), trans. W.H.S. Jones, Hippocrates(Cambridge and
London, 1948), 1: 247. See also Epid.) III, 3, 3 (Littre, III, 70; Jones, I, 241).
36. J. Hirschberg, Geschichte der Augenheilkunde im Alterthum (Leipzig, 1899),
p. 70.
37. Rosenbaum, Histoire de la syphilis (above, n. 21), p. 255.
38. H. Beh~et, "Ueber rezidivierende aphthose, <lurch ein Virus verursachte
Geschwi.ire am Mund, am Augem, und an den Genitalien," Denn. Wochenschr105
(1937): 1152-57.
39. See A. Feigenbaum, "Description of Beh~et's Syndrome in the Hippocratic
Third Book of Endemic Diseases,'' Brit. ]. Ophthalm. 40 (1956): 355-57.
40. For example, see Morb') IV, 55;Epid') II, 3, 5; Epid') VI, 2, 2; Aph.) IV, 55.
41. Lysistrata) 987L88.
42. Galen, Meth. med.) VIII, 8, 6 (Kiihn, x, 580).
43. Mu!') I, 24 (Littre, VIII, 64).
44. Galen, De foe. aff) VI (Kiihn, VIII, 441; cf. also Kiihn, IX, 267). See
P. Diepgen, Die Frauenheilkunde der Alten Welt (Munich, 1937)(Steckel)sHandbuch
der Gynakologie)XII-1), p. 199, and R. E. Siegel, Galen on the Affected Parts (Basel,
1976), pp. 192ff.
45. Mu!') II, 116-20 (Littre, VIII, 250-62).
46. Mu!') II, 116(Littre, VIII, 250-52); based on Littre's translation.
47. Diepgen, Die Frauenheilkunde (above, n. 44), p. 198.
48. M.-T. Fontanille, Avortement et contraceptiondans la medicinegreco-romaine
(Paris, 1977), p. 158.
49. Aph.) III, 24 (Littre, IV, 496). For an aphthous mouth as a bad omen in
pregnant women, see H-aen. Coacae)504 and 533(Littre, v, 700 and 706). To be
sure, it is impossible for us to distinguish aphthae in the mouth that have viral
origins from those due to thrush, but the insistence in ancient sources on the
commonness of this disease in newborns points to mycosis. In my opinion, the
two diseases coexisted in antiquity. Hippocratic knowledge of thrush is generally
accepted by historians of pediatrics, who find corroboration in the writings of
Soranus, Celsus, Aretaeus, and Galen. See D. Ghinopoulo, Padiatrie in Hellas und
Rom (Jena, 1930), pp. 57ff., and A. Peiper, Chronik der J(inderheilkunde) 2d ed.
(Leipzig, 1955),p. 23.
50. In Hipp. Aph. comm') III, 25 (Kiihn, XVII, B 627). See also Def. med') Kiihn,
XIX, 441, and the lexicons of Hesychius and Erotianus. Etymologically linked to
the verb htipto 'enflame, join, seize,' the term aphtha does not apply to deep,
stubbor~ ulcerations (whence the distinction, in Epid.) III, 7, between aphth&dea
and helkodea in the mouth) nor to blistering ulcerations (whence the distinction
between them and phluktainai 'blisters' in the ancient gynecological texts).
51. Nat. mu!') 60, 86, and 100 (Littre, VII, 398, 408, and 416); Mu!.) I, 34 (Littre,
VIII, 82); H-aen. Coacae)518(Littre, v, 704).
52. Morb') II, 50 (Littre, VII, 76).
53. See J. Jouanna's note in his edition: Hippocrate, Maladies II (Paris, 1983),
p. 254.
54. H-aen. Coacae)518(Littre, V, 704).
55. See below, chapter 7.
56. In the Hippocratic corpus, the term herpesdesignates some serious dermato-
ses that recall herpes zoster (shingles), herpes circinatus, or even perhaps the erup-
tions of smallpox, as the famous dermatologist Robert Willan thought. In my
opinion, we should think of them as indicating a cutaneous lesion having a certain
appearance, a kind of serpiginous ulcer, rather than a disease proper. None of the
uses of this word in the medical literature of the classical period refers to genital
herpes in the modern sense. See T.S.L. Beswick, "The Origin and Use of the
Word Herpes," Med. Hist. 6 (1962): 214-32.
57. Nat. mu!.) 108 (Littre, VII, 422). See also Mu!.) I, 90 (Littre, VIII, 214-18).
58. See especially Mu!.) II, 173(Littre, VIII, 354).
59. Epid') VI, 8, 21 (Littre, V, 352).
60. Mu!') I, 40 (Littre, VIII, 96-98).
61. Ulc') 14 (Littre, VI, 418; Petrequin, I, 292).
62. Celsus, De medicina) v, 28, 14B, trans. \V. G. Spencer (London and Cam-
bridge, 1983;rept. 1953),2: 161.
63. See J. D. Oriel, "Anal and Genital Warts in the Ancient World," Paleopath.
Newsletter) no. 3 (1973).
64. Nat. mul.J 66 and 83 (Littre, VII, 402 and 406). See also Galen, Meth. med.J
v (Kuhn, X, 325).
65. Diepgen, Die Frauenheilkunde (above, n. 44), p. 217, and H.-J. von
Schumann, Sexualkunde und Sexualmedizin in der klassischen Antike (Munich, 1975),
col. 34.
66. Pliny the Younger, EpistolaeJ VI, 24. The case was recently well discussed by
A. Keaveney and J. Madden in Hermes 107 (1979): 499ff., who refute the diagnosis
of syphilis proposed in Pauly-Wissowa, Real Encyclopiidieder AltertumswissenschaftJ
15: 1026.
67. Celsus, De medicinaJ VI, 18, 2. Cf. W.R. Bett, A Short History of Some Common
Diseases (Oxford, 1934), p. 35.
68. According to Palladius, Hist. Lausiaca (Migne, Patrologia GraecaJ XXXIX, col.
a
1091). See E. Jeanselme, "Le chancre mou existait-il l'Alexandrie au Ive siede?"
Bull. Soc. Fran(. Hist. Med. 14 (1920): 233-38.
69. Flavius Josephus, Against ApionJ II, 13.
70. Mul"J II, n5 and 122 (Littre, VIII, 248 and 258). See also Mul"J II, 176, where
the symptoms of purulent endometriosis are prominent.
71. Herodotus, III, 149.
Chapter Six: Leprosy

r. For the history of this research, see V. M~ller-Christensen, "New Knowl-
edge of Leprosy through Paleopathology," Intern. ]our. Leprosy 33 (1965): 603-10,
and J.G. Andersen, "Studies in Mediaeval Diagnosis of Leprosy in Denmark,"
Danish Med. Bull. suppl. 9, 16 (1969): 1-124.
2. L. Gluck, "Die Lepra der oberen Atmungswege und Verdauungswege,"
Mittheilungen und Verhandlungen der Intern. Lepra-Conferenz zu Berlin 1 (1897): 1-93.
3. V. M~ller-Christensen, Bone Changes in Leprosy (Copenhagen, 1961), and Lep-
rosy Changes in the Skull (Odense, 1978).
4. V. M~ller-Christensen, ''Evidence of Tuberculosis, Leprosy, and Syphilis
in Antiquity and the Middle Ages," Proc. XIX Intern. CongressHistory Med. (Basel)
1964) (Basel, 1966), pp. 229-37.
5. See G. Elliott Smith and F. Wood Jones, "Report on the Human Remains,"
Archaeological Survey of Nubia) Report for 1907-1908) vol. 2 (Cairo, 1910); G. Elliott
Smith and D. E. Derry, "An Anatomical Report," Archaeological Survey of Nubia)
Bull. 6 (1910); V. M~ller-Christensen and D. R. Hughes, "An Early Case of
Leprosy from Nubia," Man 62 (1966): 242-45; and Andersen, "Studies in Medi-
aeval," (above, n. 1), pp. 49 and 79.
6. T. Dzierzykray-Rogalski, "Paleopathology of the Ptolemaic Inhabitants of
Dakleh Oasis (Egypt)," ]our. Hum. Evol. 9 (1980): 71-74. ·
7. These cases (except for the last two) were published by D. Brothwell, C.
Wells, and V. M~ller-Christensen. In this connection see D. Brothwell and A. T.
Sandison, Diseases in Antiquit;y (Springfield, 1967), p. 303. Since the publication just
cited, leprous lesions were diagnosed on the lower extremities of a skeleton found
at Dorset in a Roman necropolis of the fourth century A.D. See 0. K. Skinsnes,
"An Ancient Briton Adds to the Story of Leprosy," Intern. ]our. Leprosy 44 (1975):
387ff. However, this diagnosis, which is based exclusively on the "lollipop" ap-
pearance of the phalanges, is uncertain, as E. Jonquieres (Inter. ]our. Leprosy 45
[1977]:66) has remarked in responding to Skinsnes's publication. Another British
case, from Poundbury (about A.D. 400), is also doubtful to the extent that the
diagnosis relies only on vague, noncranial lesions. Cf R. Reader, "New Evidence
for the Antiquity of Leprosy in Early Britain," ]our. Archeol. Sci. 1 (1974): 205-7.
8. Brothwell and Sandison, Diseases(above, n. 7), p. 302.
9. J. L. Angel, 'JHuman Skeletal Remains at Karata~,"Amer. ]our. Archaeol. 74
(1970): 256.
10. See I. Bloch, "Zur Vorgeschichte des Aussatzes," Verh. Berliner Ges. An-
throp. 31 (1899): 205-16 (esp. pp. 210 and 214ff.).
n. F. E. Hoggan, "The Leper Terra-cotta of Athens," ]our. Hellen. Stud. 13
(1892): 101-3.
12. Louvre Museum, terra cotta figurine E 17. The diagnosis of leprosy was
posited by F. Regnault, "Les figurines antiques devant l'art et la medecine,"
Medicina (Paris) 4 (1907): 26ff. See the description and photographic reproduction
in S. Besques, Musee Nationale du Louvre) Catalogueraisonneedesfigurines et relieft en
terre cuitegrecsJetrusquesJet romains (Paris, 1972), 3: 95, and pl. 120h. This object can
be compared with another terra cotta from Troy, dating from the first century
A.D., Louvre EID 567, where the deformity of the nose is insufficient evidence of
a leprous ailment. In the Histoiregenerate de la medecinepublished under the direc-
tion of M. Laignel-Lavastine (vol. 3 [Paris, 1948]), there is a photograph of a
Hellenistic figurine in the Louvre that, according to the caption, represents a leper.
The terra cotta in question, which comes from Smyrna and is catalogued at the
Louvre as D 14~9, does indeed call to mind leontiasis, but such a diagnosis is not
justified: the presence of a single eye on the forehead proves that it is a caricature
of the Cyclops Polyphemus.
13. Besques, Musee du Louvre) Catalogue raisonnee(above, n. 12), 3: 222, and pl.
303h. The figurine is numbered EID 1681.
14. Louvre EID 1793,E 144, and EID 1823.See Regnault, "Les figurines" (above,
n. 12), pp. 26 ff., and Besques, Musee du Louvre) Catalogueraisonnee(above, n. 12),
3: 234, pls. 3nm, 312d, and 313k.The head E 144 shows all the signs of rhinoscler-
oma. These are just a few representative examples; over twenty figurines of this
type are known.
15. Greco-Roman Museum of Alexandria and Fouquet Antiquities Collection
no. 455. See A. Panayotatou, "Terres cuites d'Egypte de l'epoque greco-romaine
et maladies," Actes VJe CongresIntern. Hist. Med. (Leiden and Amsterdam, 1927),
pp. 41-47, esp. figs. 28a and 286.
16. Panayotatou, "Terres cuites" (above, n. 15), p. 45.
17. Benaki Museum, Athens, inv. no. 12599.
18. Archeological Museum of Dijon, ex-voto no. 97 (Catalogue of S. Deyts).
See R. Bernard and P. Vassal, "Etude medicale des ex-voto de la Seine," Rev.
Archeol. de l)Est 9 (1958):332.
19. Rockefeller Museum, Jerusalem; see M. Yoeli, "A FaciesLeontina of Lep-
rosy on an Ancient Canaanite Jar," ]our. Hist. Med. 10 (1955):331-35.
20. See, in this regard, Dharmendra, "Leprosy in Ancient Indian Medicine,"
Intern. ]our. Leprosy 15 (1947): 424-30; M. Schar-Send, "Die Lepra in der alt-
indischen Medizin und Gesellschaft," in H. M. Koelbing et al., Beitrage zur Ge-
schichte der Lepra (Zurich, 1972), pp. n-33; and above all, J. Filliozat, La doctrine
classiquede la medecineindienneJ 2d ed. (Paris, 1975).
21. Susruta, NidanasthanaJ v, 12-14. Cf. Filliozat, La doctrine (above, n. 20),
pp. 82 and 103.
22. See especially NidanasthanaJ v, 7. Cf. Schar-Send, "Die Lepra," 1972,
(above, n. 20), pp. 19-23, and Filliozat, La doctrine (above, n. 20), p. 95.
23. NidanasthanaJ v, 26.
24. In this regard I refer the reader especially to the law code of Manu and the
political treatise of Kautiliya.
25. Huang-ti Nei-ching (Yellow Emperor's Manual), Su-wen, xn, 42, after the
French translation by A. Husson (Paris, 1973), p. 197.
26. See K. Chimin Wong and Wu Lien-teh, History of Chinese Medicine (Tien-
tsin, 1932), pp. 103-5; Liu Mu Chih, "Data on Leprosy in Chinese Medical Litera-
ture and Historical Records," Chin. ]our. Derrn. 4 (1956):3ff.; P. Huard and M.
Wong, La medecine chinoise au cours des siecles(Paris, 1959),pp. 19-20, 24, and 61; Lu
Gwei-djen and J. Needham, "Records of Diseases in Ancient China," in Broth well
and Sandison, Diseases in Antiqui-ty (above, n. 7), pp. 222-37, esp. p. 233.
27. Lun yiiJ VI, 8.
28. K. Chimin Wong and Wu Lien-teh, History (above, n. 26), p. 103, and Lu
Gwei-djen and Needham, "Records," (above, n. 26), pp. 236ff.
29. Pap. EbersJ no. 877. For the most recent version of his argument in favor of
the identification of leprosy in this papyrus, see B. Ebbell, Bei-triigezur iiltesten
Geschichte einiger Infektionskrankheiten (Oslo, 1967), pp. 68-84. For a critique, see
H. Grapow et al., Grundriss der Medizin der alten Argypter (Berlin, 1958),rvh: 288;
Andersen, "Studies in Mediaeval" (above, n. 1), pp. 10-14; and Koelbing et al.,
Beitriige (above, n. 20), pp. 35ff.
30. See K. Sudhoff, "Die Krankheiten bennu und sibtu der babylonisch-
assyrischen Rechtsurkunden," Arch. Gesch. Med. 4 (1911):353-69; H. E. Sigerist,
History of Medicine (New York, 1951),1: 381and 398; J. V. Kinnier Wilson, "Leprosy
in Ancient Mesopotamia," Rev. Assyr. et Archeol. Orient. 60 (1966): 47-58. The
antiquity of the isolation of persons affected by certain diseases in this region is
confirmed by a letter of Zimri-lim, king of Mari in the eighteenth century B.C. See
A. Pinet, "Les medecins au royaume de Mari," Ann. Inst. Philo!. Hist. Orient. 14
(1954-57): 123-44, esp. p. 129.
31. VAT 7525, col. 2, 11.42-45. Wilson, "Leprosy," (above, n. 30), and F.
Kocher and A. L. Oppenheim, "The Old-Babylonian Omen Text VAT 7525,"
Arch. fur Orieniforsch. 18 (1957):62.
32. Rev. Assyriol. 71, (1967): 190. Cf. also S. G. Browne, Leprosy in the Bible
(London, 1971). I note that it is incorrect to introduce into the discussion of
Mesopotamian leprosy the absence of paleopathological proof, since the number
of specimens that has been examined from this region is still too small.
33. Leviticus 13:2-46. See the detailed analysis of this text in J. Preuss, Biblische-
talmudische MedizinJ 3d ed. (Berlin, 1923),pp. 369-90.
34. For studies subsequent to Preuss, see E. Jeanselme, La lepre (Paris, 1934),
pp. 12-16; R. G. Cochrane, Biblical Leprosx 3d ed. (Glasgow, 1963); I. Goldman,
"White Spots in Biblical Times," Arch. Derrn. (Chicago) 93 (1966): 744-53; I.
Simon, "La dermatologie hebrai:que clans l' Antiquite et au Moyen Age," Rev.
Hist. Med. Hehr. 27, no. 110 (1974): 149-54, and ibid. 28, no. III (1974): 7-14.
35. Exodus 4:6, Numbers 12:106; 2 Kings 5:27.
36. E. V. Hulse, "The Nature of Biblical Leprosy and the Use of Alternative
Medical Terms in Modern Translations of the Bible," Palestine Exploration Quar-
terly 107 (1975):87-105.
37. Preuss, Biblische-talmudische Medizin (above, n. 33), p. 372.
38. For example, see Andersen, "Studies in Mediaeval," (above, n. 1), p. 15.
39. Flavius Josephus, Contra ApionJ I, 26-32, 34, and 35.
40. Contra A pion) I, 26 (esp. sec. 229).
41. Tacitus, Histories) v, 3; Justinus, Historia PhilippicaeJ XXXVI 2; Diodorus
Siculus, xxxrv, 2.
42. Flavius Josephus, Contra ApionJ I, 31(especially sees. 281-82).
43. Herodotus, I, 138.
44. Aeschylus, ChoephoroiJ 279-82. See J. Dumortier, Le vocabulaire medical
d)Eschyle et les ecrits hippocratiquesJ 2d ed. (Paris, 1975),pp. 80-83.
45. P. Chantraine, Dictionnaire ltymologique de la langue greque (Paris, 1974) 3:
630-31.
46. Nicander, Theriaca) 156and 262.
47. Morb') I, 3 (kittre, VI, 144).
48. Aph.) III, 20 (Littre, IV, 494).
49. Epid.)VI, III, 23, and Hum.) 20 (Littre, v, 304 and 500).
50. Epid') II, I, 7 (Littn~, v, 78).
51. Alim.) 20 (Littre, IX, 104). See K. Diechgraber, Pseudo-HippokratesUeberdie
Nahrung (Wiesbaden, 1973), pp. 28 and 35. This is a late treatise probably dating
from the first century A.D ..
52. Liqu.) 4 (Littre, VI, 128). The use of salt is recommended for leprous nails.
53. Epid') II, v, 24 (Littre, v, 132).
54. See Koelbing et al. Beitrage) (above, n. 20), pp. 4off.
55. H-orrh')II, 43 (Littre, IX, 74).
56. Hum') 17 (Littre, v, 498).
57. Epid') V, 17and 19 (Littre, VI, 246).
58. Aff) 35, (Littn\ VI, 246).
59. Theophrastus, Sweat) 14; Characters) 12.
60. Aristotle, H-oblemata)VII, 8 (887a).
6r. Pausanias, Descriptionof Greece)v, 5, and Suda) s.v. Lepreon. There may have
been a temple of Zeus Leukaios in his town, but it is possible that the name in the
textual tradition of Pausanias is simply a corruption of Zeus Lukaios. On sulfurous
springs, see Stephanos in Dictionnaire des sciencesmidicales (Paris, 1884), 4th ser., 10:
53off.
62. See in particular ltorrh') II, 43. Plato mentions among the diseases caused
by phlegm one that "dapples the body with white spots" (Timaeus) 85a).
63. H-aen. Coacae)502 (Littre, v, 700).
64. See Aristotle, Hist. anim.) 518a;Degener. anim') 784a. Concerning Atossa,
daughter and then wife of Artaxerxes, king of Persia from 404 to 358B.C., Plutarch
says that "her father loved her so much that, once she had become his wife, he
felt not the slightest disgust for her when her body was taken over by an alph6s))
(Life of Artaxerxes) 23, 7). The example is an instructive one because it concerns the
Persian court, where contracting lipra or leulie in their Herodotean sense meant
banishment. That the disease Plutarch calls alph6swas benign is proven by the rest
of the story of Atossa: she became her brother-in-law's lover and outlived her
father-husband despite his exceptional longevity.
65. H-orrh.)II, 43 (Littre, IX, 74).
66. See Littre, IX, 74, n. 7 of the apparatus criticus.
67. The critical edition of the text of this letter is by E. Drerup (Leipzig, 1904);
for a commentary and the dating of it, see K. Schwengler, De Aeschinis Quae
Feruntur Epistolis (Giessen, 1913).
68. Oribasius, Collectiomedica) XLV, 28. For Straton, see M. Wellman, Phiww-
gische Untersuchungen) XIV, 24, and Kind, s.v. Straton) in Pauly-Wissowa, Real-
Encyclopadie) 2d ser., vol. IV /r, col. 316.
69. After the French translation by Bussemaker and Daremberg, Oeuvres
d)Oribase (Paris, 1876), 4: 63ff.
70. Such is the opinion of the leprologist Andersen, "Studies in Mediaeval"
(above, n. 1), p. 43.
71. According to chapter 13of the Pseudo-Galenic treatise Introductio seu medicus
(Kiihn, XIV, 757), some authorities used the term elephantiasisin a generic sense,
proposing to include in it six particular forms of diseasss, nan1ely, elephantiasis(in
the narrow sense), leontiasis)ophiasis)lipra) alopekia)and lobe.Since alopekia(a disease
that, from the Pseudo-Galenic description, may be related to Aristotle's leulie)and
ophiasiscertainly have nothing to do with the leprosy complex proper, the same
thing can be stated with regard to lipra) and there is therefore no need to see this
classification as an attempted nosological identification of "leprosy" in the Hip-

pocratic sense with "leprosy" in the modern sense. Nevertheless, it may be true
that this text, which is attributed to Galen and was much appreciated in the Middle
Ages, contributed to the false explication of the classical nosological terminology
and to the confusion from which the history of leprosy has long suffered. See
Koelbing et al., Beitriige (above, n. 20), pp. 47 and 55.
72. According to Aristotle, "In the disease called sa-tyr[asisJ
. . . as the result of
an abundance of humor or of breath that has not undergone coction and pene-
trates into the parts of the forehead, the face resembles that of another animal or a
satyr"; De generatione animaliumJ 7686, after the edition and French translation by
P. Louis (Paris, 1961), p. 150. The traditional explanation (see the Liddel-Scott-
Jones lexicon s.v. saturfasisJII, and the note in Louis's edition) is that the defor-
mity that makes the disease's victims resemble satyrs is the development of frontal
exostoses. Ho,vever, J. Jouanna and J. Taillardat ("Une vox nihili," Rev. Etudes
Grecques93 (1980): 132)have well shown that this notion is only a poor interpreta-
tion of a text in Galen; through comparison with passages in the Hippocratic
corpus it is possible to deduce that the deformities in question were swelling near
or under the ears.
73. Lucretius, De rerum naturaJ vr, 1n4ff.
74. Celsus, De medicinaJIII, 25. Galen says that in his time elephantiasiswas very
common in Alexandria, rare in Germany and Mysia, and unknown among the
Scyths (Kiihn, xr, 142).
75. Pliny, Natural Histmx XXVI, I (trans. W.H.S. Jones, vol. 7, [London and
Cambridge, 1956],p. 265).
76. Pliny, Natural Hist01x xxvi, 5 (Jones, p. 271).
77. Plutarch, QJtaestionesconvivialesJVIII, 9.
78. In this regard, see the excellent analysis of Plutarch 's text by C. Mugler,
"Democrite et les dangers de !'irradiation cosmique," Rev. Hist. Sci. 20 (1967):
221-28.
79. Aretaeus of Cappadocia, Signa chron.J II, 13. See the edition of the Greek
text by K. Hude, Corpus Med. GraecorumJII, 2d ed. (Berlin, 1958),pp. 85-90, and
the modern medical commentary in Andersen, "Studies in Mediaeval" (above,
n. 1), pp. 30-40, and Koelbing et al., Beitriige (above, n. 20), pp. 43-54.
So. Aretaeus, Signachron'J II, 13, 19 (Hude, pp. 89 ff.).
81. For instance, Galen still passionately defends the Hippocratic theory that
pestilences were due to miasmata and could not pass from one human to the next.
Nevertheless, Aretaeus points out and adopts the popular notion that plague
victims and lepers are both to be avoided, since sickness is transmitted to healthy
people by the breath of the ill. With regard to the treatment of leprosy, he says
that "we fear cohabitation and communal life with them no less than if they were
victims of plague; the infection (baphe) is easily transmitted by breath" (Cura
chron'JII, 13, I [Hude, 168]).
82. This Saint Lazarus, the medieval patron of lepers, is a Biblical figure, an
ulcerous pauper who begged at the gate of the evil rich man (Luke 16:19-31);he is
not to be confused with the brother of Martha who, in the GospelaccordingtoJohn)
was miraculously brought back to life.
83. See G. Kurth, La lepreen Occident avant les Croisades(Paris, 1899); E. Jean-
selme, "Comment !'Europe au Moyen Age se protegea contre la lepre?" Bull. Soc.
Franf. Hist. Med. 25 (1931):1-155;E. Jeanselme, La lepre (Paris, 1934); J. Cougoul,
La lepre dans Fancienne France (Bordeaux, 1943); W. Dorr, Ueber den Aussatz im
Altertum und in der Gegenwart (Heidelberg, 1948); R. G. Cochrane, "The History
of Leprosy and Its Spread throughout the World," in R. G. Cochrane and T. F.
Davey, eds., Leprosy in Theory and fractice (Bristol, 1964); H. Schadewaldt, "Zur
Geschichte der Lep~a," Hautarzt20 (1969): 124-30.
84. See S. N. Brody, The Disease of the Soul: Leprosy in Medieval Literature (Ithaca
and London, 1974).
85. See Koelbing et al., Beitrage (above, n. 20), pp. 62ff.
86. For statistical data about leprosy and its geographical distribution, the au-
thority is P. Harter, frecis de leprowgie (Paris, 1968), and thereafter, the recent
epidemiological bulletins of the W odd Health Organization and reports of inter-
national conventions on leprosy. As for Greece, leprosy was still relatively common
there in the nineteeth century: in 1840 and 1843, official statistics recorded a little
more than 150 cases, to which must be added 1,000 lepers on the island of Crete.
To be sure, these figures are low. The centers of the disease were the northern part
of Corfu, in Gortynia, in the southern part of Euboea, and on the islands of
Lesbos, Thera, and to repeat, Crete. See C. Stephanos, "Grece," in Dictionnaire
des sciencesmidicales) published under the direction of A. Dechambre (Paris, 1884),
4th ser. 10: 529ff. It is not uninteresting to note that, according to Caelius Aureli-
anus (Acut.) III, 18)the physician Themison (first century) noticed the commonness
of leprosy in Crete.
87. D. Zambaco Pacha, La lepre a travers les siecleset les contrees (Paris, 1914); G.
Barbezieux, "Contribution a l'etude de l'histoire de la lepre: La lepre clans la plus
haute antiquite," Janus 19 (1914):132-49.
88. R. Chaussinand, La lepre (Paris, 1954).
89. T. A. Cockburn, "Infectious Diseases in Ancient Populations," Current
Anthropowgy 12 (1971):42-62, esp. p. 48.
90. See A. Basset, "La lepre, son interet clans l 'etude de l 'epidemiologie et de
la pathologie generale,'' Concours Medical 93 (1971):5627-37.
91. See, for instance, Andersen, "Studies in Mediaeval" (above, n. 1), p. 45,
and S. G. Browne, "How Old Is Leprosy?" Brit. Med. ]our. (1970), pp. 64off.
Chapter Seven: Tuberculosis

1. My late friend Charles Coury relied on his twofold expertise in tuberculosis
and in history to create Grandeur et diclin d)une maladie: La tubercuwse au cours des
siecles(Paris, 1972). For the ancient history of tuberculosis, nothing has yet equaled
the documentation in the sumptuous monograph by J. Jedlicka, Vyvojfthiseowgie)
nauky o tuberkuwse (Prague, 1932); this work is in Czech, which means that it is
hardly known and never used. As for other studies of the history of this disease
around the world and through time, I cite M. Piery and J. Roshem, Histoire de la
tubercuwse (Paris, 1931);A. Castiglioni, History of Tubercuwsis) (New York, 1936); R.
Bochalli, Geschichte der Schwindsucht (Leipzig, 1940); R. Burke, An Historical Chro-
nowgy ofTubercuwsis) 2d ed. (Springfield, 1955).
2. For osteoarchaeological diagnosis of tuberculosis, see D. Morse, "Tubercu-
losis," in D. Brothwell and A. T_.Sandison, eds., Diseasesin Antiquity (Springfield,
1967), pp. 249-71; T. Steinbock, Paleopathowgical Diagnosis and Interpretation
(Springfield, 1976); and M. A. Kelley and M. Y. El-Najjar, "Natural Variations
and Differential Diagnosis of Skeletal Changes in Tuberculosis," Amer. ]our. Phys.
Anthrop. 52 (1980): 153-67; D. J. Ortner and W. G. J. Putschar, Identification of
Pathowgical Conditions in Human Skeletal Remains (Washington, D.C., 1981),
pp. 141-75.
3. See Morse, "Tuberculosis" (above, n. 2), p. 250.
4. L. Pales, Paleopathowgie et pathowgie comparative (Paris, 1930), p. 226;
V. M~ller-Christensen, "Evidence of Tuberculosis, Leprosy, and Syphilis in An-

tiquity and the Middle Ages,'' ltoc. XIX Intern. Cong. Hist. Med. (Basel, 1966), pp.
229-37; P. Sager, M. Schalimtzek, and V. M~ller-Christensen, "A Case ofSpondy-
litis Tuberculosa in the Danish Neolithic Age," Danish Med. Bull. 19 (1972):
172-80.
5. P. Bartels. "Tuberkulose (Wirbelkaries) in der jiingeren Steinzeit," Arch.
Anthrop. 34 (1907): 243-55.
6. See C. Wells, Bones) Bodies) and Disease (London, 1964), p. 97.
7. G. E. Smith and M. A. Ruffer, "Pott'sche Krankheit an einer agyptischen
Mumie aus der Zeit der 21. Dynastie (um 1,000 v. Chr.)" I(rankheitserregerJ fasc. 3
(Giessen, 1910).
8. See A. J. E. Cave, "The Evidence for the Incidence of Tuberculosis in
Ancient Egypt, "Brit. four. Tub. 33 (1939):142; D. Morse et al., "Tuberculosis in
Ancient Egypt," Amer. ReJJicwResp. Diseases 90 (1964):524-41; A. P. Leca, La
midecine fgyptienne au temps des Pharaons (Paris, 1971),pp. 233ff.; and J.B. Bourke,
"The Palaeopathology of the Vertebral Column in Ancient Egypt and Nubia, "
Med. History 15(1971):370.
9. It is a small wooden statuette representing a man with a high thoracic
kyphoscoliosis and an angular projection of the sternum (Musees Royaux de Bru-
xelles, no. E 5850). For its description and arguments in favor of the diagnosis of
tuberculosis, see F. Jonckheere, "Le bossu des Musees Royaux d'Art et d'Histoire
de Bruxelles," Chronique d)Egypte 23 no. 45 (1948): 24-35. His opinion is not shared
by Leca, La midecine fgyptienne (above, n. 8), p. 235, according to whom "nonspe-
cific kyphoscoliosis can result in a similar deformity." Another example of a figu-
rine with Pott's disease was published by B. Schrumpf-Pierron, "Le mal de Pott
en Egypte 4,000 ans avant notre ere," Aesculape 23 (1933):295.
10. Michael R. Zimmerman has succeeded in demonstrating the presence of
acid-resistant bacilli in the vertebrae and extravasated blood in the lungs of a child
mummy. See M. R. Zimmerman, "Pulmonary and Osseous Tuberculosis in an
Egyptian Mummy," Bull. New York Acad. Med. 55 (1979): 604-8.
n. Coury, Grandeur et diclinJ (above, n. 1), p. 10.
12. R. Campbell Thompson, "Assyrian Prescriptions for Diseases of the Chest
and Lungs," Rev. Assyr. 31(1934): 8-22.
13. See P. Cordier, "Histoire de la medecine indienne: La phtisie pulmonaire,"
Ann. Hyg. Mid. Colon. 15 (1912):255-66 and 535-48; D. V. Subba Reddy, "Tuber-
culosis in Ancient India," Bull. Hist. Med. Hyderabad 2 (1972):156-61.
14. See T'ao Lee, ''Tuberculosis in Ancient China,'' Chin. Med. four. 61
(1942):272-80.
15. D. Morse, "Prehistoric Tuberculosis in America," Amer. Rev. Respir. Dis. 83
(1961):489-504. See also Morse, "Tuberculosis," pp. 257ff.
16. See especially J. G. Roney, Jr., "Palaeoepidemiology: An Example from
California," in S. Jarcho, ed., Human Palaeopathology(New Haven, 1966), pp. 101-
3; M. J. Allison, D. Mendoza, and A. Pezzia, "Documentation of a Case of
Tuberculosis in Pre-Columbian America," Amer. Rev. Respir. Dis. 107 (1973): 985-
91; M. Kelley, Paleopath. Newsletter) no. 26 (June, 1979), p. 4; M. Y. El-Najjar,
"Human Treponematoses and Tuberculosis: Evidence from the New World,"
Amer. four. Phys. Anthr. 51(1979): 599-618; J. Buikstra, ed., ltehistoric Tuberculosisin
the Americas (Evanston, 1981).
17. C. Coury, La midecine de l)Amirique precolombienne (Paris, 1969), pp. 84ff.
18. Ibid., p. 14.
19. Ibid., p. 194.
20. J. L. Angel, "Problems in Diagnosis," Annual Meeting of the Paleopatho-
logicalAssociation (Toronto, April 1978).
21. See A. Krause, "Tuberculosis and Public Health," Amer. Rev. Tub. 18(1928):
271-73.
22. Odyssey xr. 171-74 and 198-203; v.394-97; Iliad XIII. 663-72. Sophocles,
Antigone) 819; Euripides, Alcestis) 204 and 236.
23. Aside from the general bibliography on the history of tuberculosis cited
above in n. 1, see the following publications, which are devoted to the history of
the disease in classical antiquity: B. Meinecke, "Consumption (Tuberculosis) in
Classical Antiquity," Ann. Med. Hist. 9 (1927): 379-402; E. D. Baumann, "De
phthisi antiqua," Janus 34 (1930):209-25 and 255-72; H. Gertler and D. Schultz,
"Antike Aussagen zur Schwindsuchtsfrage," Zschr. Tuberk. 122 (1964):282-89.
24. See H. Frisk, Griechischese-tymologisches Worterbuch Heidelberg, 1969), 2:
1014-16, and P. Chantraine, Dictionnaire e-tymologique de la languegrecque (Paris,
1980), rv/2: 1201.
25. Herodotus, VII, 88, trans. A. de Selincourt (Harmondsworth, 1954; rev.
and rept. 1976), p. 471.
26. Meinecke, "Consumption" (above, n. 23), p. 381.
27. F. Kudlien, Der Beginn des medizinischen Denkens bei den Griechen (Zurich
and Stuttgart, 1967), p. no.
28. It could as easily be hemoptysis as hematemesis; classical medical texts use
derivatives of the verb emeo to designate without distinction vomiting and spitting.
29. Morb._,r, 3 (Littre, vr, 144). See R. Wittern, Die hippokratischeSchrift De
morbisr (Hildesheim and New York, 1974), pp. 9 and 10.
30. K. Deichgraber, "Zur Milchtherapie der Hippokratiker (Epid. VII)," in
Medizingeschichtein unsererZeit (FestschriftArtelt) (Stuttgart, 1971),p. 50, n. 10.
31. Pseudo-Galen, Definitiones medicae) 261 (Kiihn, XIX, 419). See C. Darem-
berg, Oeuvres choisiesd)Hippocrate) 2d ed. (Paris, 1855),p. 264, and Wittern, Die
hippokratischeSchrift (above, n. 29), p. 194.
32. Morb.) II, 49. This phthoe is a long-term disease, according to the statement
in Morb.) r, 3. However, it can be cured (if the disease is treated from the start),
which creates a semantic problem, since then the disease is not properly a subclass
of phth{sis in Morb._,r, which must be fatal.
33. Aff int.) 10 (Littre, VII, 188-90). See L. Bourgey, Observationet experience
chez les medecinsde la Collectionhippocratique(Paris, 1953),p. 149, and Coury, Gran-
deur et diclin (above, n. 1), p. 16.
34. Aff int._, (Littre, VII, 192). After the translation by Bourgey, Observation
(above, n. 33), p. 149, except for the passage on the morning cough. According to
Littre, "The cough especially lays low old people." For correction of this passage,
see R. Wittern, "Zur Krankheitserkennung in der knidischen Schrift De internis
affectionibus," in Medizinische Diagnostik in Geschichteund Gegenwart (Munich,
1978), p. 109.
35. Aff int._,12 (Littre, VII, 192-94.)
36. Morb._,n, 48 (Littre, VII, 72). After the translation by Jacques Jouanna,
based on his critical edition of the Greek text, Maladies II (Paris, 1983),pp. 183ff.
37. Morb._,49 (Littre, VII, 74-76). After the translation by Jouanna, Maladies II
(above, n. 36), pp. 185ff.
38. Morb.) 50 (Littre, VII, 76; Jouanna, 186).
39. In this regard, see R. E. Siegel, "Clinical Observations in Hippocrates: An
Essay on the Evolution of the Diagnostic Art," ]our. Mt. Sinai Hosp. 31(1964):285-
303, esp. p. 295. The morning cough is also mentioned in Aff int.) n (see above,
n. 34).
40. See A. Baffoni, Storia dellepleuriti da Ippocratea Laennec (Rome, 1947).
41. Swelling of the fingertips with curving of the nails, noticed by the Greek
physicians and not ignored since, is still an indicator, as puzzling as it is reliable, of
severe disturbances in the organism. The physiopathological mechanism in ques-

tion is unknown; in the diagnosis of pulmonary tuberculosis, they are not pathog-
nomonic, since they occur in conjunction with a variety of intrathoracic disorders.
Pulmonary suppuration remains their principal etiology, however, whence the old
proverb about the examination of the lungs beginning at the fingertips. See C.
Coury, L)hippocratismedigital (Paris, 1960), and "Le signe du doigt hippocratique,"
Pagine Stor. Med. 12, no. 2 (1968)3-12.
42. Aph'J v, n (Littn\ IV, 536), after Daremberg's translation.
43. ltaen. Coacae)426 (Littn\ v, 680). Subsequently, Celsus, Caelius Aureli-
anus, Aretaeus, and others repeat this text. See W. D. Sharpe, "Lung Disease and
the Greco-Roman Physician, Amer. Rev. Respir.Dis. 86 (1962): 178-92.
44. ltaen. Coacae)427 (Littn\ v, 680).
45. ltaen. Coacae)428 (Littn\ v, 680). Cf. Aph.J v, 12 and 14.
46. ltaen. Coacae)425 (Littre, v, 680). Cf. Aph'J v, 13.
47. Loe. Hom') 14 (Littre, VI, 302-8).
48. Morb'J I, 19 (Littre, VI, 174-76; Wittern, pp. 48-52).
49. Galen, Comm. in Hipp. Epid'J VI, 13 (Kiihn, XVII, 855). Celsus translates
phuma by tuberculum) at least when the Greek word applies to formations located
in the lungs that are manifested by coughing up blood (De medicina, IV, 5). See
also H. Dont, Die Tenninowgievon Geschwur)Geschwulst)und Anschwellung im Cor-
pus Hippocraticum (Vienna, 1968).
50. Daremberg, Oeuvreschoisies(above, n. 31), p. 282.
5r. See especially the analysis by W. Pagel, "Die Krankheitslehre der Phthisie
in den Phasen ihrer geschichtlichen Entwicklung," Beitr. J(lin. Tuberkl. 66 (1927):
66-98.
52. Aristotle, Parva naturalia) 4786. See the text edited and translated by R.
Mugnier (Paris, 1953),pp. 13off. I should note that Aristotle, in describing a disease
of cattle called krauros) mentions that "the signs of this disease are hanging ears
and loss of appetite; death comes quickly, and on dissection, the lung appears
rotten'' (Hist. anim'J VIII, 23, 604a).
53. Epid.JI, 3 (Littre, II, 604-10), in Hippocrates)trans. W. H. S. Jones (London
and Cambridge, 1923),1: 148-50.
54. For the definition of katastasis) see below, (chapter 9, the section ''The
Place and Date of the Disease of Philiscus."
56. It is amusing to see two radically opposed opinions on the frequency of
consumption in the Greek world of this time. Bruno Meinecke asserts that "Hip-
pocrates describes consumption so often and so fully that we are forced to the
conclusion that it must have been very prevalent already in his days" (Annals of
Medical History 9 [1927]: 381). Another, no less competent, specialist states that
"although there are these references to the disease [phthisis], one cannot say that
they occupy a leading place in Hippocratic writings, and we may assume that,
though tuberculosis affected the Greeks of Hippocratic days, it played no very
evident or particular part in the incidence of disease" (J. Fraser, "Tuberculosis,"
in W.R. Bett, A Short History of Some Common Diseases[London, 1934], p. 17). I
favor Meinecke's opinion. The number of times consumption is mentioned in the
Hippocratic corpus is less significant than the content of certain passages, especially
those in the Epidemics.It is also essential to keep in mind the frequency and bearing
of texts that refer to pleurisy, empyema, and other diseases whose etiology is at
least partially tuberculous.
57. It is likely that in this regard popular opinion surpassed professional knowl-
edge. The orator Isocrates alludes to the common belief that chronic consumption
is contagious (Aegin'J XIX, 29).
58. Aph.) v, 9 (Littre, IV, 534). The same statement can be found in Praen.
Coacae)431. According to Aph ') VII, 88, the ages most vulnerable to consumption
are from 18 to 30. See also Aph.) III, 29, which speaks of hemoptysis and consump-
tion as diseases that prevail an1ong young people.
59. ltaen. Coacae)513(Littre, v, 702).
60. lton-h') II, 7 (Littre, IX, 24). This statement is based on correct observation
but incorrect inversion of the causal chain: amenorrhea is the effect, not the cause,
of tuberculous disease.
61. Aer.) 4, (Littre, II, 22).
62. Morb. sacr.)5 (Littre, VI, 364; Grensemann, p. 68).
63. lton-h.) II, 5 (Littre, IX, 20).
64. Epid') III, 14 (Littre, III 96-98).
65. lton-h') II, 7 (Littre, IX, 24).
66. G. See, De la phtisie bacillairedespoumons (Paris, 1884), p. 137.
67. Epid') VII, 49 (Littre, v, 418). The same account occurs also in Epid') v, 103
(Littre, v, 258). Translation after an as yet unpublished edition by Fernand Robert.
68. Translated in accord with the text given in the V manuscript: apohrimpsies
hupopiwdies.Littre reads: apohrempsiespiwdies.
69. Littre translates "la phtisie s'etablit" [consumption took hold], a rendering
that in the context seems to me more satisfactory from a medical viewpoint. But
to reach this version, he has to eliminate the word kaz in the phrase phthfsis kaz
katiste. However, that word is attested in both M and V and the recentiores.
70. See P. Diepgen, Die Frauenheilkunde der antiken Welt (Munich, 1937), pp.
265 and 267.
71. Epid') VII (Littre, v, 378). My translation of this case history follows the as
yet unpublished critical edition by Fernand Robert.
72. The Hippocratic physicians did not know about taking the pulse, but they
understood the value of visually marking the beat of the temporal artery.
73. Celsus, De medicina) III, 22.
74. Aretaeus of Cappadocia, Signa chron.) I, 8. See the commentary by C.
Coury, "La pathologie pulmonaire clans I' oeuvre d 'Aretee de Cappadoce," ltesse
Mid. 70 (1962): 655-57. When Laennec made his magisterial description of pulmo-
nary tuberculosis in the nineteenth century, he did not omit to recall "the cough-
ing, dyspnea, purulent sputa, hectic fever, hemoptysis, marasmus, in sum, the
totality of symptoms whose image Aretaeus traced with terrifying realism"
(R. T. H. Laennec, De !)auscultationmediate (Paris, 1819),1: 57).
75. See especially Sharpe, "Lung Disease" (above, n. 43), and Gertler and
Schultz, "Antike Aussagen" (above, n. 23).
76. Aretaeus, Signa chron.)I, 8.
77. Plutarch, Life of Aratus) 52. The association of consumption with poisoning
seems to have been a popular belief iffthe Hellenistic and Roman periods. In this
connection, see the epitaph of Thermion, who died in Alexandria in the first century
(E. Bernard, Inscriptionsmitriquesde l)Egyptegrico-romaine[Paris, 1969], no. 46).
78. Plutarch, Life of Cleomenesj30. See Baumann, "De phthisi antiqua" (above,
n. 23), p. 271. In this particular case the diagnosis of pulmonary tuberculosis is
likely. But it seems to me arbitrary to interpret as such the phthfsis that, according
to Diogenes Laertius, IV, 20, made off with the Platonic philosopher Polemon of
Athens in his declining years.
79. Aelianus, Prag') 99; Suda s.v. phthoe.
80. This iconographic representation of consumption is only known to us from
writen evidence (Suda s.v. Theop6mpos).In this connection, see E. and L. Edelstein,
Asclepius (Baltimore, 1945), 1: 262ff. According to Pausanias, in the sanctuary of
Delphi there was a "sculpture in bronze of a chronic invalid" who was "utterly
wasted away and reduced almost to the state of a skeleton" (Descriptio Graeciae) x,
2, 6). See H. Pomtow, "Delphische Neufunde. III. Hippokrates und die Asklepi-
aden in Delphi," I(lio 15 (1917): 303-38, and F. Chamoux, "Perdiccas," in Hommages
aAlbert Grenier (Collection Latomus), 58 (1962): 386-96. In 1844, a bronze statuette
presenting an emaciated youth was found in the Aisne, near Soissons. Long inac-
cessible, it was finally acquired by a public collection (Dumbarton Oaks, Research
Library, no. 47.22). A Roman copy of a Greek work from the Hellenistic era, its
withered body, emaciated face, and feverish stare are presented with great skill.
The inscription "PERDIK ... " on the base of the drapery leads one to believe
that the person wasting away could be Perdiccas, sick with love. In that case the
statue is a pendant to the Delphic ex voto. However, G. M. A. Richter maintains
that the inscription alludes to Perdrix, a lame Athenian merchant taunted by
Aristophanes (Catalogue of Greek and &man Antiquities in the Dumbarton Oaks
Collection (Cambridge, 1956), p. 32). Is this tuberculous consumption, pining away
from love, or clubfoot with accompanying decline? The three diagnoses are not
mutually exclusive, but it is, unfortunately, impossible to consider the statue a
deliberate iconographic representation of pulmonary consumption. Since this dis-
ease is not marked by any external pathognomonic stigma, it is not enough to see
an artistic image of a withered person in order to consider it a representation of
tuberculosis. Likewise doubtful are the emaciated invalid on the Roman mosaic
from Lambridi (Gsell Museum in Algiers) and the Hellenistic statue of the so-
called "young consumptive" from the Fayum (Cairo Museum). For the latter, see
P. Perdrizet, "Le mort qui sentait bon," Annuaire Inst. Philol. Bruxelles 2 (1934):
719-27.
81. Itimata) case 33 (Inscr. Graecae) 1v 2 , 122). See R. Herzog, Die W underheilungen
von Epidauros (Leipzig, 1931), p. 106, and Edelstein and Edelstein, Asclepius (above,
n. 80), l: 22 7.
82. Celsus, De medicina) 111, 22.
83. Pliny the Younger, Epistolae) v, 19, 6. This treatment of pulmonary tuber-
culosis by change of climate may explain a strange find: the corpse of a young girl
of the Roman aristocracy that was mummified in Egypt during the second century
and exhumed in 1964 at the crossroads of the Via Cassia and the Via Grottarossa
in Rome. See U. Scamuzzi, "Studio sulla Mummia di bambina cosidetta Mummia
di Grottarossa," Riv. Stud. Class 12 (1964): 264-80. An appropriate paleopatho-
logical examination of the mummy could confirm or refute the hypothesis.
84. Pliny the Younger, Epistolae) VII, 9. See in this regard E. F. Leon, "A Case
of Tuberculosis in the Roman Aristocracy at the Beginning of the Second Cen-
tury," four. Hist. Med. 14 (1959): 86-88.
85. The term "tuberculosis" was invented only in 1832 (by J. L. Schonlein).
First adopted in German, the word did not enter French medical literature until
1854. Although Laennec was the pathfinder in this slow trek toward the unity of
tuberculosis, his anatomoclinical intuition did not really prevail until the advent
of bacteriology.
86. Aph°) VI, 46 (Littre, IV, 574).
87. Art°) III, 41 (Littre, IV, 178-80); my translation follows, except in one detail,
that of E.T. Withington, trans., Hippocrates) vol. 3 (Cambridge and London, 1928;
rept. 1948), Joints) 41, pp. 279-81. Where he renders the word pr6phasis "origin," I
prefer "first appearance."
88. See R.H. Major, "How Hippocrates Made His Diagnoses," Intern. Record
Med. 170 (1957): 482; M. Michler, "Die Kriippelleiden in De morbo sacro und De
articulis," Sudhofft Arch. Gesch. Med. 45 (1961): 321-23, and Coury, Grandeur et
diclin (above, n. 1), p. 62.
89. Not all the hunchbacks represented in works of art from antiquity are
tuberculous. The diagnosis can only be considered if the hunchback is at right
angles. Unfortunately, that is not enough. I have already stressed that an angular
hump on a sculpted or painted figure does not secure the diagnosis of vertebral
tuberculosis. However, such a diagnosis becomes plausible when the prevalence of
tuberculosis in the society is proven independently. The Louvre has several Helle-
nistic terra cotta torsos fron1 Smyrna that present angular kyphoses (inv. D 1177,D
1214, D 1216, D 1223, and so on). In the Greco-Roman Museum in Alexandria,
there are oil-lamps and a statuette representing hunchbacks. The British Museum
owns an ivory statuette, a Greek work of the first century, that represents a man
with Negroid features whose spine is very curved (Townley Coll. 1594-4.15.2). Also
worth mentioning is an Etruscan mirror with a drawing of a hunchbacked, emaci-
ated man on it (Tarquinia Museum).
90. Aph.) VI, 35 (Littre, IV, 572).
91. Epid._,VII, 19 (Littre, V, 390-92).
92. For the general appearance of this epitaph, its historical vicissitudes, and a
translation into French, see D. Gourevitch, "Une observation pediatrique pour
epitaphe," Echo medical 26, no. 145 (1968): 14. The diagnosis of tuberculosis is well
defended by B. Meinecke, "A Quasi-autobiographical Case History of an Ancient
Greek Child," Bull. Hist. Med. 8 (1949): 1022-31. The most recent study, which
includes a very careful medico-historical analysis and abundant bibliography, is the
thesis of H. D. Klitsch, Eine inschriftlicheI(rankengeschichtedes 3. Jht. n. Chr.: Das
Grabgedichtfur denfunfjtihrigen Lucius Minicius Anthimianus (Erlangen and Nurem-
berg, 1976).
93. Gland._,7-8 (Littre VIII, 562, ed. Joly (Paris, 1978). pp. 117ff.
94. See Bourgey, Observationet experience(above, n. 33), p. 73.
95. Pron-h._, II, 11(Littre, IX, 32). In this connection, see S. Ghinopoulo, Ptidia-
trie in Hellas und Rom (Jena, 1930), pp. 17 and 82ff.
96. Praen. Coacae)502 (Littre, v, 700).
97. Aph ._,III, 26 (Littre, IV, 498).
Chapter Eight: Leprosyand Tuberculosis

1. The conclusions of this chapter were presented to the Third European Col-
loquium of the Paleopathological Association at Caen (1980).
2. For more complete information, see C. Gernez-Rieux et al., "Les mycobac-
terioses humaines," Actes du XVI' CongresNat. de la Tubercuwse(Bordeaux, 1970),
pp. 1-70.
3. A. Cockburn, The Evolution and Eradicationof Infectious Diseases)(Baltimore,
1963), pp. 38, 73, 219-20.
4. Ibid., pp. 73 and 221ff.
5. R. Hare, "The Antiquity of Diseases Caused by Bacteria and Viruses," in
D. Brothwell and A. T. Sandison, eds., Diseasesin Antiqui-ty (Springfield, 1967),
pp. 115-31(esp. p. 127).
6. Ibid., p. 117.
7. The oldest mention of it occurs in Aristotle, Parva naturalia) 4786, and Hist.
anim.) 604a. See chapter 7, n. 52. More recent descriptions that also happen to be
more secure are given by Latin authors, especially Columella, De re rustica) VI, 14,
and Vegetius, Muwmedicina) II, 45. I should also mention the evidence from
Aretaeus of Cappadocia (Signa chron.) II, 13) on the frequency of tubercles in the
flesh of sacrificial animals. This important observation is usually misunderstood
and neglected because Aretaeus inserts it in his chapter on leprosy. For the general
history of this disease, see J. Francis, Bovine Tuberculoses(London, 1947).
8. See K. J. Donham and J. R. Leininger, "Spontaneous Leprosy-like Disease
in a Chimpanzee," ]our. Infect. Dis.) 136(1977): 132-36.
9. I refer to the situation in the nineteenth century and not to current Greece,
since it is especially instructive to look at epidemiological conditions prior to the
introduction of chemotherapy, sanitary techniques, massive urbanization, and the
admixture of populations, all of which nowadays upset the "natural history" of
infectious diseases.
10. See above, chapter 6, n. 86.
n. C. Stephanos, "Grece. Geographie medicale," in Dictionnaire des sciences
midicales)ed. A. Dechambre (Paris, 1884), 4th ser., 10: 531.
12. For the existence of a natural resistance factor (probably of genetic origin)
for the leprosy bacillus, see A. Rotberg, "Resistance et lepre," Bull. Ass. Lepr.
Lang. Franr;.(1968), pp. 141-44.
13. See J. Poirier, Recherchessur les reactionsa la lipromine en milieu lipreux et non
lepreux) thesis (Paris, 1944); A. Basset, "La lepre, son interet clans l'etude de
l' epidemiologie et la pathologie generale,'' ConcoursMedical 93 (1971):5627-37 (esp.
pp. 5628ff.).
14. See especially J. Lowe and F. McNulty, "Tuberculosis and Leprosy: Im-
munological Studies," LeprosyReview24 (1953):61-90, and H. Floch, "La reaction
de Mitsuda rendue positive par une primo-infection tuberculeuse est-elle accom-
pagnee d'une immunite relative anti-lepreuse?" Bull. Soc. Path. Exot. 47 (1954):
771-75.
15. R. Chaussinand, "Tuberculose et lepre, maladies antagonistes. Eviction de
la le pre par la tuberculose, ]our. of Leprosy 16 (1948): 431-38, and ''Quelques re-
marques concernant la theorie de l'antagonisme entre tuberculose et lepre," Acta
-iropica21 (1964):82-87.
16. See M. D. Grmek, "Preliminaires d'une etude historique des maladies,"
Annales E.S. C. 24 (1969): 1478.
17. Starting in 1964, the W odd Health Organization organized a trial vaccina-
tion of children against leprosy with BCG in Burma. By the end of March 1973,the
trial had covered 28,220 children below the age of 14. During this period (1964-73)
there were 768 new cases of leprosy recorded in the control population, of which
343 were in children who were vaccinated and 425 in children who were not. So
the protective effect of vaccination was about 20 percent. The highest degree of
protection (38 percent, which is a statistically significant number in this context)
was observed in children not older than 4 years of age at the beginning of the trial.
In other age groups, the degree of protection was not statistically significant. The
official report concludes that ''it seems unlikely that BCG vaccination can modify
the current or future manifestations of leprosy in similar (to Burma) regions." See
L. M. Bechelli et al., "BCG Vaccination of Children against Leprosy," Bull.
WH.O. 51(1974): 93-99.
18. See for instance Basset, "La lepre" (above, n. 13),pp. 5632ff., and P. Harter,
Precisde la leprologie(Paris, 1968), p. 38.
19. D. L. Weiss and V. M~ller-Christensen, "An Unusual Case ofTuberculosis
in a Medieval Leper," Dan. Med. Bull18 (1971):n-14. G. A. Hansen himself found
tuberculosis to be the most common cause of death among leprosy patients in
nineteenth-century Norway. It may be that tuberculosis did protect some people
from leprous infection at the same time as it killed the lepers themselves. W. H.
Jopling examined 500 leprous patients and none had a history of tuberculosis
before contracting leprosy; 3 of the patients contracted it afterward ("Clinical
Aspects of Leprosy," Tubercle63 [1982]: 295-305).
20. See R. and E. Blum/Health and Healing in Rural Greece(Stanford, 1965).
21. In 1938,when the general mortality in Greece had already gone down to 13.3
(it was around 30 in previous decades), the figures for officially declared causes of
death were, for every 1,000 deaths that year, 150.7 from "pneumonia," 70.3 from
tuberculosis of the respiratory tract, 52.8 from infantile diarrhea and enteritis, and
30.1 from malaria. In 1948, among the inhabitants of Athens, respiratory tubercu-
losis was officially the cause of 180 of 1,000 deaths over the year. It is worth adding
that at the same period and in the same population, extrapulmonary tuberculosis
had 25 victims for every 1,000 deaths.
22. R. and J. Dubos, The White Plague: TubercuwsisJMan) and Society (Boston,
1953).
23. M. Lurie, Resistance to Tubercuwsis: Experimental Studies in Native and Ac-
quired Defensive Mechanisms (Cambridge, 1964).
24. M. Castels et al., "Les bacilles tuberculeux de type africain," Rev. Tub. et
Pneum. 32 (1968):179.
25. See W. Hennig, Phywgenetic Systematics (Urbana, 1966); D. L. Hull, "Con-
temporary Systematic Philosophies,'' Ann. Rev. Ecol. Syst. 1, (1970): 19-54; D.
Guinot, "Examen des theories actuelles de la classification zoologique," Hist. Phil.
Life Sci. 1, (1979): 119-38(contains an important bibliographic essay).
26. J. Grober, "Zur Urgeschichte der menschlichen Tuberkulose," Med. J(lin.
49 (1954): 670-73.
27. This is a probable, not a certain, conclusion. The presence of a disease on
both sides, that is, in the Old and the New World, constitutes a real proof of its
prior origin, but the absence of its existence on the American side only supports a
presumption of its posterior origin. Some germs could have existed in the Old
W odd but not have been transmitted as such onto the American continent by
prehistoric migrants. Drawing attention to this state of affairs, Stewart has devised
the concept of a '' cold screen'' that would explain the filtering out of germs during
their journey through Siberia and Alaska. See T. D. Stewart, "A Physical Anthro-
pologist's View of the Peopling of the New World," South-west Journ. Anthrop. 16
(1960): 259-73. Nevertheless I note that a "cold screen" could not affect the
mycobacteria.
28. Tuberculosis does not seem to affect nomads; it is present in an endemic
state only among sedentary peoples. See A. J. Perzigian and L. Widmer, "Evi-
dence ofTuberculosis in Prehistoric Populations," four. Amer. Med. Ass. 241 (1979):
2643-46.
Chapter Nine: The Harm in Broad Beans

1. This chapter was originally published in History and Phiwsophy of the Life
Sciences2 (1980): 61-121;it appears here with additions and some modifications.
2. This formulation of the prohibition against broad beans is preserved in
several authors from antiquity, for instance, Plutarch, De educatione puerorumJ 17;
Diogenes Laertius, Vitae phiwsophorumJ VIII, 23; Porphyry, Vita PythagoraeJ 44;
Iamblichus, Vita PythagoricaJ 109. For the general meaning of the Pythagorean
sumbolaJ see Iamblichus, Vita) 103-5; I have used L. Deubner's critical edition
(Leipzig, 1937)and the revised and annotated edition of M. von Albrecht (Zurich
and Stuttgart, 1963).
3. The chapters on Pythagoras, Empedocles, and the Neopythagoreans in H.
Diels and W. Kranz, Die Fragmente der VorsokratikerJ 6th ed. (Berlin, 1951),are
indispensable. As for modern publications on the personality and achievement of
Pythagoras, the historical reality of the brotherhood at Croton, and the roots of
Neopythagoreanism, I would mention especially A. Delatte, Etudes sur la litterature

pythagoricienne(Paris, 1915);I. Levy, Recherchessur tessourcesde la legendede Pythagore
(Paris, 1926); K. von Fritz, PythagoreanPoliticsin Southern Italy (New York, 1940);
W. Burkert, Weisheit und WissenschaftJStudien zu Pythagoras)PhilolaosJund Platon
(Nuremberg, 1962) (Lore and Sciencein Ancient PythagoreanismJtrans. E. L. Minar
[Cambridge, 1972]); C. J. de Vogel, Pythagoras and Early Pythagoreanism (Assen,
1966); and B. L. Van der Waerden, Die PythagoreerJreligioseBruderschaftand Schute
der Wissenschaft(Zurich, 1979).
4. The basic doxography on the subject was gathered by Aulus Gellius, Noctes
AtticaeJ IV, II, 1-12.
5. See A. de Candolle, Origine des plantes cultivees(Paris, 1883),pp. 253-57.Ac-
cording to research by the school of N. I. Vavilov in Leningrad, the cradle of the
broad bean is West Asia (V. S. Muratova, 1931).
6. Iliad XIII. 589.
7. Theophrastus, De causisplantarumJ IV, 14.
8. The author of De mulierum affectibus uses the broad bean (specifically, the
seed) as a unit of size to specify the dosage of medications. He is at pains to make
it clear that he is speaking of a "Greek seed" (Mul.J 46; Littre, VIII, 106) in one
case and of an "Egyptian broad bean" in the other (Mul.J 181;Littre, VIII, 364).
The "Egyptian broad bean" is also mentioned in Acut. (sp.)J 53 (ed. Joly, 92). I
disagree with the opinion of Littre who, following the Renaissance commentators
and acceding to the authority ofJ. H. Dierbach, thinks that certain attestations of
kuamos without modifiers in the Hippocratic treatises designate the "Egyptian
broad bean" (identified with Nymphaea nelumbo).
9. Dioscorides, Materia medicaJII, 105 and 106. For the modern identification
of the plants at issue, see J. Berendes, Des Pedanios DioskuridesArzneimitteltehre
(Stuttgart, 1908), p. 209.
10. This opinion is still maintained by J. H. Dierbach, Die Arzneimittel des
Hippokrates (Heidelberg, 1824), pp. 2off. See the discussion in A. Benedicenti,
MalatiJ medici efarmacisti (Milan, 1947), 1: 93ff.
II. Pliny, Natural Histo'IXXVIII, 30, 12. See J. Andre, Lexique des termesde bota-
nique en latin (Paris, 1956), s.v. Faba.
12. See for example "Pythagore et les haricots," Chronique medicate27 (1920):
244ff.; 28 (1921):185and 375-77; 33 (1926): 184-87; 34 (1927): 88-90 (contains the
question of a doctor from Alger on the real hygenic value of the Pythagorean
precept according to which "one must abstain from beans" and the answer of Dr.
A. Lebeaupin to the effect that it cannot be beans, followed by a discussion on
the dangers sometimes posed by the consumption of common broad beans and
also of toxic European beans).
13. Chronique medicate33 (1926): 185-87.
14. See for example Dioscorides, IV, 68.
15. Empedocles, fr. 141 (D-K I, 368); text transmitted by Aulus Gellius, Noctes
AtticaeJ IV, II, 9.
16. Callimachus, fr. 126 Pfeiffer; text transmitted by Aulus Gellius, NoctesAtti-
caeJIV, II, 2.
17. Cicero, De divinationeJI, 62.
18. Iamblichus, Vita Pyth'J 109.
19. Diogenes Laertius, VIII, 19; see also Suda s.v. Pythagoras.
20. To be found in Diogenes Laertius, VIII, 33.
21. Artemidorus, OnirocriticonJI, 68.
22. Porphyry, De abstinentiaJ IV, 16; see also Diogenes Laertius, VIII, 33.
23. Pausanias, DescriptioGraeciaeJI, 37, 4. Also according to Pausanias, in Ar-
cadia it was said that Demeter provided all vegetables except broad beans (Descriptio1
GraeciaeJVIII, 15,1).
24. Gregory ofNazianzus, OrationesJXXIII, 535;Rufinus, In Greg. Naz. Orat.J
IX, 10; Didymus according to GeoponicaJII, 35, 8; Plutarch, Symp°JII, 3, 1. See
OrphicorumfragmentaJ fr. 291 Kern.
25. GeoponicaJII, 35, 8.
26. See W. Rathmann, Quaestiones Pythagoreae Orphicae EmpedocleaeJthesis
(Halle, 1933),and K. Kerenyi, Pythagorasund Orpheus) 3d ed. Zurich, 1950).
27. L. von Schroder, "Das Bohnenverbot bei Pythagoras und im Veda," Zschr.
I(unde Morgen/and 15 (1901): 187. This opinion is qualified by several more recent
authorities, for instance A. C. Andrews, "The Bean and Indo-European Totem-
ism," Amer. Anthrop. 51(1949):274-92.
28. See de Candolle, Plantes cultivies (above, n. 5), p. 254. Admittedly, in the
Susrutasarrihita(Sii:trasthanaJXLVI, 19-20) there is a discussion of the dietetic prop-
erties of several kinds of shimvaJ a term usually translated "broad bean," but the
seeds in question come from other legumes than Viciafaba.
29. Herodotus, II, 37 (trans. A. de Selincourt, rev. A. R. Burn [Harmonds-
worth, 1972; rept. 1976], p. 144). Some similar information probably based on this
can be found in Diodorus Siculus, I, 89; Plutarch, Symp.JVIII, 8, 2 and De Is. et
Osir°Js; Porphyry, De abstinentiaJII, 25.
30. For specimens discovered in tombs and for the mention of the broad bean
in hieroglyphic texts, see V. Loret, Laflore pharaoniqueJ2d ed. (Paris, 1892), p. 93.
A significant quantity of broad beans was offered by Ramses III to the priests of
Memphis and Heliopolis. The botanical determination of the seeds found by
archaeologists is not at issue, but there is controversy about the precise meaning
of the Egyptian term iwr.f.tJ which is usually translated "broad bean." Recent
lexica (Erman-Grapow, Cerny, Charpentier) identify the term not with Greek
kuamos but instead with dolikhos 'calavance' (Vigna sinensis Endl.). There is no
Egyptian text that forbids the eating of a legume or pronounces one harmful.
31. Isocrates, BusirisJ28; Herodotus, II, 123.
32. Herodotus, II, 123(trans. Selincourt, p. 178); see also II, 81, and IV, 95.
33. According to Origenes, Contra CelsumJv, 41.
34. According to Aul us Gellius, NoctesAtticaeJ IV, 11,12.
35. Tertullian, De animaJ 31,asserts that Pythagoras forbade his disciples to walk
across a field of broad beans.
36. Diogenes Laertius, VIII, 45. See also Anthologia PalatinaJ VII, 122.
37. Diogenes Laertius, VIII, 39. Diogenes also relates a similar story from Her-
mippus to the effect that Pythagoras was killed by Syracusans unhappy with the
help he had given the Agrigentans. The details that interest us are not changed in
this variant; death results from the fact that the philosopher, while fleeing, "ar-
rived at a field of broad beans that he did not wish to cross.'' See also Suda) s. v.
Pythagoras.
38. Iamblichus, Vita Pyth.J 189-94.
39. This text comes from a lost work by Aristotle himself or one of the collab-
orators to whom he entrusted the task of studying the Pythagorean tradition. It is
cited from Diogenes Laertius, VIII, 34.
40. Iamblichus, Vita Pyth.J 109.
41. Plutarch, Symp.JII, 3, 1, and Clement of Alexandria, StromataJ III, 3.
42. Pliny, Natural HistoJX XVIII, 118-19.The offerings for the dead mentioned
by Pliny were made during the archaic rites of the Lemuria.
43. Lucian, Somnium (Gallus)) 5-6.
44. Lucian, Vitarum auctioJafter the French translation by L. Humbert.
45. Geoponica)II, 35, 6. The same information can be found in Pliny, who speaks
of the "mournful letters on the flower of the broad bean" (Natural Histo-;x XVIII,
II9).
46. Porphyry, Vita Pyth.J 44.
47. Plutarch, De educationepuerorum) 17.
48. Iamblichus, Vita Pyth.J 206.
49. Aul us Gellius, NoctesAtticae) IV, II, 4-5.
50. Aulus Gellius, NoctesAtticae) IV, II, 10.
5r. Artemidorus, Onirocriticon)I, 68.
52. Cicero, De divinatione) I, 62; see also Aulus Gellius, NoctesAtticae) IV, n, 3.
53. Pliny, Natural Histo-;x XVIII, II8.
54. Diogenes Laertius, VIII, 23.
55. Geoponica)II, 35, 3-4. See also Tertullian, De anima) 48.
56. Clement of Alexandria, Stromata) III, 3. In the Geoponica)the broad bean is
accused of making livestock infertile.
57. The context of this phrase is something as follows: "When will I be served,
along with vegetables dressed in greasy lard, this bean, Pythagoras's sister?" (Hor-
ace, Satires) II, 6, 63). The Latin word cqgnatadoes not just mean "sister" but also
and always specifies a blood relation.
58. See J. Jouanna, "Presence d'Empedocle clans la Collection hippocratique,"
Bull. Assoc. Guill. Budi 44 (1961):452-63.
59. &gimen) II, 45, 1 (trans. W. H. S. Jones, Hippocrates(Cambridge and Lon-
don, 1943), 4: 315).
60. Acut. (sp.)J 64 (Littre, II, 518), and Epid.J II, 6, 7 (Littre, v, 134). In these
cases it is the broad bean as such that is recommended as medicine. Other Hippo-
cratic passages stipulate the therapeutic use of the "Egyptian" bean, that is, the
seeds of the pink lotus.
6r. Dioscorides, II, 105.
62. Epid'J VII, 82 (Littre, V, 436-38).
63. Acut. (sp.)J47 (Littre, 484-86; Joly, 89).
64. Epid'J II, 4, 3 (Littre, v, 126), and Epid.J VI, 4, II (Littre, v, 310). My
translation is perceptibly different from Littre's. I wish to thank J. Jouanna for
collating this passage in the manuscripts of the Hippocratic corpus.
65. For the subsistence crisis in Ainos, see J. M. F. May, Ainos) Its History and
Coinage (Oxford, 1950). For the dating of Epidemics II-IV-VI, see chapter 12, the
section "The Author, Date, and Locale of this Epidemiological Report."
66. Galen, In Hippocratis libros I et II Epid. commentaria) ad foe. The passage
concerning Ainos is known only in the Arabic translation; see the edition of E.
Wenkenback and F. Pfaff, Corpus Medicorum Graecorum) v, 10, 1 (Leipzig and
Berlin, 1934), pp. 338-40. See also the commentary on the parallel passage in Corp.
Med. Graec.Jv, 10, 2, 2, ed. Wenkenbach and Pfaff (Berlin, 1956), p. 219.
67. See Dierbach, Arzneimittel (above, n. 10), pp. 41ff., de Candolle, Plantes
cultivees (above, n. 5), pp. 85ff.; and the Liddell-Scott-Jones Greek-EnglishLexicon
s.v. One should not follow R. Joly in his edition of the treatise &gimen. He rejects
Littre and translates 6robos(II, 45, 2) by the French word vesce)a term with too
broad a meaning if it applies to all plants of the genus Vicia) or too imprecise if it
is limited to common vetch (Vicia sativa). The identification of 6roboswith bitter
vetch is supported by descriptions in Theophrastus and Dioscorides.
68. Dioscorides, II, 108.
69. Pliny, Natural HistOJXXXII, 153.
70. Galen, In HippocratislibrosI et II Epid. commentaria (above, n. 66), p. 339.
71. A. Cantani, "Latirismo (Lathyrismus) illustrato da tre casi clinici," Il Mor-
gagni 15(1873):745-95.
72. See J. C. Huber, "Historische Notizen iiber den Lathyrismus, "Friedrich)s
Blatter fur gerichtliche Medizin (1886),pp. 34-36, and R. H. Major, "How Hippoc-
rates Made His Diagnoses," Intern. Ree. Med. 170 (1957):481.
73. B. Schuchardt, "Zur Geschichte und Casuistik des Lathyrismus," Dtsch.
Arch. J(lin. Med. 40 (1886-87): 312-41.
74. See J. Borg, G. Mazars, and B. Sacko, "Apropos de la neurotoxicite de
Lathyrus sativus, plante alimentaire et medicinale de l 'Inde,'' Les medecines -traditi-
onellesde t>Asie (Actes du Colloque de Paris, 1979) (Strasburg, 1981),pp. 103-10.
75. J. Andre, L)alimentation et la cuisine a Rome (Paris, 1961), p. 38 (new ed.
[Paris, 1981],p. 37).
76. Pliny, Natural HistOJJyxvrrr, 103.
77. Anaxandrides, fr. 41, 43; Alexis, fr. 162, 12.
78. Particularly an amino acid, diaminopropionate. See E. D. Schilling and
F. M. Strong, "Isolation, Structure, and Synthesis of a Lathyrus Factor from L.
odoratus," ]our. Amer. Chem. Soc. 76 (1954):2848; rept. in Nutr. Rev. 23 (1976): 242.
See also Borg, Mazars, and Sacko, "A propos de Lathyrus" (above, n. 74); E.
Massa, "An Overview of Lathyrism," Rev. Neurobiol. 18 (1972): 181-206; and I. E.
Liener, ed., Toxic Constituents of Plant RJodstuffi (New York, 1980), pp. 239-63.
79. See M. Streifler and D. F. Cohn, "Chronic Central Nervous System Tox-
icity of the Chickling Pea (Lathyrus sativus)," Clin. Toxicology18 (1981):1513-17.
So. "Favas verdes produsindo ictericia," &vista Universal Lisbonese (1843),p. 515.
81. A. Mina La Grua, Sopra l)itterizia endemica e su le malattie ordinarie dei conta-
dini di Castelbuono (1856);cited and discussed by G. Sansone, A. M. Piga, and G.
Segni, Il favismo (Turin, 1958),pp. 9ff.
82. See Sansone et al., Ilfavismo (above, n. 81), pp. 1off.
83. G. Montano, "Del favismo o intossicazione fabacea," Atti del XI Congr.
Med. Internationale (Rome, 1894), 3: 301-4.
84. See Mule Bertolo, "Zafara, o itterizia particolare prodotta dalle particelle
odorifere della pianta fava," ltatica del medico (1901) (this article was originally
published in 1873in a local political journal of Caltanisetta).
85. In Sansone, Piga, and Segni, Ilfavismo (above, n. 81), p. 11.
86. Montano, "Del favismo" (above, n. 83), p. 304; see also Sansone, Piga,
and Segni, Ilfavismo (above, n. 81), p. 14.
87. Especially worth citing are the research of Fermi in Sardinia (1905)and the
critical review of prior research undertaken by A. Gasbarrini, "Su di una forma
ancora insufficientemente conosciuta di anemia acuta febbrile con itterizia ed emo-
globinuria (il favismo)," RJlia Clin. Chim. Microscop. (Salsomaggiore) 4 (1912-14):
374-89.
88. This acid was to have been liberated into the organism from "cyanogenetic
glucosides" that were supposed to exist in broad beans. See M. R. Marquet,
Intoxications alimentaires par certaines ligumineuses: Gesses)feves) haricots cyanogenetiques
(Paris, 1944). In reality, cyanhy~dric poisoning has only been reported after the
ingestion of certain other beans (for example, Phaseolus lunatus). The hypothesis of
this toxic mechanism as the fundamental phenomenon in favic idiosyncrasy was
put forth by analogy, not as the result of direct observations on the broad bean
itself.
89. A. Turchetti, "Forme poco frequenti di emoglobinuria da farmaci in corso
di infezione malarica," llifonna medica 62 (1948): 325-28; this article is cited as a
precursor by E. Beutler, ''The Hemolytic Effects of Primaquine and Related Com-
pounds," Blood 14 (1959):103-39.
90. P. E. Carson et al., "Enzymatic Deficiency in Primaquine Sensitive Eryth-

rocytes,'' Science 124 (1956):484ff.
91. See W. H. Crosby, "Favism in Sardinia (Newsletter)," Blood 11(1956):91ff.;
A. Szeinberg et al., "Studies on Erythrocytes" Blood 12 (1957): 603-13; W. H.
Zinkham et al., ''A Deficiency of Glucose-6-phosphate Dehydrogenase Activity in
Erythrocytes from Patients with Favism," Bull. Johns Hopkins Hosp. 102 (1958):169-
75; P. Larizza et al., "L'individualita bioenzimatica dell'eritrocita favico: Sopra
alcune anomalie biochimiche ed enzimatiche delle emazie nei pazienzi affetti da
favismo e nei loro familiari," Haematologia 3 (1958):251-59;G. Sansone and G. Segni,
"Nuovi aspetti dell'alterato biochimismo degli eritrociti favici: Assenza pressoche
completa della glucosio-6-P-deidrogenasi," Boll. Soc. Ital. Biol. Sper. 15 (1958):327-
29; U. Carcassi, Eritroenzimopatie ed anemie emolitiche (Pisa, 1959).
92. K. L. Roth and A. M. Frumin, "Studies on the Hemolytic Principle of
the Fava Bean," ]our. Lab. Clin. Med. 59 (1960): 695-700. See J. Ducas, Le deficit
en glucose-6-phosphatedishydrqgenase) thesis (Nancy, 1961).
93. J. Y. Lin and K. M. Ling, "Studies on Favism: 1. Isolation of an Active
Principle from Fava Beans," four. Rmnosan Med. Ass. 61 (1962): 484-89. For new
research on these substances, see E. Bottini et al., "Presence in Vicia faba of Different
Substances with Activity in Vitro on Gd - Med. Red Blood Cell Reduced Gluta-
thione," Clin. Chim. Acta 30 (1970): 831-34; and J. Jamalian, "Favism-inducing
Toxins in Broad Beans (Vicia faba)," Plant modsfor Human Nutrition 27 (1977):207-
11.
94. R. Lederer, "A New Form of Acute Hemolytic Anaemia: Baghdad Spring
Anaemia," Trans. Roy.Soc. Trop. Med. Hyg. 34 (1941):387-94.
95. See Ducas, Le deficit (above, n. 92), and A. Orsini et al., "Le favism," Sem.
Hopit. (Paris, 1961),pp. 557-70.
96. S. H. Boyer et al., "Electrophoretic Heterogeneity ofGlucose-6-phosphate
Dehydrogenase and Its Relationship to Enzyme Deficiency in Man," It-oc. Nation.
Acad. Sci. USA 48 (1962): 1868.
97. H. N. Kirkman et al., "Functionally Abnormal Glucose-6-phosphate De-
hydrogenases," Cold Spring Harbor Symp. Quant. Biol. 29 (1964): 391.
98. P. R. McCurdy et al., "A Chinese Variant of Glucose-6-phosphate Dehy-
drogenase," four. Lab. Clin. Med. 67 (1966): 374.
99. A. Yoshida, G. Stamatoyannopoulos, and A. G. Motulsky, "Biochemical
Genetics of Glucose-6-phosphate Dehydrogenase Variation,'' Ann. N. r Acad. Sci.
155(1968): 868-79; E. Beutler, "Drug-induced Hemolytic Anaemia," Pharmacol.
Rev. 21 (1969):73-103; J. C. Dreyfus, "Bases moleculaires des anomalies enzyma-
tiques genetiques," Biochimie 54 (197?,):559-71; J. C. Kaplan, "Remarques sur les
enzymopathies genetiques du globule rouge," Biochimie 54 (1972): 765-73; J. C.
Kaplan, ''Defective Molecular Variants of Glucose-6-phosphate Dehydrogenase
and Methaemoglobin Reductase," four. Clin. Path. 27 (Suppl. Roy. Coll. Path. 8)
(1974): 134-41; and G. Schapira et al., Pathologie moliculaire (Paris, 1975).An excel-
lent, up-to-date overview by E. Beutler can be found in J. B. Stanbury, J. B.
Wyngaarden, and D.S. Frederickson, The Metabolic Basis of Inherited Diseases (New
York, 1978), pp. 1430-51.
100. B. Childs et al., "A Genetic Study of a Defect in Glutathione Metabolism
of the Erythrocite," Bull. Johns Hopkins Hosp. 102 (1958):21-37; E. Sartori, "On the
Pathogenesis of Favism," four. Med. Genet. 8 (1971):462-67; V. A. McKusick,
Mendelian Inheritance in Man) 3d ed. (Baltimore, 1971).
101. See G. Stamatoyannopoulos et al., "On Familial Predisposition to Fav-
ism," Amer. four. Hum. Genet. 18 (1966): 253-63; Sartori, "Pathogenesis" (above,
n. 100); E. Beutler, "Abnormalities of the Hexose Monophosphate Shunt," Sem.
Hemat. 8 (1971):311-47.
102. See Sansone, Piga, ·::ind Segni, Il favismo (above, n. 81); Ducas, Le deficit
(above, n. 92); Orsjni et al., "Le favisme" (above, n. 95); G. Lugassy, Le deficit en
glucose-6-phospho-deshydrqgenase) thesis (Paris, 1979), pp. 311-47.
103. S. A. Doxiadis et al., "Glucose-6-phosphate Dehydrogenase Deficiency, a
New Etiological Factor of Neonatal Jaundice," Lancet (1961),I, pp. 297-301.
104. See chapter 10.
105. See Sansone, Piga, and Segni, Il favismo (above, n. 81); J. Bernard and J.
Ruffiee, Hematologiegeqgraphique (Paris, 1966-72); M. A. Belsey, "The Epidemiol-
ogy ofFavism, Bull. WH.O. 48 (1973):1-13; Lugassy, Le deficit (above, n. 102); M.
Benabadji et al., "Heterogeneity ofG6PD Deficiency in Algeria," Hum. Genet. 40
(1978):177-84.
106. See the publications cited above by G. Sansone, G. Segni, P. Larizza, U. E.
Carcassi, M. E. Belsey, and others.
107. See C. Choremis, L. Zannos-Marioulea, and M.D.C. Kattamis "Fre-
quency of Glucose-6-phosphate Dehydrogenase Deficiency in Certain Highly Ma-
larious Areas of Greece," Lancet (1962), I pp. 17ff.; A. C. Allison et al., "Deficiency
of Erythrocyte Glucose-6-phosphate Dehydrogenase in Greek Populations," Ann.
Hum. Genet. 26 (1963):237-42; G. Stamatoyannopoulos et al., "The Distribution
of Glucose-6-phosphate Dehydrogenase Deficiency in Greece," Amer. ]our. Hum.
Genet 18 (1966): 296-308.
108. Doxiadis et al., "Neonatal Jaundice" above, n. 103).
109. See C. A. Kattamis, "Some Clinical and Biochemical Aspects ofFavism in
Childhood," Ann. Soc. Be{geMed. Trop. 49 (1969): 289-304. In the decade following
Carson's ground-breaking publication, more than 500 cases of favism were treated
in a large pediatric hospital in Greece. To all appearances, this "epidemic" reflects
changing medical knowledge and not nosological reality. The historian of medicine
should ask himself what went on beforehand in the diagnosis and treatment of
children with favism. By what means was this disease rendered "invisible"?
no. See C. G. Gasperini et al., "Osservazione sulle manifestazioni allergiche
locali e generali nel favismo ed il favismo nell 'Isola di Rodi," Gior. Ital. Mal. Esot.
Trop. 4 (1931):49, and C. A. Kattamis et al., "G6PD Deficiency and Favism in the
Island of Rhodes (Greece)," ]our. Med. Genet. 6 (1969): 286-91.
111. G. R. Fraser et al., "Thalassemia, Abnormal Hemoglobins, and Glucose-
6-phosphate Dehydrogenase Deficiency in the Arta Area of Greece," Ann. N. Y
Acad. Sci. 119(1964): 415-35. See also Stamatoyannopoulos et al., "Distribution in
Greece" (above, n. 107).
112. M. Siniscalco et al., "Favism and Thalassemia in Sardinia and Their Rela-
tionship to Malaria," Nature 190 (1961):1179ff.;A. C. Allison, "Malaria and Glu-
cose-6-phosphate Dehydrogenase Deficiency," Nature 197 (1963): 609; A. G.
Motulsky, "Hereditary Cell Traits and Malaria," Amer. ]our. Trop. Med. Hyg. 13
(1964): 147-58. See also chapter 10, below.
113. U. Bienzle et al., "Glucose-6-phosphate Dehydrogenase and Malaria,"
Lancet (1972), I pp. 107-10.
114. Lugassy, Le deficit (abov~, n. 102), pp. 60-61 and 64.
115. See for instance U. E. Carcassi, ''The Interaction between Beta-thalassemia,
G6PD Deficiency, and Favism," Ann. N. Y Acad. Sci. 232 (1974): 297-305.
116. Thanks especially to the work undertaken by G. Stamatayannopoulos et
al., "Genetic Diversity of the 'Mediterranean' Glucose-6-phosphate Dehydrogen-
ase Deficiency Phenotype," ]our. Clin. Invest. 50 (1971):1253-61.
117. For an overview of the humanists' commentaries, see C. T. Menke, De
Leguminibus Veterum (Gottingen, 1814),esp. the chapter entitled "Faba Pythagori-
cis Vetita." A reading of this particularly learned thesis in medicine highlights the
contrast between the abundance of traditional imaginings on this subject and the
paucity of newer ideas.
n8. K. Sprengel, Versuch einer pragmatischen Geschichte der Arzneykunde (Halle,
1792), vol. r. In the French edition of this work (Paris, 1815,1: 229) broad beans are
mistaken for beans (Phaseolus sp.); the error lies with the translator, not Sprengel
himself.
n9. A. B. Krische, De Societatis a Pythagore in Urbe Crotoniatarum Conditae Scopo
Politico Commentatio (Gottingen, 1830). See also C. Holk, De Acusmatis sive Symbolis
Pythagoricis) thesis (Kiel, 1894).
120. R. and E. Blum, Health and Healing in 'Rural Greece (Stanford, 1965),p. 78.
121. F. Lenormant, "Faba," in C. Daremberg and E. Saglio, Dictionnaire des
antiquitesgrecques et romaines (Paris, 1896), 2: 947.
122. E. Bourquelot, "Remarques apropos des feves de Pythagore," C. R. Soc.
Biol. (Paris) 56 (1904): 861ff.
123. J. G. Frazer, Totemism (Edinburgh, 1887);The Golden Bough (London, 1890).
124. E. Rohde, Psyche (Freiburg im Breisgau, 1898); trans. from the 8th ed. by
W. B. Willis (London and New York, 1925)and often reprinted.
125. See especially J. G. Frazer, The Golden Bough) 3d ed., (London, 19n), 1: n7-
19, 214, etc.
126. S. Reinach, Cultes) mythes) et religions (Paris, 1905).
127. Especially J. Larguier de Bancels, "Surles origines de la notion de l'ame a
propos d'une interdiction de Pythagore," Arch. Psycho!. 17 (1918): 58-66; A.
Lebeaupin "Pythagore et les haricots" (above, n. 12), pp. 244ff.; L. Piniatoglou,
"The Pythagorean Tabous," Acta Greek Anthrop. Soc. (1934), p. 36; M. Tierney,
"A Pythagorean Tabu,"Melanges Boisacq (Brussells, 1938),pp. 317-21;L. Bonuzzi,
"Ancora su Pitagora e il suo influsso sulla medicina greca," Acta Med. Hist. Pat.
15(1968-69): 9-16.
128. As examples, I cite H. E. Sigerist, History of Medicine (New York, 1961), 2:
96, and G. Sarton, History of Science (Cambridge, 1960), 1: 201.
129. A. Delatte, "Faba Pythagorae Cognata," Serta Leodiensia (Liege, 1930),
pp. 31-57.
130. M. Detienne, "La cuisine de Pythagore," Arch. Socio!. Rel. 15, no. 29
(1970): 141-62; Les jardins d)Adonis (Paris, 1972) (The Gardens of Adonis) trans. J.
Lloyd [Hassocks and Sussex, 1977]); Dionysos mis a mort (Paris, 1977) (Dionysos Slain)
trans. M. and L. Muellner [Baltimore, 1979]).
131. Reinach, Cultes) mythes) et religions (above, n. 126), 1: 44.
132. Ibid., 1: 43-48.
133. Larguier de Bancels, "Surles qrigines de l'ame" (above, n. 127), pp. 58-66.
A like opinion is maintained by Andrews, "The Bean and Inda-European" (above,
n. 27), p. 289, who finds the tabu's origin in the primitive experience of flatulence
(which is felt as the presence of souls in the broad beans) and not of totemism.
134. See E. Jones, "Die Empfangnis der Jungfrau Maria <lurch das Ohr," Jahrb.
Psychoanal. 6 (1914):135.
135. Delatte, "Faba" (above, n. 129), p. 3r.
136. Ibid., p. 56.
137. Detienne, Dionysos Slain) pp. 6off. and 85 (French ed., pp. 146-47, 192)
(above, n. 130).
138. Ibid., p. 86ff. (French ed., pp. 193-94). For the text of this cultic regula-
tion, see F. Sokolowski, Les lois sacreesde PAsie Mineure (Paris, 1955),no. 84. As for
mint, which is still a condiment for broad beans in modern Greece, see G. Daux,
"L'interdiction rituelle de la men the," Bull. Corresp. Hell. 81 (1957):1-5.
139. J. Schumacher, Antike Medizin) (Berlin, 1940), pp. 59-63.
140. S. Veras's account was published in Italian as "Il favismo era conosciuto
dai Greci antichi," in the book by Sansone, Piga, and Segni, II favismo (above, n.
81), pp. 5-7. Amo11g those who see a relationship between the Pythagorean tabu
and favism, I especially mention T.H.D. Arie, "Pythagoras and Beans," Oxford
Med. School Gaz. n (1959):75-81. See also E. Lieber, "Favism in Antiquity," I(oroth
5 (1970): 331-35,and "The Pythagorean Community as a Sheltered Environment
for the Handicapped," in Inter. Symp. Society, Medicine) and Law (Jerusalem) 1972)
(Amsterdam, 1973), pp. 33-41; and R. S. Brumbaugh and J. Schwartz, "Pythagoras
and Beans: A Medical Explanation" Class. Hiorld 73 (1980): 421ff.
141. See Arie, "Pythagoras" (above, n. 140), and the letter by H. A. Waldron
published in Brit. Med. ]our. (June 16, 1973).
142. M. Enrique Laval, "La prohibicion pitagorica de comer habas. Una nueva
explication?" Anal. Chi!. Hist. Med. n (1969): 79-97.
143. C. N. Ballas, "The Pythagorean Prohibitions," Acta Congr. Inter. XXIV
Hist. Artis Med. (Budapest, 1976), 2: 1343-46.
144. Diogenes Laertius, VIII, 2, 70. See 0. Bernhard, "Ueber Malariabekamp-
fung im klassischen Altertum," in Neubu1lJer)s Festschrift (Vienna, 1928), pp. 44-
46.
145. See Strabo, Geog..,VI, 1, 2, and 2, 4.
146. In their studies of the history of malaria in Greece, W.H.S. Jones and P.
Fraccaro stress the importance of the disease in southern Italy.
147. See Andre, L)alimentation (above, n. 75), p. 35.
148. Athenaeus, DeipnosophistaeJ54ff.
149. Pliny, Natural History; XVIII, 118;see Andre, L)alimentation (above, n. 75),
p. 36.
150. See Belsey, "Epidemiology ofFavism" (above, n. 105), pp. 1-13.
151. S. Katz, "Un exemple d'evolution bioculturelle: La feve," Communications
31 (1979): 53-69 (esp. p. 57).
152. Ibid., p. 66.
153. E. Giles, "Favism, Sex-linkage, and the Inda-European Kinship System,"
Southwest ]our. Anthrop. 18 (1962): 286-90.
154. See de Vogel, Pythagoras (above, n. 3), pp. 232-44, and Schumacher, An-
tike Medizin (above, n. 139), pp. 57-63.
155. Iamblichus, Vita Pyth.., 163.
156. Plutarch, Degenio SocratisJ580c. On the abandonment of the "miraculous"
by thinkers after Socrates and especially on Plutarch's rationality, see the spirited
discussion by H. Pourrat, Le sage et son demon (Paris, 1950).
157. "Rapport sur les conferences d'Histoire de la n1edecine et des sciences
biologiques," Annuaire de /JE.P.H.E.J IV section) pour [Jannie 1975-1976 (Paris, 1975),
p. 814.
Chapter Ten: PoroticHyperostosisJ

HereditaryAnemiasJ
and Malaria
1. Most of the material in this chapter was published in Annales E. S. C. (1975),
pp. 1152-85.It has been revised to take account of new research, in particular, an
inquiry by the Paleopathology Association (Detroit, 1977), a study of cribra orbi-
talia in Europe, an overview of Mediterranean research by J. L. Angel (1978), and
the work of M. M. Wintrobe (1980). The section on the history of malaria is new.
2. See C. Toldt, "Ueber Welcker's Cribra orbitalia," Mitt. Anthrop. Gesell.
Wien 16 (1886): 20, and H. Welcker, "Cribra orbitalia: Ein ethnologisch-diagnos-
tisches Merkmal am Schadel mehrerer Menschenrassen," Arch. Anthrop. 17 (1888):
1-18. In this connection it is interesting to read Virchow's anthropological lucubra-
tions, "Ueber einige Merkmale niederer Menschenrassen am Schadel, Abh. l+euss.

Akad. Wiss? Phys. I(l.) 2 Abth. (1875), pp. 1-130. Virchow knew of symmetrical
porous osteophytosis among American Indians; recent examination of a Peruvian
skull in his old collection in Berlin has confirmed the diagnosis.
3. B. Adachi, "Die Porositat des Schadeldaches," Zschr. Morph. Anthr. 7 (1904):
373; F. Wood Jones, "Report on Human Remains," Archeol. Survey of Nubia)
Report for 1907-1908 (Cairo, 1910), pp. 1-375.
4. J. Saint-Perier, "Lesions osseuses d'un squelette d'enfant trouve dans un
milieu gallo-romain," Bull. Mem. Soc. Anthr. Paris 5 (1914): 31-36.
5. A. Hrdlicka, "Anthropological Field Work in Peru in 1913; With Notes on
the Pathology of Ancient Peruvians," Smithsonian Inst. Misc. Coll. 61 (1914): 1-69
(esp. pp. 57-60). For a historical estimate of this research, see S. Jarcho, Human
Palaeopathology(New Haven and London, 1966), pp. 16ff.
6. H. U. Williams, "Human Paleopathology, with Some Original Observa-
tions on Symmetrical Osteoporosis of the Skull," Arch. Path. Labor. Med. 7 (1929):
839-902. The same parallel between symmetrical osteoporosis and familial anemia
is made in the article by B. F. Feingold and J. T. Case, "Roentgenologic Skull
Change in Anemias of Childhood; A Few Notes on Similar Findings among Skulls
of Peruvian Indians," Amer. ]our. Roentg. 29 (1933): 194-202.
7. S. Moore, "Bone Changes in Sickle Cell Anemia, with Note on Similar
Changes in Skulls of Ancient Maya Indians,'' ]our. Missouri Med. Ass. 26 (1929):
561-64.
8. E. A. Hooton, The Indians of PecosPueblo: A Study of Their Skeletal Remains
(New Haven, 1930), esp. pp. 306-30; see also "Skeletons from the Cenote of
Sacrifice at Chichen Itza, '' in The Maya and Their Neighbors (New York, 1940),
pp. 272-80.
9. L. Pales, Paliopathologie et pathologie comparative (Paris, 1930), p. 250 and
pl. 43, 2.
10. R. L. Moodie, RoentgenologicalStudies of Egyptian and Peruvian Mummies
(Chicago, 1931).
n. J. L. Angel, "Skeletal Changes in Ancient Greece," Amer. ]our. Physical
Anthrop. 4 (1946): 69-97.
12. For example, see R. Fraser, "The Problem of Osteoporosis: A Critical
Review," ]our. BoneJoint Su13. (1962), pp. 485-95.
13. H. Mueller, "Het voorkomen van de zgn. osteoporosis symmetrica cranii
op Java en haar verband met de rhachitis," Geneesk. TiJdsch. Neder. Indie 74 (1934):
1084-93.
14. J. L. Angel, "Porotic Hyperostosis, Anemias, Malarias, and Marshes in the
Prehistoric Eastern Mediterranean," Science153 (1966): 760-63.
15. In particular, see D. S. Carlson, G. J. Armelagos, and D. Van Gerven,
"Factors Influencing the Etiology of Cribra Orbitalia in Prehistoric Nubia," ]our.
Hum. Evol. 3 (1974): 405-10. In an examination of 285 skulls from Nubia, they
discovered just one case of the association of cribra orbitalia with porotic hyperos-
tosis (of the cranial vault). The former ailment was reported in 21.4 percent of the
specimens and the latter in o. 7 percent.
16. See especially W.G.J. Putschar, "Problems in the Pathology and Palaeo-
pathology of Bones," in Jarcho, Human Palaeopathology(above, n. 5), pp. 56-65.
17. Hrdlicka, "Anthropological Fieldwork" (above, n. 5), and Wood Jones,
"Human Remains" (above, n. 3).
18. Hooton, Indians (above, n. 8), and the critical remarks of Jarcho, Human
Palaeopathology(above, n. 5), p. 22.
19. See Mueller, "Het voorkomen van osteoporosis" (above, n. 13).
20. H. Hamper! and P. Weiss, "Ueber die spongiose Hyperostose am Schadel
aus Alt-Peru," VircfaowsArch. Path. Anat. 327 (1955):629-42.
21. Williams, "Paleopathology" (above, n. 6), p. 900.
22. Pales, Paliopathologie (above, n. 9), p. 250.
23. H. E. Sigerist, History of Medicine) vol. 1 (New York, 1951),p. 47.
24. T. B. Cooley and P. Lee, "A Series of Cases of Splenomegaly in Children
·with Anemia and Peculiar Bone Changes," Trans. Amer. Ped. Soc. 37 (1925): 29;
T. B. Cooley et al., ''Anemia in Children with Splenomegaly and Peculiar Changes
in the Bones," Am. ]our. Dis. Child. 34 (1927): 347-63. For the history of this
discovery, see W. W. Zuelzer, "Thomas B. Cooley (1875-1945),"]our. Pediatr. 49
(1956):642-50, and D. J. Weatherall, "Toward an Understanding of the Molecular
Biology of Some Common Inherited Anemias: The Story of Thalassemia," in M.
M. Wintrobe, Blood) Pure and Eloquent (New York, 1980), pp. 373-414.
25. See the bibliographical documentation in V. Chini and C. M. Valeri,
"Mediterranean Hemopathic Syndromes," Blood 4 (1949): 989-1013. Cooley de-
scribed what is now considered the homozygous, severe form of a hereditary
anemia particularly widespread in certain Mediterranean lands. Its discovery was
brought about with relatively unsophisticated technical means available to any
physician. How did it happen, then, that this anemia was recognized as a disease
suigeneris by American physicians before being detected by physicians in Greece or
Italy? Weatherall ("Thalassemia" [above, n. 24], p. 376) proposes a simple and
convincing, if paradoxical, explanation: Cooley noticed the disease in Detroit since
it was a rarity there and so represented something worth noting, while in Italy it
escaped medical perception precisely because it was too familiar.
26. For the formation of this nosological concept, see G. H. Whipple and
W. L. Bradford, '' Mediterranean Disease-Thalassemia (Erythroblastic Anemia of
Cooley)," ]our. Pediatr. 9 (1936):279-3n; G. Astaldi et al., La talassemia)·Morbo di
Cooleye forme affini (Pavia, 1951);R. M. Bannerman, Thalassemia: A Survey of Some
Aspects (New York, 1961);and Weatherall, "Thalassemia" (above, n. 24), pp. 373-
414. For the determination of the thalassemic trait, see D. Hammond et al.,
"Definition of Cooley's Trait or Thalassemia Minor: Classical, Clinical, and Rou-
tine Hematology," Ann. N. Y Acad. Sci. n9 (1964): 372-89. For an up-to-date
survey, see especially the monograph of D. J. Weatherall and J. B. Clegg, The
Thalassemia Syndromes) 2d ed. (Oxford, 1972).
27. Definitive proof of the hereditary nature of thalassemia was provided by V.
Angelini, "Primi risultati di ricerche ematologiche nei familiari de'ammalati di
anemia di Cooley," Minerva Med. 28 (1937): 331ff., and J. Caminopetros, "Re-
cherches sur l'anemie erythroblastique infantile des peuples de la Mediterranee
orientale; Etude Anthropologique, etiologique et pathogenique; La transmission
hereditaire de la maladie, '' Ann. Med. 43 (1938):27-61 and 104-25. The genetic
details of this disease were laid out by W. N. Valentine and J. V. Neel, "Hema-
tologic and Genetic Study of Transmission of Thalassemia (Cooley's Anemia;
Mediterranean Anemia)," Arch. Intern. Med. 74 (1944): 185-96, and by H. Leh-
mann, "Variations in Human Hemoglobin Synthesis and Factors Governing Their
Inheritance," Brit. Med. Bull. 15 (1959):401-46. See also D. L. Rucknagel, "Cur-
rent Concepts of the Genetics of Thalassemia," Ann. N. Y Acad. Sci. n9 (1964):
463-49.
28. For clarification of the physiopathological mechanism of the thalassemias
and their classification, see Lehmann, "Variations in Hemoglobin" (above, n. 27);
P. Fessas, "Forms of Thalassemia," in J.H.P. Jonxis and J. F. Delafresnaye, Ab-
nonnal Hemoglobins (Oxford, 1959); A. Fiehrer, "Les nouvelles donnees sur les
anemies mediterraneennes: Les beta et alpha-thalassemies," Rev. Path. Comp. 64
(1964): 273-80; M. L. Freedman, "Thalassemia, an Abnormality in Globin Chain

Synthesis," Amer. ]our. Med. Sci. 267 (1974): 256-65; G. Schapira et al., Pathologie
moliculaire (Paris, 1975).
29. For the clinical picture and the facies of thalassemic children in modern
Greece, see J. Caminopetros, "Recherches" (above, n. 27), pp. 27-61, and S.
Charokopos, "Considerations sur l'anemie de Cooley chez l'enfant grec," Pedia"trie
(Lyon) 10 (1955): 535-38.
30. See J. P. Caffey, "Cooley's Anemia: A Review of the Roentgenographic
Findings in the Skeleton," Amer. four. Roentg. 78 (1957): 381-91; J. E. Moseley, Bone
Changes in Hematologic Disorders (&entgen Aspects) (New York and London, 1963);
V. Bismuth and R. Benacerraf, "Etude radiologique des manifestations osseuses
des anemies hemolytiques hereditaires," Ann. Radio!. Sem. Hop. 10 (1967): 559-74.
For the microscopic structure of the skull ("hair-on-end" pattern), see especially
the old piece by L. Belloni and P. Fornara, ''Istogenesi del cranio a spazzola nel
morbo di Cooley," Minerva Pediatrica 7 (1955): 1638-45.
31. R. Lucot-Branlard, Contribution a Fetude des lisions osseusesde la thalassemie)
thesis (Paris, 1969), and M. F. Tardivel, Contribution a Fetude du diagnostic du trait
thalassemique par la radiographie du crane) thesis (Paris, 1971).
32. See for example the articles by J. B. Moseley, "The Paleopathological Rid-
dle of Symmetrical Osteoporosis," Amer. ]our. Roentg. 95 (1956): 135-42, and by
J. L. Angel, "Porotic Hyperostosis or Osteoporosis Symmetrica," in D. Brothwell
and A. T. Sandison, Diseases in Antiquiry (Springfield, 1967), pp. 378-89.
33. Notably J. B. Moseley, "Radiographic Studies in Hematologic Bone Dis-
eases: Implications for Paleopathology," in Jarcho, Human Palaeopathology (above,
11. 5), pp. 121-30.
34. J.B. Herrick, "Peculiar Elongated and Sickle-shaped Red Blood Corpuscles
in a Case of Severe Anemia," Arch. Intern. Med. 6 (1910): 517-21. For the history
of this discovery, see the autobiographical account of J. B. Herrick, Memory of
Eighry Years (Chicago, 1940). For a general sketch of the history of research on
Herrick's disease in our time, see especially C. L. Conley, "Sickle-cell Anemia-
The First Molecular Disease," in Wintrobe, Blood) Pure and Eloquent (above, n.
24), pp. 319-71.
35. The discovery ofhemoglobin Sand of its role in the physiopathology of an
inherited disease was made in 1949 by Linus Pauling, H. Itano, S. J. Singer, and
I. C. Wells. It represents a milestone in the history of contemporary medicine. See
L. Pauling et al., "Sickle-cell Anemia: A Molecular Disease," Science no (1949):
138-42. This publication marks the debut of molecular pathology.
36. For the fundamental hematological aspects of sickle cell anemia, see B. J.
Culliton, "Sickle-cell Anemia: The Route from Obscurity to Prominence," Science
178 (1972): 138-42; H. Lehmann and R. G. Huntsman, Man)s Haemoglobins (Ox-
ford, 1974); Actes du Symposium sur la drepanocytose (Abidjian, 1975); Schapira et al.,
Pathologie moliculaire (above, n. 28); H. F. Bunn et al., Human Hemoglobins (Phila-
delphia, 1977); Conley, "Sickle-cell Anemia" (above, n. 34).
37. For the clinical aspects of the disease, besides the publications cited in the
previous note, see G. R. Serjeant, The Clinical Features of Sickle Cell Disease (Am-
sterdam, 1974). For its genetics, the basic discoveries are published in the following
articles: W. H. Taliaferro and J. G. Huck, "The Inheritance of Sickle-cell Anemia
in Man," Genetics 8 (1923): 594-98; J. V. Neel, "The Inheritance of Sickle Cell
Anemia," Science no (1949): 64-66; A. C. Allison, "Population Genetics of Ab-
normal Human Haemoglobins," Acta Genet. 6 (1957): 430-34; and D. N. Ruck-
nagel, ''The Genetics of Sickle Cell Anemia and Related Syndromes,'' Arch. In-
tern. Med. 133 (1974): 595-610.
38. See K. R. Diebert, "Roentgen Changes in Sickle Cell Anemia," Amer. ]our.
Roentg. 82 (1959): 501-4; Mo'seley, Bone Changes (above, n. 30); and G. Charmot,
'' Aspect radiologique des lesions osseuses clans la maladie drepanocytaire,'' Ann.
Soc. Be{geMid. Trop. 49 (1969): 199-204.
39. See chapter 9, above.
40. Luan Eng Lie Injo, "Chronic Deficiency Anemia with Bone Changes
Resembling Cooley's Anemia," Acta Haemat. 19 (1958): 263-68; H. Burko et al.,
"Skull Changes in Iron Deficiency Anemia Simulating Congenital Hemolytic Ane-
mia," Amer. ]our. Roentg. 86 (1961): 447-52; P. Lanzkowsky, "Radiological Features
in Iron Deficiency Anemia," Amer. ]our. Dis. Child. 116 (1968): 16-29; P. Lanz-
kowsky, "Osseous Changes in Iron Deficiency Anemia-Implications for Paleo-
pathology," in Porotic Hyperostosis:An Enquiry (Detroit, 1977), pp. 23-34.
41. G. Gurrarino and M. Erlandson, "Premature Fusion ofEpiphyses in Cool-
ey's Anemia," Radiology83 (1964): 656-64; R. Lucot-Branlard, Contribution (above,
n. 31); H. G. Poynton and K. W. Davey, "Thalassemia," Oral Su13. 25 (1968):
564-76.
42. H. Nathan and N. Haas, "On the Presence ofCribra Orbitalia in Apes and
Monkeys," Amer. ]our. Phys. Anthrop. 24 (1966): 351-60; 0. P. Hengen, "Cribra
Orbitalia: Pathogenesis and Probable Etiology," Homo 22 (1971): 57-75; Carlson,
Armelagos, and Van Gerven, "Cribra Orbitalia" (above, n. 15).
43. See especially J. Lallo, G. J. Armelagos, and R. P. Mensforth, "The Role
of Diet, Disease, and Physiology in the Origin of Porotic H yperostosis,'' Hum.
Biol. 49 (1977): 471-83; L. M. Debra and G. J. Armelagos, "Paleoepidemiological
Methods and Porotic Hyperostosis," Paleopathology Newsletter) no. 24 (1978),
pp. 14-17.
44. Determination of the level of iron was used by a research team in Pisa on
bone remains from Carthage. In a sample of 24 skulls dated to the third century
B.C., of which 13 had cribra orbitalia, a direct relation was revealed between the
presence of the lesion and a drop in the level of iron in the bones. The age and sex
distribution of the cases of cribra orbitalia suggests a diagnosis of acquired iron
deficiency anemia for this population. Cf. G. Fornaciari and F. Mallegni, "Cribra
orbitalia in un campione de Punici di Cartagine,'' Quad. Sci. Antrop. (Padua) 5
(1980): 106-21, and G. Fornaciari et al., "Cribra Orbitalia and Elemental Bone Iron
in the Punics of Carthage,'' Ossa 8 (1981): 63-77. Since iron is a factor in the
synthesis of two amino acids, hydroxylysine and hydroxyproline, that are found in
bone protein, it can be argued that their reduced concentration in skeletons with
porotic hyperostosis supports a diagnosis of iron deficiency anemia. This was con-
vincingly demonstrated by D. W. von Endt and D. J. Ortner, "Amino Acid
Analysis of Bone from a Possible Case of Prehistoric Iron Deficiency Anemia from
the American Southwest," Amer. ]our. Phys. Anthrop. 59 (1982): 377-85.
45. The basic data are in the following publications: Phaedon Fessas, "Hered-
itary Anemias in Greece," in J.H.P. Jonxis et al., Abnormal Haemoglobins (Oxford,
1959), pp. 260-66; B. Malamos et al., "Types ofThalassemia-trait Carriers as Re-
vealed by a Study of Their Incidence in Greece," Brit. ]our. Haemat. 8 (1962): 5-14;
N. A. Barnicot et al., "Haemoglobin Types in Greek Populations," Ann. Hum.
Gen. 26 (1963): 229-36; J. Bernard and J. Ruffie, Himatologie giographique (Paris,
1966), 1: 89; N. Matsaniotis and C. Kattamis, "Thalassemias, a Social Problem in
Greece," Ann. Soc. Be{geMid. Trop. 49 (1969): 223-30.
46. N. Spiropoulos et al., "Anemie mediterraneenne (erythroblastique) ou
thalassemie ou anemie de Cooley et anemie a cellules falciformes," Sang 25 (1955):
610; Charokopos "L'anemie de Cooley chez l'enfant grec" (above, n. 29).
47. Matsaniotis and Kattamis, "Thalassemias" (above, n. 45).
48. Malamos et al., "Types ofThalassemia-trait Carriers" (above, n. 45).
49. See the publications of Barnicot et al. and Matsaniotis and Kattamis cited
above, n. 45, as well as A. Gouttas, "Anemies hemolytiques en Grece," Ann. Soc.

Be{geMed. Trop. 49 (1969): 185-92.
50. G. R. Fraser et al., "Thalassemia, Abnormal Hemoglobins, and Glucose-
6-phosphate Dehydrogenase Deficiency in the Arta Area in Greece,'' Ann. N. Y
Acad. Sci. n9 (1964): 415-35. The numbers given in parentheses are indicative, not
absolute. The distribution of anemic traits in this area is not homogeneous. A
study by C. Choremis, L. Zannos-Marioulea, and M.D.C. Kattamis shows that
the levels are different among inhabitants of the plains as against the mountain-
dwellers (' 'Frequency of Glucose-6-phosphate Dehydrogenase Deficiency in Cer-
tain Highly Malarious Areas of Greece," Lancet [1962], I, pp. 17ft).
51. The delta trait, that is, the hemoglobinosis in which genetic error affects the
production of the delta chain, was discovered in Greece by Phaedon Fessas in 1962.
52. See Barnicot et al., "Haemoglobin Types" (above, n. 45).
53. Choremis, Zannos-Marioulea, and Kattamis, "Frequency" (above, n. 50).
54. For the worldwide distribution of the thalassemias, see A. I. Chernoff,
"The Distribution of the Thalassemia Gene: A Historical Review," Blood 14 (1959):
899-912; G. Sannie, Ripartitiongeographique de la thalassemie; Son interpretation) thesis
(Paris, 1964); Bernard and Ruffie, Hematologie (above, n. 45), 1: 88-89; Weatherall
and Clegg, Thalassemia Syndromes (above, n. 26); and Weatherall, "Toward an
Understanding" (above, n. 24).
55. A. G. Maratchev, "Gemoglobinopatii v Tadzhikistane," Vestn. Akad. Med.
Nauk (1965),p. 86.
56. E. Silvestroni and I. Bianco, "The Distribution of Microcythaemias (or
Thalassemias) in Italy," in Jonxis et al., Abnormal Haemoglobins (above, n. 45),
pp. 242-59.
57. See L. C. Brumpt, "A propos de l'anemie de Cooley: Thalassemie ou
sinemie?" Bull. Acad. Nat. Med. 139(1955):333-36.
58. I. Gatto, "Sulla ereditarieta della malattia di Cooley," Minerva Med. 39
(1948): 194-98, and "Origine della thalassemia," Pensiero Scientifuo 3 (1960): 413-19.
59. E. C. Zaino, "Paleontologic Thalassemia," Ann. N. Y Acad. Sci. n9 (1964):
402-12.
60. G. R. Fraser et al., "Glucose-6-phosphate Dehydrogenase Deficiency, Ab-
normal Haemoglobins, and Thalassaemia in Yugoslavia,'' four. Med. Gen. 3 (1966):
35-41.
61. A. Orsini and L. Badettii, "Nosologie, etiologie et pathologie des thalasse-
mies," Rapports au xve Congres des Pedia-tresde Langue fram;aise (Marseille, 1955),
pp. 1-130.
62. For such a critique, see J. F. Pays, La thalassemie eurasiatique. Essai d)anthro-
popathologie (Paris and Toulouse, 1971),pp. 136-40.
63. E. Silvestroni, I. Bianco, and N. Alfieri, "Sulle origini della microcitemia
in Italia e nelle altre regioni della terra," Medicina (Parma) •2 (1952):187-216.
64. See Caminopetros, "Recherches" (above, n. 27), p. 106.
65. Brumpt, "L'anemie de Cooley" (above, n. 56), p. 333.See also Thor Peng
Thong, La thalassemie au Cambodge (Paris, 1958).
66. L. C. Brum pt, "Les splenomegalies chroniques africaines," Rev. Sci. Med.
162 (1964): 78-87 (esp. p. 85).
67. Pays, La thalassemie (above, n. 62).
68. L. C. Brumpt and J.-F. Pays, "Signification anthropologique de la thalas-
semie eurasiatique," Maroc Med. 52 (1972):499-506.
69. Pays, La thalassemie (above, n. 62), p. 198.
70. P. Bugard, L)etat de maladie (Paris, 1964), p. 99.
71. Brumpt and Pays, "Signification anthropologique" (above, n. 68).
72. 0. Tonz et al., "New Mutation in a Swiss Girl Leading to Clinical and
Biochemical {J-Thalassemia Minor," Humangenetik 30 (1973):321-27.
73. F. B. Livingstone, Abnormal Hemoglobins in Human Populations (Chicago,
1967); Bernard and Ruffie, Hematologie (above, n. 45), 1: 72-77; and cf. the
publications cited in n. 36, above.
74. H. Lehmann and M. Cutbush, "Sickle-cell Trait in Southern India," Brit.
Med. ]our. (1952), r, pp. 404-5; H. Lehmann, "Distribution of the Sickle-cell
Gene: A New Light on the Origin of the East Africans," Eugenics Review46 (1954):
3-23; R. Singer, "The Origin of the Sickle Cell," S. A.fr. ]our. Sci. 50 (1954):287-91.
On Lehmann's research, see especially Conley, "Sickle-cell Anemia" (above, n.
34), pp. 329-30.
75. See the articles by P. Brain in Man 53 (1953):154, and Brit. Med. ]our. (1973),
III, p. 294.
76. E. C. Bucchi, "Is Sickling a Veddit Trait?" Anthropologist 1 (1955):25-29.
77. Bernard and Ruffie, Hematologie (above, n. 45), 1: 72-77.
78. For contact between Greece and black Africa during the classical period,
see especially A. Bourgeois, La Greceantique devant la negri-:ude(Paris, 1971).
79. J.B.S. Haldane, "The Rate of Mutation of Human Genes," ltoceedings of
the VIIIth International Congress on Genetics (Stockholm, 1948), Hereditas) suppl.
(1949), and "Disease and Evolution," llicerca scientifica 19 (1949): 68-76.
80. A. C. Allison, "Protection Afforded by Sickle-cell Trait against Sub-tertian
Malarial Infection," Brit. Med. ]our. (1954)r, pp. 290-94.
81. The mechanism in question is well explained by A. G. Motulsky, "Meta-
bolic Polymorphism and the Role of Infectious Diseases in Human Evolution,"
Hum. Biol. 32 (1960): 28-62; A. C. Allison, "Polymorphism and Natural Selection
in Human Populations," Cold Spring Harbor Symp. Quant. Biol. 29 (1964): 137-49;
and T. Dobzhansky, Mankind Evolving (New Haven, 1963). For general thoughts
on this subject, see J. Bernard and J. Ruffie, "Origine du polymorphisme hema-
tologique chez l'homme et dynamique des populations," Annales E.S. C. 34 (1979):
1324-43.
82. See especially the mathematical model proposed by S. L. Wiesenfeld,
"Sickle-cell Trait in Human Biological and Cultural Evolution," Science 157(1967):
n34-40.
83. H. Lehmann, "The Maintenance of the Haemoglobinopathies at High
Frequency," in Jonxis et al., Abnormal Haemoglobins (above, n. 45), pp. 307-21;
A. G. Motulsky, "Hereditary Red Cell Diseases and Malaria," Amer. ]our. Trop.
Med. 13(1964): 147-58.
84. A critical survey can be found in the thesis of D. Hesse-Turner, Die Bedeu-
tung der Thalassaemie und des Glucose-6-PhosphatedehydrogenaseMangels fur die Path-
ologieund Epidemiologie der Malaria (Tiibingen, 1967).
85. See G. Montalenti, E. Silvestroni, and I. Bianco, "Nuove indagini sul
problema della microcitemia,'' R. C. Accad. Naz. Lincei) 8, t. 14 (1959):183-88, and
C. Menini, "Osservazioni storiche sulla malaria nel Ferrarese, nei suoi rapporti con
talassemia e favismo," Episteme 4_.(1970):234-40.
86. See the publications cited in n. 45, above, and the study by G. Stamatoy-
annopoulos and P. Fessas, ''Thalassemia, Glucose-6-phosphate Dehydrogenase
Deficiency, Sickling, and Malarial Endemicity in Greece," Brit. Med. ]our. (1964),
I, pp. 875-79.
87. C. Choremis et al., "Three Inherited Red-cell Abnormalities in a District
of Greece: Thalassemia, Sickling, and Glucose-6-phosphate Dehydrogenase Defi-
ciency," Lancet (1963),I, pp. 907-9.
88. D. G. Rokhlin, Bolezni drevnih lyudei (Moscow, 1965).
89. Carlson, Armelagos, and Van Gen;ren, "Factors Influencing" (above,

n. 15).
90. C. Wells, Bones) Bodies) and Disease (London, 1964), p. 114.
91. A. Marcsik and F. Kosa, "Pathological Aspects of Paleoanthropological
Finds," Acta CongressusIntern. Hist. Artis Med. (Budapest) I974) (Budapest, 1976),
2: 1301-7.
92. See T. Lodge, "Thinning of the Parietal Bones," in Broth well and Sandi-
son, Diseasesin Antiqui-ty (above, n. 32), pp. 405-16.
93. A good update can be found in F. L. Dunn, "On the Antiquity of Malaria
in the Western Hemisphere," Hum. Biol. 37 (1965):385-93, and J. de Zulueta and
S. C. Ayala, "Malaria in Pre-Columbian America?" PaleopathologicalNewsletter) no.
23 (1978),pp. 12-15.
94. M. Y. El-Najjar, B. Lozoff, and D. J. Ryan, "The Paleoepidemiology of
Porotic Hyperostosis in the American Southwest: Radiological and Ecological
Considerations," Amer. ]our. Roentg. 125(1975):918-24; M. Y. El-Najjar et al., "The
Etiology of Porotic Hyperostosis among the Prehistoric and Historic Anasazi In-
dians of Southwestern United States," Amer. ]our. Phys. Anthrop. 44 (1976): 477-
88; and M. Y. El-Najjar, "Maize, Malaria, and the Anemias in the Pre-Columbian
New World," Yearb. Phys. Anthrop. 20 (1976): 329-37.
95. P. Graziosi, "Gli uomini paleolitici della Grotta di San Teodoro (Mes-
sina)," Riv. Sci. l+eistor. 2 (1947): 123.
96. Gatto, "Sulla ereditarieta" (above, n. 58).
97. A. Ascenzi, "Thalassemie et lesions osseuses. Avec discussion d' exem-
plaires paleopathologiques italiens," Actes du JerColl. Fran{. de Paleopath. (Lyon)
1973) (Lyon, 1975),pp. 169-85.
98. E. Benassi and A. Toti, "Osservazioni sulle ossa rinvenute negli scavi della
necropoli di Spina. Conferma all'origine razziale della talassemia," Att. Mem.
Accad. Stor. Arte San.) 2d ser., 24 (1958):16-28. See also P. Di Pietro, Breve storia
dell)ematologia (Padua, 1958). It may be well to recall that Spina is in the region
where F. Rietti made his first description of thalassemia minor.
99. Ascenzi, "Thalassemie" (above, n. 97).
100. A. Ascenzi and P. Balistreri, "Porotic Hyperostosis and the Problem of
the Origin ofThalassemia in Italy," ]our. Hum. Evol. 6 (1977):595-604; A. Ascenzi,
"A Problem in Paleopathology: The Origin of Thalassemia in Italy," Virchows
Arch. Path. Anat. 384 (1979): 121-30.
101. G. Fornaciari and F. Mallegni, "Iperostosi porotica verosimilmente talas-
semica in due scheletri rinvenuti in un gruppo di tombe del III secolo a.C. di San
Giovenale (Viterbo)," Quad. Sci. An-wop. (Padua) 4 (1980): 21-50.
102. J. L. Angel, "Human Biological Changes in Ancient Greece, with Special
Reference to Lerna," Yearb. Amer. Philos. Soc. (1957),p. 269.
103. J. L. Angel, "Osteoporosis: Thalassemia?" Amer. ]our. Phys. Anthrop. 22
(1964): 369-72.
104. Angel, "Porotic Hyperostosis" (above, n. 14).
105. J. L. Angel, The People of Lema: Analysis of a ltehistoric Aegean Population
(Princeton and Washington, D.C. 1971),pp. 77-84.
106. J. L. Angel, "Anemias of Antiquity: Eastern Mediterranean," in Porotic
Hyperostosis: An Enquiry (Detroit, 1977), pp. 1-5; "Porotic Hyperostosis in the
Eastern Mediterranean," MCV Quart. 15(1978):10-16.
107. Angel, Peopleof Lema (above, n. 105), p. 55.
108. Angel, "Porotic Hyperostosis" (above, n. 14), p. 762, and Brothwell and
Sandison, Diseasesin Antiquity (above, n. 32), pp. 380-81.
109. See J. L. Benson, Bamboula (Philadelphia, 1972).
no. J. L. Angel, "Human Skeletal Material from Franchthi Cave," Hesperia 38

(1969): 380-81. ,
nr. J. Mellaart, <;atal Huyuk: A Neolithic Town in Anatolia (New York, 1967);
J. L. Angel, "Early Neolithic Skeletons from <_::atalHiiyiik: Demography and
Pathology,'' Anatolian Studies 21 (1971): 77-98; J. L. Angel, ''Early Neolithic People
of Nea Nikomedea," in I. Schwidetzky, Die Anfange des Neolithikums vom Orient
bis Nordeuropa. Fundamenta. (Cologne, 1973), pt. B, vol. 3, pp. 103-12; Angel,
"Porotic Hyperostosis in the Eastern Mediterranean (above, n. 106).
n2. Angel, "Porotic Hyperostosis in the Eastern Mediterranean" (above, n.
106).
n3. M. J.Mellink and J. L. Angel, "Excavations at Karata~-Semayiik in Lycia,"
Amer. ]our. Archaeol. 72 (1968): 243-63; Angel, "Porotic Hyperostosis in the Eastern
Mediterranean" (above, n. 106).
n4. Angel, People of Lerna (above, n. 105), p. 79.
n5. G. Brauer and R. Fricke, "Zur Phanomenologie osteoporotischer Ver-
anderungen bei Bestehen systemischer hamatologischen Affektionen," Homo 31
(1980), 198-2n.
n6. Angel, "Porotic Hyperostosis in the Eastern Mediterranean" (above,
n. 106), p. n.
n7. Brit. Med. ]our. (1973), I, p. 488, and II, p. 489.
n8. Aff int.J 32 (Littre, VII, 248).
n9. The diagnosis of homozygous thalassemia suggested by P. Brain (Brit.
Med. ]our. [1973], III, p. 294) does not seem to me acceptable, given the character
and localization of the pain and the rarity of ulcerations in thalassemics.
120. Itaen. CoacaeJ 333 (Littre, v, 656). This passage was mentioned in connec-
tion with thalassemia in studies by Caminopetros, "Recherches" (above, n. 27),
Bannerman, Thalassemia: A Survey (above, n. 26), and Weatherall, "Toward an
Understanding" (above, n. 24). Weatherall opts for a diagnosis of iron deficiency
anemia and not thalassemia.
12r. See especially P. Lanzkowsky, "Investigation into the Aetiology and Treat-
ment of Pica," Arch. Dis. Child. 34 (1959): 140-8.
122. Louvre Museum, inv. nos. EID 564 and EID 1899-1905.
123. L. J. Bruce-Chwatt, "Paleogenesis and Paleo-epide1niology of Primate
Malaria," Bull. WH. 0. 32 (1965): 363-87, and P. F. Mattingly, "Origins and Evo-
lution of the Human Malarias: The Role of the Vector," Parassitologia 15 (1973):
160-72. The hypothesis that the first center for the evolution of simian and human
malarias was not Africa but the jungles of Southeast Asia cannot be set aside. See
G. R. Coatney et al., The ltimate Malarias (Washington, D.C., 1971). From my
point of view, it makes no difference; in either case, Africa was affected long before
Europe.
124. Bruce-Chwatt, "Paleogenesis" (above, n. 123), pp. 377 and 384; P. F.
Russell, Man)s Mastery of Malaria (London, 1955); F. L. Dunn, "On the Antiquity
of Malaria in the Western Hemisphere," Hum. Biol. 37 (1965): 385-93; L. J. Bruce-
Chwatt and J. de Zulueta, The Rise and Fall of Malaria in Europe (Oxford, 1980).
125. P. C. Garnham, Malaria Parasites and Other Haemosporidia (Oxford, 1966),
and P. F. Mattingly, "Evolution of the Malarias: The Problem of Origins," Pa-
rassitologia 18(1976): 1-8. For the current state of knowledge concerning malaria
parasites and the clinical course, pathology, and epidemiology of malaria, see
L. J. C. Bruce-Chwatt, Essential Malariology (London, 1980).
126. Angel, "Porotic Hyperostosis in the Eastern Mediterranean" (above,
n. 106), p. 14.
127. D. Ferembach, La necropoleepipaliolithique de Taforalt (Maroc Oriental); Etude
des squelettes humains (Rabat, 1962).
128. P. G. Shute, "Failure to Infect English Specimens of Anopheles Maculi-

pennis var. Atroparvus with Certain Strains of Plasmodium Falciparum of Tropical
Origin," ]our. Trop. Med. Hyg. 43 (1940): 175-87; J. de Zulueta, C. D. Ramsdale,
and M. Coluzzi, "Receptivity to Malaria in Europe," Bull. WH. 0. 52 (1975):109-
n; Bruce-Chwatt and de Zulueta, The Rise and Fall of Malaria (above, n. 124),
pp. 13-15.
129. J. de Zulueta, '' Malaria and Mediterranean History,'' Parassitologia15(1973):
r-15; Bruce-Chwatt and de Zulueta, The Rise and Fall of Malaria (above, n. 124),
pp. 13-16and 34.
130. These ideas were first intuited by Sir Ronald Ross (1857-1932),who won the
Nobel Prize in medicine for his research on the role of mosquitoes in the transmis-
sion of malaria. He conceived them during a trip to Greece. See R. Ross, "Malaria
in Greece," four. Trop. Med. 9 (1906): 341-47. Not knowing how to document his
historical conjectures, Ross enlisted his friend Jones, a classical philologist. See
W.H.S. Jones, Malaria) a Neglected Factor in the History of Greeceand &me (Cam-
bridge, 1907) (with a valuable introduction by Ross); "Malaria and History," Ann.
Trop. Med. Parasit. r (1908): 529-46; "Dea Febris-A Study of Malaria in Ancient
Italy," Ann. Archaeol. Anthrop. 2 (1909): 97-124; Malaria and Greek History
(J\rlanchester, 1909); "Ancient Documents and Contemporary Life, with Special
reference to the Hippocratic Corpus, Celsus, and Pliny,'' in Science)Medicine) and
History (Essaysin Honor of C. Singer) (Oxford, 1953),r: roo-no. See also P. Fraccaro,
1
"La malaria e la storia degli antichi popoli classici," Atene e &ma 22 (1919):57-88,
and N. Toscanelli, La malaria nelfantichita e la fine degli Etruschi (Milan, 1927).
131. This is a significant fact because his homeland, Boeotia (especially the Lake
Copais basin, which was a swamp at the time), fulfilled all the geographic condi-
tions for rampant malaria. Actually, Hellenistic sources consider certain Boeotian
locales like Onchestos to be fever centers. In Plutarch's time, this region of central
Greece was devastated by malaria. See Jones, Malaria and Greek History (above, n.
130), pp. 25ff., 40, and 54.
132. This observation by Jones (ibid., p. 26) is absolutely true, but the argu-
ment he makes of it is vitiated by the extreme rarity of texts preserved from the
period in question. Even the early date of the poetry attributed to Theognis is
doubtful.
133. Theognis, I, 174. See Jones, Malaria and Greek History (above, n. 130), and
B. A. Van Groningen, Theognis) Le premier livre) edite avec un commentaire (Amster-
dam, 1966), pp. 68-69.
134. Aristophanes, Wasps) 277, 281, and 812;Acharn') n65.
135. Jones, Malaria and Greek History (above, n. 130), p. 35.
136. Ibid., pp. 35and 42-43.
137. See J. P. Cardamatis, "La malaria a Athenes depuis les temps les plus
recules jusqu'a nos jours," Atti Soc. Studi Mal. 8 (1907)':rn, and "Du paludisme
clans la Grece continentale depuis les temps les plus recules jusqu'a la periode
macedonienne," Arch. Schiff Trop. Hyg. 19 (1915):273-301; A. Cawadias, "Le palu-
disme clans l'histoire de l'ancienne Grece," Bull. Soc. Fra~. Hist. Mid. (1909),
pp. 158-65; G. A. Livadas, "Malaria in Ancient Greece," Riv. Parassit. 20 (1959):
299-304.
138. It is worth recalling that at the time when Ross and Jones were publishing
their historico-epidemiological hypotheses, malaria affected about a third of the
population of Greece. Malaria-induced mortality was very high there, and the
influence of the disease on economic and cultural life was especially devastating.
On the history of malaria in Greece in modern times, see G. D. Belos, L)histoire du
paludisme en Grecedepuis f Antiquite jusqu)a la dicouverte de L~eran (Paris, 1933),and
G. A. Livadas and J. C. Sphangos, Malaria in Greece(Athens, 1941).
139. As an example, see F. Regnault, ''La decadence de la Grece expliquee par
la deforestation et l;.mpaludisme," l+esseMid. (1909), pp. 729-31, and "Du role
du depeuplement, du deboisement et de la malaria clans la decadence de certaines
nations," Rev. Scient. 52 (1914):46ff.
140. The American geographer Ellsworth Huntington provided the first de-
tailed elaboration of the hypothesis that seeks to explain fluctuations in the malarial
endemic in Greece by climatic cycles. Its recrudescence in Hellenistic times was
due to the period of aridity that began around 300 B.C. See E. Huntington, "The
Burial of Olympia: A Study in Climate and History," Geograph.four. 36 (1910):
657ff. For a contemporary view of the relation between malaria and climate in
Greece, see especially G. Panessa, "Recenti studi di interesse paleoclimatologico
riguardanti la Grecia," Ann. ScuolaNorm. Sup. Pisa) 3d ser., III, 12 (1982):1601-14.
141. See A. Celli, Malaria (London, 1900), and Storia della malaria nell)Agro
lwmano (Rome, 1925).
142. See Jones, Malaria and Greek History (above, n. 130), pp. 62-73; Kind,
"Malaria," in Pauly-Wissowa, &al-Encyclopadie)14 (1928),col. 830-846; G. Sticker,
"Fieber und Entziindung bei den Hippokratikern," Arch. Gesch.Med. 20 (1928):
150-74; 22 (1929-30): 313-43, 361-81;23 (1930):40-67; G. Chairopoulos, Le paludisme
dans la Greceantique) thesis (Lyon, 1930); H. E. Sigerist, History of Medicine) vol. 2
(Oxford, 1961),pp. 328-32.
143. See chapter n, below, and M. Grmek, "Les ruses de guerre biologiques
clans l' Antiquite," Rev. Et. Grecques92 (1979): 141-63. Introduction of the falcipa-
rum type could have been brought about or facilitated by Xerxes' Persian army.
For this hypothesis, see C. Laderman, "Malaria and Progress: Some Historical and
Ecological Considerations," SocialScienceand Medicine 9 (1975):587-94 (esp. p. 592).
144. Jones, Malaria and GreekHistory (above, n. 130), p. 76.
145. Ibid., pp. 95-100 and 132.
146. L. W. Hackett, Malaria in Europe (London, 1937),and E. H. Ackerknecht,
Geschichteund Geographicder Wichtigsten J(rankheiten (Stuttgart, 1963),p. 80.
Chapter Eleven: The HippocraticConceptionof Disease

1. The essence of this chapter was published in Actualitis himatologiques1978)
12th ser. (Paris, 1978), pp. 293-315.
2. Epid°JI, patient l (Littre, II, 682-84).
3. The numbers in parentheses refer to parts of the discussion in the next
section of this chapter.
4. H. Kuehlewein, HippocratisOperaquaeferuntur omnia (Leipzig, 1894), 1: 180-
245.
5. W.H.S. Jones, Hippocrates(Cambridge and London, 1923),1: 146-2n (specif-
ically, p. 186).
6. Galen, In HippocratisEpidcmiarum librum primum commentaria (Kuhn, XVII,
A. 1-302). Critical edition by E. W enkebach and F. Pfaff, CorpusMedicorum Grae-
corum) v, 10, 1 (Leipzig and Berlin, 1934).
7. E. Littre Oeuvrescompletesd)Hippocrate(Paris, 1840), 2: 598-717, esp. pp. 683-
85.
8. Desmars, Epidimiques d)Hippocrate(Paris, 1767); C. Daremberg, Oeuvreschoi-
siesd)Hippocrate)2d ed. (Paris, 1855),pp. 426-27; L. Bourgey, Observationetexpirience
chez les midecins de la Collectionhippocratique(Paris, 1953),p. 200; and M. Martiny,
Hippocrateet la midecine (Paris, 1964), p. 192.
9. R. Fuchs, Hippokrates)Siimtliche Werke (Munich, 1897),2: 99-128; G. Sticker,
Hippokrates) Der Volkskrankheiten erstes und drittes Buch (Leipzig, 1923); R. Kapferer
and G. Sticker, Die Werke des Hippokrates (Stuttgart and Leipzig, 1939), vol. 11;W.
Mueri, Der Arzt in Altertum) 3d ed. (Munich, 1962), p. 97; and H. Diller, Hippok-
rates) Schriften (Reinbek bei Hamburg, 1962), pp. 30-31.
10. F. Adams, The Genuine Works of Hippocrates (London, 1849 rept. Baltimore,
1939), pp. 111-12,and Jones, Hippocrates (above, n. 5), p. 187.
11. M. Vegetti, Opere di Ippocrate (Turin, 1965),p. 314.
12. In this regard, see the statement of A. Foesius, Oeconomia Hippocratis Alpha-
beti Serie Distincta (Frankfurt, 1588),col. 471.
13. Daremberg, Oeuvres choisies (above, n. 8), p. 461, n. 20.
14. P. Berrettoni, "Il lessico tecnico de I e III libro delle Epidemie ippocra-
tiche," Annali Scuola norm. Pisa 39 (1970): 36.
15. Galen, In Hippocratis (above, n. 6), comm. III.
16. V. Langholf draws my attention to another Hippocratic passage in the Use
of Liquids (Littre, VI, 118,12), where the expression khrofen an kalisai means "to
bring back the color of the skin.''
17. Vegetti, Opere di Ippocrate (above, n. 11),p. 314.
18. See K. Deichgraber, Die Epidemien und das Corpus Hippocraticum (Abh. lteuss.
Akad. Wiss:, Phiwl.-hist. J(l.) no. 3) (Berlin, 1933;2d ed., Berlin and New York,
1971).
19. For the Hippocratic notion of katastasis) see Jones, Hippocrates (above,
n. 5), p. 141, and 0. Temkin, "Die Krankheitsauffassung von Hippokrates und
Sydenham in ihren Epidemien, '' Sudhofft Arch. Gesch. Med. 20 (1928):327-52.
20. Littre, II, 642.
21. Littre, II, 650-54.
22. J. Pouilloux, Recherches sur Phistoire et les cultes de Thasos) vol. 1, De la
fondation de la cite a 196 avant ].-C. (Paris, 1954), p. 249. See also J.-E. Dugand,
"Hippocrate a Thasos.et en Grece du Nord," in Corpus Hippocraticum (Actes du
Colwque de Mons) 1975) (Mons, 1977), pp. 233-45.
23. Inscriptiones Graecae) vol. XII, fasc. VIII, 1909, nos. 271and 277.
24. Deichgraber, Die Epidemien (above, n. 18), p. 16.
25. See above, n. 22.
26. There is no need to keep silent about a further difficulty that Deichgraber
does not take into account: the Hippocratic patient is once expressly referred to as
the son of Antagoras (Littre, II, 665), while the theor6s with the same name is the
son of Aristocleides. The commonness of the name "Philiscus" in Thasian epig-
raphy makes the identifcation fairly unlikely. It would be even easier, and com-
pletely gratuitous as well, to try to identify our patient with a certain Philiscus of
Thasos nicknamed "The Savage," a recluse who indulged in apiculture. We know
nothing of the man, except that he wrote a treatise on bees that was used by
Hyginus and that is cited by Pliny the Elder (Natural Histmx XI, 9).
27. The term kausos is known to derive from the verb kaio 'burn' and to belong
to the same family as, for instance, the verb "cauterize" and the adjective "caus-
tic." Besides Chantraine's Dictionnaire ltymowgique) p. 435, see the study by R.
Stromberg, Griechische Wortstudien: Untersuchungen zur Benennung von Tieren)
Pflanzen) I(orperteilen und I(rankheiten (Goteborg, 1944), p. 87.
28. Aff.) II (Littre, VI, 214).
29. The basic descriptions of this clinical entity are in Epid.) I, 9 (Littre, II, 650-
52), Acut. (app.)) 1 (Littre, II, 394-98, and Joly, 68-69), Morb.) II, 63 (Littre, VII,
96-98), and Aff.) 11(Littre, VI, 214-18). For the latter two texts, one should consult
the critical edition, translation, and philological commentary of J. J ouanna, Hip-
pocrate et !)Ecole de Cnide (Paris, 1974), pp. 274-77, 286, and so forth. For other
details on kausos) see Littre, Oeuvres completes (above, n. 7). I, 612; II, 28, 50, 232,
368, 418, 600, 618-20, 636, 640-42, 685, 666; III, 60, 66, 70, 80, 90, 98, 102, 108, 118,
122, 130; IV, 496, 509, 522, 570; V, 72, 100, 168-70, 294, 380, 392, 408, 458, 462, 514,
530-32, 608-10, 716; VI, 144-46, 194, 200-202, 222; VII, 156-60.
30. For instance Epid°JI, patient 2; Epid°JIII, first list, patients 10 and 12; second
list, patients 1, 5, 7, 9, and 12; Epid.JVII, 10, 20, and 42.
31. R. Wittern, Die hippokratischeSchrift De morbisI (Hildesheim and New York,
1974), pp. 190-91.
32. See the old bibliography in C. F. Fuchs, "Der Causos des Hippokrates,"
Archiv des Vereinsfiir wiss. Heilkunde 2 (1866), and Sticker, Hippokrates(above, n. 9),
p. 92.
33. J.-B. Germain, Les Epidemiesd)Hippocratepeuvent-ellesetre rapporteesa un cadre
nosologique? thesis in medicine (Paris, 1803).
34. F. Adams, The SevenBooksof Paulus Aegineta (London, 1844), 1: 260-62.
35. Littre, Oeuvrescompletes(above, n. 7), pp. 566-71; E. Beaugrand, "Causus,"
in Dictionnaire encyclopedique des sciencesmidicales (Paris, 1874) 13:391-92.
36. C. A. Wunderlich, Geschichteder Medizin (Stuttgart, 1859),p. 9.
37. W.H.S. Jones, Malaria and Greek History (Manchester, 1909). To be sure,
Jones admits that certain concrete Hippocratic cases of kausosare complications of
malaria.
38. Sticker, Hippokrates(above, n. 9), pp. 91-93. The use of the terms kausosand
puretos kaus7Jdes in the parallel versions of Aff.J 11, and Morb°J II, 63, proves against
Sticker that the two expressions are utterly synonymous.
39. W. MacArthur, "Historical Notes on Some Epidemic Diseases Associated
with Jaundice," British Medical Bulletin 13 (1957):146-49.
40. See in particular Stromberg, GriechischeWortstudien (above, n. 27), pp. 87-
88.
41. Galen, De atra bile) 4 (CorpusMed. GraecorumJv. 4, 1, p. 76).
42. See n. 29, above.
43. Aretaeus of Cappadocia as well as Galen, Paul of Aegina, and Alexander of
Tralles all describe kausosand distinguish two clinical forms of it, one "authentic"
and the other "false." The split made by these late authors is not exactly the same
as those in Acut. (app.)J 1. From the standpoint of modern differential diagnosis,
none of these divisions is really useful.
44. See Adams, The SevenBooks (above, n. 34), p. 261.
45. M. Meyerhof, "Thirty-three Clinical Observations by Rhazes (circa 900
A.D.)," Isis 23 (1935):321-72, esp. p. 347.
46. M. Vust-Mussard, "Remarques sur les livres I et III des epidemies; Les
histoires de malades et le pronostic," Etudes de Lettres (Lausanne), 3d, ser., 3 (1970):
67-69.
47. See Temkin, "Die Krankheitsauffassung" (above, n. 19). Fernand Robert
has provided an excellent review of the role of prognosis in the relatively recent
portions of the Epidemics: "La prognose hippocratique clans les livres v et VII des
Epidemies," in Le Mondegrec (Hommage a Claire freaux) (Brussels, 1975), pp. 257-
70.
48. See E. Vintr6, Hip6cratesy la nosolog[ahippocratica(Barcelona, 1972), esp.
p. 125.
49. For a formal analysis of the clinical observations attributed to Hippocrates,
see P. Lafn Entralgo, La historia cl[nicaJ 2d ed. (Barcelona, 1963), and Vintr6,
Hip6crates(above, n. 48), pp. 107-44. There have been attempts at critical analysis
of these observations in the light of modern medical knowledge, notably by R. E.
Siegel, "Clinical Observations in Hippocrates: An Essay on the Evolution of the
Diagnostic Art," Journal of the Mount Sinai Hospital 31(1964): 285-303.
50. Epidemics)I, 8 (Littre, II, 640-42; Jones, I, 167).
51. Epidemics)I, 9 (Littre, II, 650-54; Jones, I 173).

52. Vust-Mussard, "Remarques sur les Epidemies" (above, n. 46), p. 72.
53. Jtorrh.J I, 39 (Littre, v. 520). The same text recurs in Coac.J49.
54. ltogn._,12 (Littre, II, 142).
55. Epidemics)I, patients 1, 2, and 3; Epidemics)III, second list, patients 2, 3, 9,
10, 11,13, and 14.
56. M. Sorre, "Complexes pathogenes et geographie medicale," Annales de
Geographie42 (1933):1-18.
57. H. Scott, History of TropicalMedicine (London, 1939), 1:252-78; J.W.W. Ste-
phens, BlackwaterFever:Historical Survey and Summary of ObservationsMade Over a
Century (Liverpool, 1937).
58. See A. Guillon, "La fievre bilieuse hemoglobinurique," in Grandesepidimies
tropicales(Paris, 1935),7: 60-75.
59. Fluctuation in the terminology reflects the uncertainties: the disease is called
"pernicious melanuric remittent fever" (Duchassaing, 1850), "miasmatic haema-
turia" (Cummings, 1859), "bilious hematuric fever" (Barthelemy-Benoit, 1865;
Veillard, 1867), "bilious melanuric fever" (Berenger-Feraud, 1874), "black urine
fever" (Pellarin, 1876), "hemospherinic paludal fever" (Karamitsas, 1882), "hemo-
globinuric bilious fever" (Corre, 1883), "blackwater fever" (Easmon, 1884),
"Schwarz,vasserfieber" [blackwater fever] (Plehn, 1895),etc.
60. A. Antoniades, "Peri t6n haimorragi6n kal idfos tgs haimatourfas epl t6n
dialeip6nt6n puret6n,'' Iatriki Ephemerfs1 (1858-59):161-63.
61. C. Stephanos, "Grece (Topographie medicale)," in Dictionnaire encyclope-
dique dessciencesmedicates)4th ser. (Paris, 1884), 10:500.
62. Seen. 60, above.
63. G. Karamitsas, "Perl el6dous haimosphairinikou puretou (Athens, 1882);
G. Karamitsas, ''Sur la fievre hemispherinurique, '' Archives de medecinenavale (1882),
pp. 153-56.
64. See the publications cited in the previous note and Scott, TropicalMedicine
(above, n. 57).
65. See Scott, TropicalMedicine (above, n. 57), 1: 266-69.
66. E. Marchiafava and A. Bignami, Sulle febbre malariche estivo-autumnali
(Rome, 1892).
67. See for example Guillon, "La fievre bilieuse" (above, n. 58), and B. Mae-
graith, Pathologicalltocessesin Malaria and BlackwaterFever(Oxford, 1948).
68. For instance, G. Charmot and L. J. Andre, "Infection palustre et reactions
immunitaires," Medecine tropicale26 (1966): 115-30;F. Vachon et al., "Acces per-
nicieux palustre," in Annie en reanimation 1 (1970): 105-70; G. Saimot et al., "Les
Formes aigues graves du paludisme, '' Med. mal. infect. 1 (1971):9-22.
69. See chapter 9, above.
70. J. Gear, "Autoantigens and Autoantibodies in the Pathogenesis of Disease
with Special Reference to Blackwater Fever," Trans. Royal Soc. Trop.Med. 39 (1945):
301.
71. Jones, Hippocrates(above, n. 5), p. 144.
72. R. Joly, Le niveau de la sciencehippocratique(Paris, 1966), p. 220.
73. H. Foy and A. Kondi, "Researches on Blackwater Fever in Greece: Intro-
duction and History," Annals Trop.Med. Parasit. 29 (1935):383-93.
74. Epidemics)III, second list, patient 3 (Littre, III, 112-16).
75. See R. H. Major, "How Hippocrates Made His Diagnoses," International
Ree. Med. 170 (1957):479-85.
76. L. F. Barker, A Study of Some Fatal Cases of Malaria (Baltimore, 1895),
pp. 6-8.
77. See for instance F. Vachon et al., "L'insuffisance renale aigue du paludisme
pernicieux," NouvellePresseMedicate 2 (1973):1035-39.
78. Cf. Germain, Les Epidemiesd)Hippocrate(above, n. 33).
79. Littre, Oeuvrescompletes(above, n. 7), 2:539-67.
80. Ibid., 2:543.
81. Wunderlich, Geschichteder Medizin (above, n. 36).
82. Stephanos, "Grece" (above, n. 61), p. 500.
83. J. P. Cardamatis, De lafievre hemoglobinuriqueobserveeen Grece(Athens, 1901).
84. Foy and Kondi, "Researches on Blackwater Fever" (above, n. 73).
85. Scott, TropicalMedicine (above, n. 57).
86. Martiny, Hippocrate (above, n. 8).
87. Sticker, Hippokrates (above, n. 9), pp. 90-91.
88. R. E. Siegel, "Epidemics and Infectious Diseases at the Time of Hippoc-
rates: Their Relation to Modern Accounts," Gesnerus17 (1960): 77-98.
89. See R. F. Timken-Zinkann, "Black Bile: A Review of Recent Attempts to
Trace the Origin of the Teaching on Melancholia to Medical Observations," Med-
ical History 12 (1968):288-92.
Chapter Twelve: The ((Cough of Perinthus))

r. The text of the first part of this chapter was given before the Third Hippo-
cratic Colloquim (Paris, 1978). The second part has not been published before.
2. Epidemics)VI, 7, 1, (Littre, V, 330-36).
3. In this translation, the numbers enclosed in parentheses refer to the headings
in the commentary that follows it. Word in square brackets are my own editorial
comments.
4. After the appearance of my study in the acts of the Paris colloquium, D.
Manetti and A. Roselli published a critical revision of the Greek text and an Italian
translation of the sixth book of the Epidemics (Florence, 1982). Their work did not
lead me to change my translation or conclusions.
5. Cf. W. Braeutigam, De HippocratisEpidemiarum Libri Sexti Commentatoribus)
thesis (Konigsberg, 1908).
6. E. Wenkebach and F. Pfaff, Galeni in Hippocratis Epidemiarum librum VI
commentaria I-VIII (Berlin, 1956), (Corpus Medicorum Graecorum) v, 10, 2, 2). The
part concerning the "cough of Perin thus" occurs on pp. 386-401. It was translated
into German by F. Pfaff from an Arabic text by Hunain ibn Ishaq. On this subject,
consult F. Pfaff, "Die nur arabisch erhaltenen Teile der Epidemienkommentare
Galens und die -Oberlieferung des Corpus Hippocraticum," Sitzungs-Bericht. Ber-
lin. Akad. Wiss. (1931),p. 558.See also B. Alexanderson, "Bemerkungen zu Galens
Epidemienkommentaren," Eranos 65 (1967): n8-45.
7. Palladii scholia in Hippocratisde popularibus Morbis librum VI) in F. R. Dietz,
Scholia in Hippocratem et Galenum (Konigsberg, 1834), 2: pp. 1-204 (rept. Amster-
dam, 1966); and C. D. Pritchet;Johannis Alexandrini commentaria in sextum librum
HippocratisEpidemiarum (Leiden, 1975).
8. E. Littre Oeuvres completesd)Hippocrate (Paris, 1846), 5: 331-37; Th. Pusch-
mann, "Die Influenza im Alterthum," Wien. I(lin. Wschr. 6 (1893):239-42; R.
Fuchs, Hippokrates) Stimtliche Werke (Munich, 1897), 2: 280-82; R. Kapferer (with
the collaboration of G. Sticker), Die Werke des Hippokrates)vol. 2 (Stuttgart, 1939),
pp. xn/75-xn/77.
9. Cf. F. Robert, "Prophasis," Rev. Etudes Grecques89 (1976):317-42.
10. Aristotle, De generatione animalium) v. 1; 780a. See S. Byl, Recherchessur les
grands -traitesbiologiquesd)Aristote (Brussels, 1980), pp. 87ff. and 281.
II. Bitot, "Memoire sur une lesion conjonctivale non encore decrite coincidant
avec l'hemeralopie," Gaz. Hebd. Med. Chir. 10 (1863):284-88.
12. Epidemics)VI, 7, 10 (Littre, V, 342).
13. Humors) VII, trans. W.H.S. Jones, in Hippocrates)vol. 4 (London and Cam-
bridge, 1931;rept. 1953).
14. Epidemics)II, 3, I (Littre, v, 100); II, I, 5 (Littre, v, 74); II, 3, II (Littre, V.
n4); VI, 21 (Littre, V, 160); and VI, 2, 19 (Littre, V, 286).
15. Epidemics)IV, 21 (Littre, V, 160).
16. Littre, Oeuvrescompletes(above, n. 8), 5: 16-17; see Aristotle, Meteorologica)1,
6, 8.
17. Littre, Oeuvrescompletes(above, n. 8), 5: 16-17; see Thucydides, III, 87.
18. H. Grensemann, "Die Krankheit der Tochter des Theodoros," Clio Medica
4 (1969): 72.
19. Aristotle, Meteorologica)1, 6, 9; 6, 10; and 7, 10. See also Diodorus Siculus,
xv, 50, and Seneca, Natural QJ4cstions)VII, 16.
20. Gundel, s.v. J(ometen in Pauly-Wissowa, Real-Encyclopadie) xx1, col. I183;A.
Pingre, CometographieOU traite historiqueet theoriquedes cometes(Paris, 1783),1: 259;
G. Bigourdan, "Les cometes; Liste chronologique de celles qui ont apparu de
l'origine a 1900," Annuaire du Bureau des Longitudes (1927), app. A.
21. V. Di Benedetto, "Principi metodici di Ep., II, IV, v1," in CorpusHippocra-
ticum) Colloquedu Mons (Mons, 1977), p. 261.
22. See the lists of Pingre and Bigourdan cited above, as well as the work of
A. S. Yamamoto, lteliminary General Catalogueof Comets (Kyoto, 1936). A comet
appearing in 400 B.c. is mentioned in them, but that is probably the result of the
incorrect interpretation of a late text.
23. Novas and supernovas are stellar explosions that appear irregularly in the
sky. Their brightness can equal or exceed that of first-magnitude stars. On average,
one supernova and ten very bright novas are seen in a century. For antiquity, the
lists of such events are very incomplete. See D. H. Clark and F. R. Stephenson,
The Historical Supernovae(Oxford, 1977).
24. Epidemics)IV, 45 (Littre, v, 186). For the critical edition of this text and a
philological commentary, see V. Langholf, Syntaktische Untersuchungenzu Hippok-
rates-Texten(Abhandl. Akad. Wiss:, Mainz) (Wiesbaden, 1977), pp. 106ff. and 164ff.
25. K. Deichgraber, Die Epidemienund das CorpusHippocraticum(Abhandl. lteuss.
Akad. Wiss. Philol.-hist. J(l:i no. 3) (Berlin, 1933),pp. 74ff. H. Grensemann has told
me per litteras of his doubts concerning Deichgraber's conclusions.
26. Xenophon, Anabasis) VII, 1, 5; 2, 10 and 24; and, especially, VII, 7, 1-14.
27. Compare Epidemics)IV, 53, and VII, 7, 10, with Anabasis) VII, 1, 13.
28. According to Langholf, Syntaktishce Untersuchungen (above, n. 24), p. 15,
this Alkibiades could also be a place name.
29. Deichgraber, Die Epidemien (above, n. 25), p. 75.
30. Book 4 differs from books 2 and 6 in some details of language, style, and
method, but most philologists still maintain the compositional unity of the group
2-4-6.
31. I have a preference for the last decade of the fifth century B.C., even though,
since Deichgraber, general opinion has favored the first decade of the fourth.
Because of its being repeated in the handbooks, the relatively recent date of the
"cough of Perinthus" appears as an established fact based on external historical
evidence. Here is an eloquent example: in his monumental history of Greek liter-
ature, Albin Lesky states that the dating of Epidemics)II, IV, and VI, is based on
the description of an epidemic that took place in Perinthus between 399 and 395.
A fine example of circular argument! See A. Lesky, GeschichtedergriechischenLiter-

atur) 3d ed. (Bern and Munich, 1971),p. 552.
32. E. Oberhummer, s.v. Perinthus) in Pauly-Wissowa, &al-Encyclopiidie)XIX,
1, col. 806.
33. Epidemics) II, 3, 1 (Littre, v, 100, 1).
34. On this subject, see above, chapter n.
35. F. Robert, "Les Adresses de malades clans les Epidemies II, IV et VI," in
Collectionhippocratiqueet son roledans l)histoirede la medecine)Colloque de Strasbourg
(Leiden, 1975),pp. 173-94, esp. p. 174.
36. A. A. Nikitas, Untersuchungen zu den Epidemienbuchen II-IV-VI des Corpus
Hippocraticum) thesis (Hamburg, 1967; published 1968).
37. Robert, "Les adresses" (above, n. 35), pp. 174 and 184.
38. The similarity has not escaped the notice of philologists: see Deichgraber,
Die Epidemien (above, n. 25), pp. 26 and 33.
39. According to Bourgey, the accounts of the epidemic constitutions of Per-
inthus and Thasos are "of the same order," but there is "greater mastery" shown
in the latter. This judgment is the result of a common prejudice in favor of
Epidemics) I and III; it is inspired by an overall judgment of the treatises and not a
meaningful comparison of the actual texts in question. See L. Bourgey, Observation
et experiencechez les medecinsde la CollectionHippocratique (Paris, 1953),p. 38.
40. Galen, De difficultate respirationis)I, 8 (Kuhn, VII, 854-55), and In Hipp.
Epid. librum VI commentaria) praefatio (Kuhn, xvIIa, 796; CMG v, 10, 2, 2, p. 5).
4r. Galen, De difficultate respirationis)III, 1 (Kuhn, VII, 890). See also the pas-
sages from Galen cited in the previous note as well as De crisibus)2 (Kuhn, IX, 859).
Though he concedes the possibility that book 6 contains additions from other,
post-Hippocratic authors, Galen is nevertheless convinced that its basic elements
derive from original Hippocratic manuscripts revised by Thessalus (cf. CMG) v,
10, 1, p. 310, and v, 10, 2, 2, p. 5). He has the same opinion of book 2, but he
considers book 4 to be at some remove from the actual teaching of Hippocrates
himself. See Nikitas, Untersuchungen (above, n. 36), p. 15.I can add that the treatise
Presbeutikosconfirms the presence ofThessalus at the side ofHippocrates during his
visit to Thessaly (Littre, IX, 418).
42. J.-E. Dugand, "Hippocrates a Thasos et en Grece," in Corpus Hippocrati-
cum (above, n. 21), pp. 233-45 (esp. p. 244). For the close geographic, political,
and commercial links between Perinthus and Cyzicus, see Louis Robert, "Des
Carpathes a la Propontide. VII. De Perinthe a Apamee, Cyzique et Claros," Studii
classice(Bucharest) 16 (1974): 61-80.
43. Littre, Oeuvrescompletes(above, n. 8), 5: 264. Italics mine.
44. I have borrowed this definition from F. Robert ("Les adresses" [above, n.
35]), who emphasizes the usage of the word in the plural and is obviously distressed
by the need to consider the case of Perinthus as an exception. There is no valid
medical reason to give this case a special meaning that conflicts with the semantic
unity of the word "epidemic" and its derivatives in the Hippocratic treatises, but
Littre's authority was such that his mistake was perpetuated in study after study,
especially once it was reinforced by the statements ofDeichgraber on the nature of
the "Hustenepidemie" of Perin thus.
45. Littre, Oeuvrescompletes(above, n. 8), 5: 265. Italics mine.
46. P. Bretonneau, "Surles moyens de prevenir le developpement et les progres
de la diphterie," Arch. gen. de med.) 5th ser. 5 (1855):1-14; 6 (1855):257-79.
47. P. Bretonneau, Des inflammations specialesdu tissu muqueux et en particulier
la diphtherite) ou inflammation pelliculaire (Paris, 1826).
48. V.P.A. Maingault, De la paralysiediphterique (Paris, 1860).
49. E. Littre, "De la diphterite et de la paralysie consecutive a la diphtherite

clans les oeuvres d'Hippocrate," Gaz. mid. de Paris) 3d ser. 16 (1861):353-56.
50. A. Gubler, Des paralysiesdans leurs rapportsavec les maladies aigues) et speciale-
ment desparalysiesastheniques)diffuses) des convalescents(Paris, 1860).
51. Littre, Oeuvrescompletes(above, n. 8), 10: ii-viii.
52. See N. Mani, "Die Nachtblindheit und ihre Behandlung in der griechi-
schen-romischen Medizin," Gesnerus10 (1953):53-58.
53. D. Gourevitch, "Le dossier philologique du nyctalope," Hippocratica(Paris,
1979), pp. 167-87.
54. For example, Epidemics) IV, 12 (Littre, v, 150). In this regard, see J. Hirsch-
berg, Geschichteder Augenheilkunde im Alterthum (Leipzig, 1899), p. 88.
55. Littre, Oeuvrescompletes(above, n. 8), 10: vii.
56. Epidemics)II, 2, 24 (Littre, V, 94-98).
57. A. Souques, "Nature diphterique des paralysies post-angineuses attribuees
par les auteurs hippocratiques a la luxation spontanee des vertebres cervicales,''
Bull. Soc. Franf. Hist. Mid. 27 (1933):77-98; A. Souques, Etapes de la neurologiedans
PAntiquitegrecque (d)Homerea Galien) (Paris, 1938),pp. 57-68.
58. Littre, Oeuvrescompletes(above, n. 8), 5: 95.
59. A. Souques, "Nature diphterique de l'epidemie de toux de Perinthe,"
Bull. Soc. Franf. Hist. Mid. 28 (1934): 151-55.
60. See F. Schnurrer, Die I(rankheiten des Menschen-Geschlechtshistorisch und
geqgraphisch beobachtet (Tiibingen, 1823), 1: 45; A. Corradi, Annali delle epidemie
occorsein Italia dalleprime memoriefino al 1850(Bologna, 1865),1: 18ff.
61. Livy, History ofRnme) IV, 52.
62. G. Phokas, Peri tes par) Hippokrtitei epidemikis Gripes (Influenza) (Athens,
1892).
63. A. Laboulbene, "Sur un memoire de M. le Dr. Phokas: La grippe epide-
mique (Influenza) clans les ecrits hippocratiques," Bull. Acad. Mid. (Paris), 3d ser.,
28 (1892): 86-93; T. Puschmann, "Die Influenza im Alterthum," Wien. I(lin.
Wschr. 6 (1893):239-42.
64. A. G. Panayotatou, L)hygiene chez lesanciens Grecs(Paris, 1923),pp. 68-71.
65. H. Favier, "La dengue et la maladie de Perinthe," Gaz. hebd. mid. chir.) 2d
ser. 23 (1886):534-37 and566-69 (esp. p. 536).
66. E.W. Goodall, "On Infectious Diseases and Epidemiology in the Hippo-
cratic Collection," ltoc. Rny. Soc. Med. 27 (1934):525-34 (esp. p. 531).
67. J. R. Paul, A History of Polwmyelitis (New Haven, 1971),pp. 14-16.
68. A. Ilvento, Storia delle grandi malattie epidemiche con specialeriguardo al/a
malaria (Rome, 1938),p. 147.
69. See for instance F. Henschen, The History and Geqgraphyof Diseases (New
York, 1966), p. 46; G. Rosen, A History of Public Health (New York, 1948), p. 31;
W. I. Beveridge, Influenza) the Last Great Plague (New York, 1977), p. 25.
70. P. Lepine, "Les grandes epidemies de grippe," Vie medicateSI (1970): 4151.
71. Chamseru, "Recherches sur la nyctalopie ou l'aveuglement de nuit, maladie
qui regne tous les ans clans le printemps aux environs de La Roche-Guyon,"
Memoires de la SocieteRnyalede Medecine 8 (1786)(actually published in 1790): 130-78.
72. Ibid., p. 141.
73. Ibid., pp. 143-44.
74. Littre, Oeuvrescompletes(above, n. 8) 10:ii. See also Deichgraber, Die Epide-
mien (above, n. 25), pp. 34ff.
75. Nikitas, Untersuchungen (above, n. 36), p. 141.
76. I have chosen this case just as an example. Among the clinical histories
that, traditional opinion to the contrary notwithstanding, do not belong to the
"cough of Perinthus," I can cite Epidemics) IV, 25, 36, 47, SI, 52; VI, 1, 12.
77. Epidemics)II, 2, 8 (Littre, v, 88). For the establishment of the Greek text,
see Deichgraber, Dje Epidemien (above, n. 25), p. 69, and Langholf, Syntaktische
Untersuchungen (above, n. 24), p. rr7.
78. A diagnosis suggested by Goodall, "Infectious Diseases" (above, n. 66),
p. 53r.
79. First considered in the last century, especially by Puschmann, "Die Influ-
enza" (above, n. 63), p. 240, the diagnosis of neurotic paralysis is vigorously stated
by I. Veith, Hysteria) the History of a Disease(Chicago, 1965).
80. Epidemics)IV, 50 (Littre, V, 190).
Sr. Nikitas, Untersuchungen (above, n. 36), p. 135.
82. Epidemics)VI, 3, 8 (Littre, v, 296).
83. The distinction between IV, 50, and VI, 3, 8, was well drawn by Nikitas,
Untersuchungen (above, n. 36), pp. 132ff.I also owe the right diagnosis of the case
in VI, 3, 8, to him; Galen in his commentary on the passage had already suspected
it (CMG) v, 10, 2, 2, p. 149).
84. Epidemics)VI, 7, 9-10 (Littre, V, 342).
85. This way of interpreting the text agrees with Galen's commentaries (see
above, n. 83, p. 433), as well as those of Palladius and Foes.
86. Epidemics)IV, 53 (Littre, V, 192-94).
87. See especially Beveridge, Influenza (above, n. 69); M. M. Kaplan and R. G.
Webster, "The Epidemiology of Influenza," ScientificAmeriCan 237 (1977): 88-106;
and W. G. Laver, The Origin of Pandemic Influenza Viruses(Ne\V York, 1983).
88. For the chronology of the great influenza epidemics of the past, see above
all E. S. Thompson, Influenza) or Epidemic Catarrhal Fever (London, 1890); J. F.
Townsend, "History of Influenza Epidemics," Ann. Med. Hist. 5 (1933):533-47;
and G. Cavina, L)influenza epidemicaattraversoi secoli(Rome, 1959).For the serolog-
ical archaeology of influenza and the cyclical return of its epidemics, see N. Masu-
re!, "De wederkomst van het varkens-influenzavirus," Nederl. T&'dschr.Geneesk.
120 (1976): rr23-25; M. Bader, "Influenza Cycles," JAMA 237 (1977): 2813; and
especially Beveridge, Influenza (above, n. 69), pp. 77-79.
89. Dionysius of Halicarnassus, Antiquitates &manae) vn, 68; Livy, History of
Rome) IV, 52; Cassius Dio, History of Rome) XLV, 17. I consider it useless to include
the evidence of Diodorus Sicul us (XIV, 70) concerning the plague that smote the
Carthaginian army before Syracuse in about 396/395 B.c., although Townsend,
"History of Influenza" (above, n. 88), believes it was influenza. It is hard to
believe that Townsend has read the original text of Diodorus, since it mentions
symptoms that have nothing in common with the influenza syndrome.
90. The term k6ruza sometimes refers to the humor that flows from the nose,
sometimes to the flux itself, that is, the rhinorrhea (see Galen, Comm. II in Progn.)
Kuhn, XVIII, B, 180; Sympt. caus.) Kuhn, VII, 263). When phlegm from the brain
flows into the nostrils, that is k6ruza in the strict sense; when it flows from the
palate and then down the throat, it is called kattirrhos(cold with pharyngitis); when
the phlegm reaches the trachea, that is brtingkhos(cold with laryngitis and tracheo-
bronchitis). In this connection,.,.see C. Daremberg, Oeuvreschoisiesd)Hippocrate)2d
ed. (Paris, 1855),p. 170.
9r. Prognostic)14 (Littre, II, 146), and Praen. Coacae)393 (Littre, V, 670).
92. Ancient Medicine) 18 (Littre, I, 614; in Jones's edition, p. 46).
93. Aer.J10 (Littre, II, 46); Jones, I, 100.
94. For the clinical mention of labial herpes and genital herpes, see above,
chapter 5, "Other Diseases of the External Genitals."
95. Praen. Coacae)618 (Littre, v, 728).
96. See for example Rosen, History of Public Health (above, n. 69); J. F. H.
Broadbent, "The Acute· Infectious Diseases," in W. R. Bett, A Short History of
Some Common Diseases(Oxford, 1934), pp. 8-9; and E. H. Ackerknecht, Geschichte

und Geographieder wichtigstenI(rankheiten (Stuttgart, 1963), p. 60.
97. Morb.., III, 7 (Littre, vn, 124-26). See P. Potter, Hippokrates)Ueberdie J(ran-
kheiten III (Berlin, 1980), pp. 74-77 and 108.
98. Epidemics) I, r (Littre, 11, 600-602). See E. Ebstein, "Klassische Kranken-
geschichten: 11. Der Mumps bei Hippokrates," I(indertirtzlicheltaxis) 2 (1931):140-
41.
99. 0. Hamburger, "Un cas de paralysie infantile dans l'Antiquite," Bull. Soc.
Franf. Hist. Med. IO (1971):407-12.
100. Paul, Poliomyelitis(above, n. 67), p. 14.
101. See M. Michler, Die J(lumpfusslehre der Hippokratiker (Wiesbaden, 1963)
(Sudhoffi Archiv. Beiheft 2), p. 47.
102. Aristotle, Nie. Ethics) no2b.
103. A.H.W. Adkins, "Paralysis and Akrasia in Eth. Nie.," American Journal
of Philology97 (1976): 62-64.
105. Aretaeus of Cappadocia, Signa ac..,I, 9. See R. Bayeux, La diphterie depuis
Aretee le Cappadocienjusqu)en 1894 (Paris, 1899); B. 0. Hagedorn, Die Diphtherie in
ihremgeschichtlichenAujbau (Leipzig, 1919).
106. Dent.) 24 and 31; after the edition and translation by R. Joly (Paris, 1978),
pp. 224-25.
107. Acut (app.)) 10 (Joly, 72-73); Morb.) n, 26 (Jouanna, 159-69). Diphtheritic
sore throat is also described in Morb.) III, 10. In this connection, see Potter,
Hippokrates(above, n. 97), pp. 76-79 and no-n.
108. De visu) 7 (Joly, 171).
109. Bretonneau, "Surles moyens," (above, n. 46), p. 7.
ChapterI3: Diawgue benveena Phiwwgistand a Physician

1. K. Deichgraber, Die Epidemien und das CorpusHippocraticum)(Abhandl. lteuss.
Akad. Wiss.) Philol-hist. J(lasse)no. 3) (Berlin, 1933;rpt. Berlin, 1971),pp. 16 and 172.
2. I hereby thank my friend for his permission to use the presentation we made
together before the colloquium at Mons. See M. Grmek and F. Robert, "Dialogue
d'un medecin et d'un philologue sur quelques passages des Epidemies vn," Corpus
Hippocraticum (Mons, 1977), pp. 275-90. Although the first part of this chapter is
the same as the presentation at Mons, the second part contains significant addi-
tions, including a translation and a medical commentary on several clinical histories
that excel in the concision of their style and the astuteness of their medical
observations.
3. I alone am responsible for the content and final form of this chapter, but I
should perhaps stress that, although it represents a joint effort, the lion's share of
the work was naturally the philologist's.
4. E. Littre, Oeuvrescompletesd)Hippocrate(Paris, 1846), 5: 378-81.
5. Mis Codex Marcianus Graecus268 (tenth century), and Vis Codex Vaticanus
Graecus276 (twelfth century).
6. A tonic spasm is a continuous muscular contraction, while a clonic spasm is
a convulsion with a series of contractions.
7. Littre, v, 404-6. Parallel versions of these three case histories occur in
Epidemics)v, 74-76 (Littre, v, 246-48). For the first of these cases, see the section
"Four Concise Case Histories" in this chapter.
8. For instance, in Morb.) III, 12 (Littre, vn, 132; Potter, 80), "the tetanuses"
(hoi tetanoi) that lay low a patient and produce trismus, strabismus, and paralysis
in the limbs correspond to the modern definition of the tern1, if not from the
conceptual standpoint at least as regards the pathological substratum. The "teta-
nus" spoken ofin Acut. (app.)J37, 1 (Joly, 85)is, by contrast, only a special affection
of the hip.
9. This notion has been elaborated in F. Robert, "La prognose hippocratique
clans les livres v and VII des Epidemies," Le mondegrec (Hommage a Claire Preaux)
(Brussels, 1975),pp. 257-70. It is much indebted to the notion of "Kolleghefte" or
course notebooks used by Deichgraber in the book cited above (n. 1) and applied
by him to several texts in the Hippocratic collection.
10. Such is the case for the wife of Hermoptolemos; see the section "A Case
of Typhoid Fever" in this chapter.
II. Plato, PhaedrusJ253e, 2; Galen, in Kuhn, XVIII, B, 164.
12. For instance, see J. S. Milne, Su1lJicalInstruments in Greekand Rnman Times
(London, 1907), pp. 51-83, pls. II-14.
13. This important problem was the object of a stimulating, masterly discussion
by Jean Irigoin at the Mons colloquium, "Le role des recentiores clans l'etablisse-
ment du texte hippocratique," CorpusHippocraticum (above, n. 2), pp. 9-17.
14. History of Animals) 492b, 32, and Problems)902,b, 22.
15. Epidemics)VII, II (Littre, V, 382-86).
16. After consulting the Hippocratic concordance prepared by G. Maloney and
his Canadian team, I persist in this translation ofgiu6o by "cripple." Granted that
a certain softening in the word's meaning took place in medical parlance, the usual
translation of the word as ''weaken'' goes a little too far.
17. See F. Robert, "Prophasis," Revue des etudesgrecques89, (1976): 317-42.
18. Berrettoni's lexicon ("II lessico tecnico del I e II libro delle Epidemie ippo-
cratiche," Annali delta Scuola normale sup. di Pisa 39 [1970]: 9off.), after providing
"coma" as an initial translation of the word komaJgoes on to show the difference
between koma and sleep, since, in the former, consciousness is maintained.
19. The Hippocratic physicians were familiar with the automatic, unconscious,
and continual gesturing that affects some patients with fever: they seem to search
for nonexistent objects, grasp at their bedclothes, and tear out the threads of their
blankets. The physicians saw these gestures, with good reason, as bad signs for the
future, as likely precursors of death in patients who were delirious, victims of
phrenitis, pneumonia, or any other kind of acute fever. See Prqgn.J4, and Prorrh.J
I, 34. The term "carphology" (Gr. karpholog{a),which was introduced as a modern
medical term by Philippe Pinel, is used several times by Galen. He suffered an
attack of fever with this symptom during his own childhood. See Galen, Comment.
I in Hipp. Prqgn.J23 (Kuhn, XVIII, B, 71-75), and especially De loc. aff.J IV, 2 (Kuhn,
VIII, 226-27).
20. According to the Suda) there were three gegrees of I?ental disturbance: the
strongest was paraphrosuneand the weakest paralerosJwith lerosbetween the two.
2r. In Aff int.J 39-43 (Littre, VII, 260-74), distinctions are drawn between five
diseases called tuphos. None of them really corresponds to typhoid fever, although
we cannot rule out the possibility that the observation of actual cases of typhoid
contributed to the formulation of these five clinical profiles that, to the eyes of a
modern physician, are actually heterogeneous. The Hippocratic term tuphos has no
equivalent in modern pathology and should not be rendered by the word "ty-
phus." Typhoid fever was not understood as a clinical entity in classical antiquity.
Instead, it was confused with various forms of phrenitis, kausosJand bilious fever.
See G. Ongaro, "Evoluzione storica del concetto di tifo," Riforma medicaJno. 6
(1967).
22. There can be no doubt as to the presence of salmonella infections in the
ancient populations of the Mediterranean. I see the symptoms of salmonellosis in
several cases of kausosdescribed in Epidemics)I and III. The tenth patient in Epidem-
ics) I (the Clazomenean man who suffered from a strong, unremitting fever for
forty days) is a good example (Littre, II, 704-8). For arguments in favor of the
existence of salmonelloses in classical Greece, see E. W. Goodall, "On Infectious
Diseases and Epidemiology in the Hippocratic Collection," ltoc. Royal Soc. Med.
(London) 27 (1934):525-34 (esp. pp. 526-29).
23. The name "nystagmus" is given to slight, rhythmic, involuntary jerks of
both eyeballs (rarely, of just one). There is a succession of jerks, with the alterna-
tion of a rapid jerk in one direction and a slower jerk that returns the eye to its
original position. The direction is conventionally defined by the rapid one. Thus
the Hippocratic writer is defining left, horizontal nystagmus. It is the sign of a
lesion either in the inner ear or in the central nervous system (vestibular centers).
Infections can be its etiology. This complication, then, agrees with the diagnosis
of salmonellosis. I note in passing that the Greek word nustagmos occurs in ancient
texts in a different sense from the one just given. It refers to the way a sleeping or
dro\tvsy person shakes his head. The general view has been that the first clinical
descriptions of nystagmus date only from the eighteenth century. See G. Bilan-
cioni, "Alcuni spunti sulla storia del nistagmo," Boll. Ist. Stor. Ital. Arte San. 11
(1931):1-14.
24. Epidemics) VII, 116 (Littre, v, 462); parallel account in Epidemics) v, 101
(Littre, V, 258).
25. On the diagnosis of this case and the mammary cancer in antiquity, see P.
Diepgen, Die Frauenheilkunde der Alten Welt (Munich, 1937),p. 237, and G. Grassi,
Storia dei tumori nella antichitagreco-romana (Rome, 1941), pp. 28-30.
26. De mulierum affectibus) II, 133(Littre, VIII, 282); see also Gland.) 17 (Littre,
VIII, 572).
27. Herodotus, III, 133.
28. Among the modern authors who opt for a cancer diagnosis, I can cite J.
Korbler, Geschichteder I(rebskrankheit (Wien, 1973),pp. 8-10. The opposite opinion
has been put forth by A. T. Sandison, ''The First Recorded Case of Inflammatory
Mastitis: Queen Atossa of Persia and the Physician Democedes," Medical History 3
(1959):317-22. A prudent position, very critical of the cancer diagnosis, was adopted
by P. Menetrier and R. Houdry, ''La guerison du cancer de sein de la reine
Atossa," Bull. Soc. Franr;.Hist. Mid. 15(1921):285-89.
29. See M- P. Duminil, ''Le sens de ichor clans les textes hippocratiques, ''
Corpus Hippocraticum (above, n. 2), pp. 65-76; and, going beyond the medical
context, J. Jouanna and P. Demont, "Le sens d'ich6r chez Homere et chez
Eschyle en relation avec les emplois du mot clans la collection hippocratique,"
Revue des etudes anciennes 83 (1981):197-209.
30. Epidemics)VII, 36 (Littre, v, 404); parallel account in Epidemics)v, 74 (Littre,
v, 246-48).
31. Epidemics) VII, 113(Littre, v, 460-62); parallel account in Epidemics) v, 100
(Littre, V, 256-58).
32. Epidemics) VII, 32 (Littre, v, 400-402); parallel account in Epidemics) v, 60
(Littre, v, 240).
33. See S. Winkle, "Der Tetanos im Altertum," Die gelben Hefte 18 (1970):
916-28.
34. See F. Robert, "La bataille de Delos (Hippocrate, Epidemies, v, 61 and VII,
33)," Etudes deliennes (Suppl. I du Bulletin de CorrespondanceHellinique) (Athens,
1973),pp. 427-33.
35. Epidemics)VII, 46 (Littre, V, 414).
36. For the history of epilepsy in antiquity, see especially 0. Temkin, The
Falling Sickness)2d edition (Baltimore, 1971).
37. For the translation of Greek peripneumon{a(or peripleumon{a)by the modern
term ''pneumonia,;' see above, p. 307.
38. Epidemics)VII, 13 (Littre, v, 388).
39. Epidemics)VII, 55 (Littre, v, 422).
40. See A. Papanicolaou, Glossikaiereunaiepi tou CorpusHippocraticum(Athens,
1965).
Index
Abrasion of teeth. See Dental abrasion Andersen, J. G., r6r, 396

Abscess: dental, 122; of the lung, 131; peri- Andre, J., 4m, 413
nephric, r3r; periodontal, r2r; peritonsillar, Andre, L. J., 430
130 Andrews, A. C., 4n
Abscessed osteitis of tibia and fibula, r2r Angel, J. L., 49, 52, 55, 65, 71, 72, 74, 79, 82,
Abul-Qasim, 172 92, 96, 99, ror, m2, m3, ro6, n4, n5, n7,
Achondroplastic dwarfism, 75 122, 182, 247, 259, 270, 275, 370, 379, 380,
Ackerknecht, E. H., 49, 142, 144, 357, 370, 382, 383, 384, 387, 388, 397, 402, 418, 420,
371, 394, 427, 436 424, 425; major publications, 372 n. 31
Acromegaly, 75 Angelini, V., 419
Actinomycosis, 86 Anglicus, Bartholemew, 84
Actuarius, J., 297 Angular kyphosis, ambiguous symptom of
Adachi, B., 248, 418 Port's disease, 178
Adamantiades-Beh<_;etsyndrome, 146 Ankylosing spondylitis, 50, 84; common-
Adams, F., 291, 428, 429 ness in Egypt, 85; genetic factors in, 85
Adenitis, sub maxillary, 130 Ankylosing vertebral hyperostosis, 80, Sr
Adkins, A. H. W., 436 Ankylosis: of hip, 71; of thumb, 122
Aeneas, 29 Anophelesatroparvtts)279
Aerogenes)129 Anopheles labranchiae)279
Aeschylus, 164 Anophelessacharmn)279
Agenesia, dental, 74 Antinoos of Bithynia, 83
Ajax, Telamonian, 30 Antoniades, A., 298, 302, 430
akreton)286 anastasis)313
Albarracin Teul6n, A., 361 Aphasia ( = amphasfe), 34
Alcaeus, 43 Aphthae, in Hippocrates, 149
Alcibiades, m8, 316 Aphthosis, 149
Alciphron, 69 Apion, 151
Alcmaeon of Croton, 40 Appelboom, T., 379
Alexanderson, B., 431 Archilochus of Paros, 42-43
Alexis ofThurioi, age of, I09 Aretaeus of Cappadocia, 143, 192; describes
Alfieri, N., 259, 422 leprosy, 170
Allison, A. C., 232, 265, 415, 420, 423 Arie, T. H. D., 417
Allison, M. J., 402 Ariston, 42
Alpagut, B., 79, 378 Aristophanes, 35, 146
alph6s) 165 Aristotle, ro7, 3n
Anacreon, 44 Armelagos, G. J., 247, 383, 418, 421
anakaleomino) 287 Arrows of Apollo and Artemis, 22
441
442 INDEX
Arta region, Greece, 255, 266 Beaugrand, E., 291, 429

Artemidorus Capito, 312 Bechelli, L. M., 408
Arthritis: degenerative (seeOsteoarthritis); Bechterew's disease. See Ankylosing
septic, 83; terminology of, 378 spondylitis
Ascenzi, A., 269, 424 Becker, M. J., 384
Ascites, symptom of tuberculosis, 195 Beer, G. de, 383
Asclepius, 46 Beh<;et's disease, in Hippocratic writings,
Astaldi, G. , 419 146, 394
asthma) 33 Bejel, 134, 139, 142
asthrndzo) 33 Belloni, L., 420
Asymmetry of face in statues: aesthetic, 68; Beloch, J., 382
cerebral, 67; realistic, 68 Belos, G. D., 426
Atchison, J., 358 Belser, C. W., 159
Atossa: wife of Artaxerxes, 399 n. 64; wife Belsey, M. A., 415
of Darius, 350 Benabadji, M., 415
aukhin, 312, 343 Benacerraf, R., 420
Austin, M. A., 381 Benassi, E., 269, 380, 387, 424
Average age at <leath, 99; in classical period, Benedetti, A., 373
ro3; in Crete vs. mainland, ro3; and gen- Benedicenti, A., 4ro
der, roo, ro3; historical analysis, roo; in- Benson, J. L., 56, 424
creases in, ror; from Neolithic to Roman Benveniste, E., 46, 369
times, ro4; and progress, roo; and social Berard, J., 383
status, ro2; statistical value of, ro5 Berendes, J., 4ro
Average height: fluctuations in, ro9; and ge- Berg, F. D., 148
netic polymorphism, no; vs. heredity and Bergen's syndrome, 154
nutrition, no; and nutritional changes, no Bergfelder, T. , 382
Average life expectancy, ro4 Bergh off, E., 357
Average lifespan, ror Bergmark, G., 83, 379
Ayala, S. C., 424 Berheim, F., 362
Beriberi, 75
Bacteroides)129 Bering Strait, 137, 139
Bade, P., 375 Bernard, E., 405
Bader, M., 435 Bernard, J., 264, 415, 421, 423
Badetti, L., 258, 422 Bernard, R., 397
Baffoni, A., 357, 391, 403 Bernhard, 0., 417
Bailey, N. T. J., 358 Berrettoni, P., 286, 428, 437
Balanced polymorphism, 232, 265; malaria Besques, S., 156, 397
and, 266; sickle cell anemia, thalassemia, Beswick, T. S. L., 395
266 Bett, W. R., 357
Balistreri, P., 269, 424 Beutler, E., 413, 414
Ballas, C. N., 239, 417 Beveridge, W. I., 434
Bannerman, R. M., 419 Bianco, I., 256, 259, 422, 423
Barbezieux, G., 173 Bible, ro8
Barbillon, 391 Bieber, M., 374
Barghoorn, E. S., 120, 388 Bienzle, U., 232, 300, 415
Barker, L. F., 430 Bignami, A., 299, 430
Barlow's disease, 77 Bigourdan, G., 432
Barnicot, N. A., 421 Bilancioni, G., 438
Bartels, P., 179, 402 Bilharziasis, absence of, 89
Bartonellosis, 57 Biological constancy, 8, 9; from classical to
Bartsocas, C. S., 25, 56, 70, 375, 376, 384 modern Greece, 14
Basedow's disease, 12-13 Biraben, J. N., roo, 384
Basset, A., 4or, 408 Birthrate, ro2
Baty, J. M., 259 Bisel, S. C., n3
Baudouin, M., 378 Bismuth, V., 420
Baumann, E. D., 359, 366, 391, 403 Bitot, 432; Bitot's pearly spots, 3n
Bayeux, R., 436 Bitter vetch: food in famine, 222; known
Bazin, E., 391 toxicity of, 223
BCG vaccination, 203 Black, F. L., 383
INDEX 443
Black urine, 285-86, 295 counts of, 233-38; Orphism and, 218; other
Blackwater fever: etiol'ogies, 299-300; falcip- ancient accounts of, 220; overdetermined,
arum malaria and, 299; in Greece, 297-98; 240; political explanation of, 219; psychic
Hippocratic vs. modern accounts, 3or-2; disturbance and, 221; Pythagorean interest
history, 296-99; quinine intoxication and, in health and, 243; in ritual and myth,
298 214; sacred diets and, 243; silence of Hip-
Blastomycosis, 86, 157 pocrates on, 221; structuralist account of,
Blegen, C. W., 56 237-39; trampling and, 216; transmigration
Blindness: and eye disease, 26; of poets, of souls and, 215, 218; unhealthy in Hippoc-
25-26 rates, 222; Viciafaba L., 2u-12
Bloch, I., 157, 392, 397 Broadbent, J. F. H., 435
Blum, Eva and Richard, 14, 233, 359, 409, Broca, P., 48
416 Brodny, M. L., 144, 394
Bobin, V. V., 122 Brody, S. N., 4or
Bochalli, R., 4or Brommer, J., 385
Bogen, J. E., 67 Brommer, S., 385
Bollack, J., 389 Bronchopneumonia, 131
Bonhorst, C. W., 387 Brothwell, D. R., 138, 370, 380, 385, 392, 393,
Bonnard, A., 43 396
Bonuzzi, L., 416 Broth well, P., 385
Border, J., 335 Brown, J. A. C., 358
Borg, J., 413 Brown, T. S., 368
Boross, M. M., 376 Browne, S. G., 160, 161, 398, 4or
Bottini, E., 414 Bruce-Chwatt, L. J., 277, 425
Botto-Micca, A., 361 Brumbaugh, R. S., 417
Bouba. SeeYaws Brumpt, L., 257, 261, 265, 422
boubones,146 Brundin, L., 141
Bouillaud's disease, 83 Bucchi, E. C., 264, 423
Boule, M., 80 Buchanan, W.W., 379, 380
Bourgeois, A., 423 Buess, H., 361, 378
Bourgey, L., 403, 427, 433 Bugard, P., 262, 422
Bourke, J.B., 378,402 Bungarten, J. I., 375
Bourquelot, E., 234, 416 Bunn, H. F., 420
Boyer, S. H., 414 Burckhardt, J., 381
Brabant, H., 385 Burger, M., 358
Brachia! plexus, lesion of, 32 Burke, R., 4or
Bradford, W. L., 249, 419 Burkert, W., 4ro
Braeutigam, W., 431 Burko, H., 421
Brain, P., 264, 423, 425 Burleigh, R., 393
Brandenburg, D., 369 Burn, A. W., 384
Brass, W., 380 Burnet, M., 383
Brauer, G., 375, 425 Butzer, K. W., 380
Braumiiller, 0., 361 Byl, S., 431
Bread and wine, m
Breitinger, E., 55, 58, 372 Cabanes, A., 360
Brendel, A., 37, 366 Caffey, J. P., 420
Bretonneau, P., 321, 338, 433 Callimachus, 213
Brewerton, D., 380 Calvus, 313
Broad bean tabu, 2n-44; as antidote to ma- Cambyses II, 41, 76, ro8; death of, 126
laria, 242; Aristotle's accounts of, 217; Ar- Cameron, R., 120
istoxenus' account of, 219; basic issues Caminopetros, J., 260, 419
about, 240; celery tabu and, 234; Dioscor- Camp, J.M., 381
ides on, 221; eating and, 213; in Egypt, 215; Cancer: breast, 72, 350; bronchopulmonary,
favism and, 233-44; flatulence and, 219; 187; of the penis, 151; of the vulva, 151
folk medicine and, 233; in Horace, 220; Candida albicans) 148
Iamblichus account of, 217; in India, 214; Candidiasis, in Hippocratic writings, 147
L-dopa in, 221; Lucian 's account of, 218; Candolle, A. de, 215, 4ro
medical accounts of, 238-39; modern ac- Cantani, A., 223, 413
Capasso. L., 375
444 INDEX
Capecchi, V., 359 Chini, V., 259, 419

Caplan, A. L., 357 Chiron, 46
Caramalea, V. V., 384 Chlamydia trachomatis)150
Carate. SeePinta Chlorosis, 148
Carbuncle, 122 Choremis, C., 415, 422, 423
Carcassi, U. E., 266, 414, 415 Churchill, E. D., 389
Cardamatis, J. P., 302, 426, 431 Cicero, 26, 219
Caries, dental. SeeDental caries Cirrhosis of the liver, 42, 44
Carlson, D. S., 246, 418 Cladogram of mycobacteria, 208
Carney, T. F., 392 Clark, D. H., 432
Carphology, 348 Clarke, S. , 381
Carr, H. G., 56, 387, 388 Clauss, M., 384
Carson, P. E., 414 Clavolella, M., 369
Casanova, A., 358 Cleanthes, age of, ro9
Case, J. T., 418 Cleft lip, unilateral, 70
Caskey, J. L., 56, 64, 270 Clegg, J. B., 419
Cassidy, C. M., 378 Cleidocranial dysplasia, 70
Castellani, A. , 134 Clement, A. J., 387
Castels, M., 409 Clementelli, M., 384
Castiglioni, A., 4or Clement of Alexandria, 220
Cataleptic rigidity, 20 Cleomenes, 44
Caughy, D. E., 379 Clostridium) 129
Cauvin, J., 381 Clostridium tet:ani)341
Cavaignac, E., 382 Clubbing, 131, 187
Cave, A. J. E., 402 Clubfoot, 71; acquired, 337; congenital, ro
Cave gout, 48 Coatney, G. R., 425
Cavina, G. , 435 Cochrane, R. G., 161, 398, 400
Cawadias, A., 426 Cockburn, T. A., 137, 174, 199, 209, 359, 378,
Cebriones, 29 381, 392, 4or, 407
Celli, A., 280, 427 Code ofHammurabi, 159
Celsus (physician), 35, 150, 169, 193 Cohen, M. N., 381
Celsus, 215 Cohen, R., 383
Cepellini, R., 265 Cohn, D. F., 413
Cerebral hemiatrophy, 69 Colar-blindness, 27
Cerebral palsy, 68 Coluzzi, M., 426
Chairemon, 162 Comby, J., 249
Chairopoulos, G., 427 Common cold, 334-36
Chamoux, F., 381, 406 Concussion, 30
Chamseru, Dr., 327-28, 338-39, 434 Condylomas, 150
Chancroid, 150 Conley, C. L, 420
Chantraine, P., 389, 398, 403 Constantine IX of Byzantium, 84
Charbonneaux, J., 381 Constitution, epidemic, 288, 320, 338; of
Charcot, J.-M., 377 Thasos, 189, 289, 302
Charitonidis, S., 374 Consumption, 178; ambiguous symptom of
Charles, R. P., 56, 58, 63, 65, 71, 79, ro3, n4, pulmonary tube,rculosis, 179; in Homer,
II7, 372, 384, 388 36; in Sophocles, 36
Charmot, G., 421, 430 Convulsions, 41
Charokopos, S., 420 Cooley, T. B., 249, 419
Chaussinand, R., 203, 4or, 408 Coombs' test, 300
Cheiloschisis, 70 Cootjans, G., 387
Chernoff, A. I., 422 Copeland, D. D., 357
Cheyne-Stokes breathing, 287, 302 Cordier, P., 402
Chickenpox, in Hippocrates, 335-36 Corlieu, A., 369
Chickling pea or vetch, as food in famine, Cornarius, 309
223 Corner, G. W., 249
Chih, L. M., 398 Corradi, A., 434
Children, malnutrition of, n2-13 Corre, A., 298
Childs, B., 414 Corvisier, J. N., 383
INDEX 445
Corynebacteriumdiphtheriae) 337 Dental caries, 115-18:and childhood disease,
Coryza, 334-36 118;in classical period, 117;factors in, 116;
Casson, H., 369 and fluoride in water, 118;in Greece vs.
Cough of Perin thus, 305-39; diphtheria in, Europe, 116;measurement of, 116;rate of,
321, 324; encephalitis in, 326; influenza in, 117;and sugar in diet, 117
325, 333-34; Littn~'s view, 319; one or many Derry, D. E., 396
diseases, 320, 328; other cases of, 330-33; Desmars, 427
other diagnoses of, 326-27 Desnos, E., 391
Cougoul, J., 400 Detienne, M., 235, 237-39, 416
Courbin, P., 56 Dharmendra, 397
Courtade, A., 391 Diabetes, 12
Coury, C., 195, 401, 402, 405 Diarrhea, in pneumonia, 354
Cranial capacity, of Greek skulls, 54 Di Benedetto, V., 315,318,432
Cribra orbitalia, 246 Dichuchwa, 134, 139, 142
Crohns, H., 369 Diebert, K. R., 420
Crosby, W. A., 228, 414 Diepgen, P., 148, 150, 357, 391, 395, 405, 438
Croup, 313 Dierbach, J. H., 410
Culliton, B. J., 420 Dietz, F. R., 431
Cunnilingus, retribution for, 143 Dietz, S., 56
Cushing's syndrome, 75 Diffuse idiopathic skeletal hyperostosis
Cutbush, M., 264, 423 (DISH), 81
Dikaios, P., 56
Daly, L. W., 390 Diller, H., 428
Dambassis, J. N., 368 Dimbleby, G. W., 385
Daremberg, C., 27, 36, 37, 43, 45, 188, 189, Dio Chrysostom, 26
234, 286, 361, 368, 403, 427, 435 Diodorus Siculus, 163
Dark Ages, 91 Diomedes, 30
Dastugue, J., 56, 69, 372, 374, 380, 388 Dioscorides, 314
Dausset, J., 50, 370 Diphtheria: consecutive paralyses in, 321;
Daux, G., 56, 416 history of, 337-38; in Hippocrates, 338
Davey, K. W., 421 Di Pierro, P., 424
Debra, L. M., 421 Diplococcus)121
Deevey, E. S., 382 Disease names of unknown etymology, 360
Deficiency: of amino acids, m; G6PD, 228- n. 3
29, 265; nutritional, in children, 112;of vi- Diseases: absence of evidence for, 155,162;
tamin A, 75, 112;of vitamin B2 , 75; of vita- acquired and inborn in, 192; anatomical
min C, 77; of vitamin D 2 , 75; of vitamins, concept of, 2; antagonistic, 3; climate and,
III
38, 39; clinical concept of, 2; concept of, 1;
Deformity, pathological, 45: absence of, 76; demographic thresholds for, 87, 98; deter-
artistic representation of, 377 n. 114; mining vs. predisposing causes of, 226;
caused by rickets, 75; pelvic, 76; skull, 74 distribution of, 3, 4; division of labor and,
Degenerative joint disease. SeeOsteoarthritis 89; eastern Mediterranean and, 92; envi-
Deichgraber, K., 184, 289, 316-17, 331,340, ronment and, 11-14, 24, 93, 135,210, 253;
399, 403, 428, 432, 436 environment and conception of, 1; evi-
Delatte, A., 216, 235, 236, 410, 416 dence from art and its interpretation, 155;
Delpeuch, A., 379 heredity and, 41; Hesiodic concept of, 39;
Democedes of Craton, 351 Hippocratic conception of, 40; impurity
Demographic analysis of epitaphs, 105; sta- and, 35; individual constitution and, 190;
tistical value of, 106-7 Inda-European conception of, 46; infec-
Demographic density: decline in second tious, 13, 85-86; innate and acquired in, n,
century B.C., 98; of eastern Mediterra- 210; invisibility of, 210, 224, 241-42; mate-
nean, 95; in Greece vs. Europe, 95; and rial evidence for, 2; morality and, 5; myth-
the Inda-European invasion, 96; from ical concept of, 5; nature and, 359 n. 16;
Neolithic to Bronze Age, 96; by regions, Neolithic revolution and, 88; ontology of,
97; statistical accuracy and, 97 1, 40, 293; Pythagorean concept of, 40;
Demont, P., 438 race and, 90, 246; relative frequency of, 3,
Dental abrasion, 78, 113;in men vs. women, 4, 14; of sailors, 36; silence about, 11,12,
115;relative measurement by a single ob- 33, 36; social change and, 139, 173, 200~201,
server, 114; subjective measurement of, 114 282-83; spread of by sea, 89; suppurative,
INDEX
Diseases (con)t) Enselme, J., 388

n9; in symbiosis, 3; systematic view of, 2; Entamoeba hystolytica)354
terminology, 167, 249, 289, 345 Enterobacteriaceae, n9
DISH, 81 Enterococcus)n9
Disk, herniated, 82 Eosiniphilic granuloma, 86
Dislocation of hip, congenital, Hippocratic Epeios, 30
account of, 71 Ephorus, 25
Dobzhansky, T., 423 eptalos)45
Doerr, W., 359 Epidemics, and overpopulation, 90
Domen, R. E., 379 Epilepsy, grand mal, 40, 353-54
Donham, K. J., 408 Epitaphs, demographic analysis of, ro5,
Dont, H., 404 I06-7
Dorian invasion, 91 Ergotism, n3
Dorr, W., 400 Ericson, J. E., 370
Down's syndrome, 9, 75 Erlandson, M., 421
Doxiadis, S. A., 231,415 Ernstedt, P. V., 360
Drepanocytosis. SeeSickle cell anemia erus[pelas)129
Dreyfus, J. C., 414 Erysipelas, 129
Dropsy, 41 Escherichiacoli) n9
du Boulay, J., 359 Esper, J. F., 47, 370
Dubos, J., 409 Esser, A., 363
Dubos, R., 359, 409 Etienne, R., 384
Ducas, J., 414 Eubacteria) 120
Duday, H., 56, n8, 372, 387, 388 Euler, H., 387
Dugand, J.-E., 428, 433 Euryalos, 0., 30
Duguet's angina, 130 Evolution, of host, 136
Duminil, M.-P., 389, 483
Dumortier, J., 398 faba) faba Aegyptia) 212
Dumezil, G., 238 Facies leontina, 157
Dunn, F. L., 424, 425 Facies leprosa, 154
Durand, J. D., 384 Fahlbusch, W., 392
Dysentery, amebic, 354 Fasbinder, H., 391
Dysostosis cleidocranialis, 25 Favier, H., 326, 434
Dzierzykray-Rogalski, T., 154, 396 Favism, 224-44, 230; advantage of, 232; an-
tiquity of defect, 232; discovery in Siciliy
Eade, 323 of, 225; etioloy of, 228; first discovered in
Ebbell, B., 159, 363, 390, 398 Portugal, 224; in folk medicine, 225, 233;
Ebstein, W., 376, 377 G6PD deficiency and, 228; geographic dis-
Edelstein, E., 405 tribution of, 231;hemoglobinuria in, 225-
Edelstein, L., 405 26; hemolytic quinine intolerance and,
Edema: hypoproteinemic, 41; starvation, 39, 227; heredity and, 226, 230; other plant-
43; vulvar, 43 induced hemolyses and, 229; primaquine
El-Najjar, M. Y., 268, 393, 4or, 402, 424 intolerance and, 227; toxic factor in broad
elephantia) 169 beans and, 227
elephantiasis)first Greek term for true lep- Fay, T., 70, 374
rosy, 168 Fecundity (demogtaphic statistic), 97
eliphas) 168 Feigenbaum, A., 394
elephasnwrbus) 169 Feingold, B. F., 418
Empedocles, 124, 2n, 220 Felty's syndrome, 83
empuema) 124 Ferembach, D., 425
Empyema, 131,187 Fessas, P., 419, 421, 423
Enamel, dental: hypoplasia, n8; mottled, n8 fu:us) 150
Encephalitis, viral, 335 Fiehrer, A., 419
Endemic diseases, and history, 16 Filliozat, J., 397
Endemic syphilis, 134, 139, 142 Finet, A., 398
Endemic trachoma, 150 Finley, M. I., 381
Endt, D. W. von, 421 Fischer, P. M., 373
Engelhardt, H. T., 357 Fistula, vesico-rectal, 45
Enrique Laval, M., 239, 417 Flaceliere, R., 381
INDEX 447
Fleger, J., 394 Garrod, A., 72
Floch, H. 408 Bab. garabu) 159
Floguet, A., 366 Gasbarrini, A., 413
Foesius, A., 287, 308, 389, 428 Gas gangrene, 287, 390 n. 57
Fontanille, M.-T., 395 Gasperini, C. G., 415
Food and nutrition, 111 gastrfe) 45
Food supply, 113 Gatto, I., 257, 268, 422
Forestier, J., 378 Gazzaniga, M. S., 374
Formation of pus, 119; Hippocratic views Gear, J., 300, 430
on, 124, 125, 126; thought "normal," 119 Gejvall, N. G., 6, 70, 375, 393
Fornaciari, G., 113, 270, 421, 424 Gemmill, C. L., 359
Fornara, P., 420 Genetic polymorphism, no: and average
Forrestier, J., 80 height, no; of Greeks, 94; and health, 90
Fossil bones, infected with pyogenic bacte- Gengou, 0., 335
na, 51, 120 Germain, J.-B., 291, 429
Foy, H., 302-3, 430 Gernez-Rieux, C., 407
Fraccaro, P., 426 Gertler, H., 403
Fractures, 59; of collarbone, 59; distribution Ghinopoulo, D., 395
by gender, 59; of femur, 61; frequency of, Ghinopoulo, S., 359, 407
57; of frontal bone, 57; greenstick, 61; of Giedke, A., 369
humerus, 59; inflammation around, 62; of Gigantism, 75
jaws, 59; of metacarpals, 61; of nose, 59; Gilbert, J. B., 360
from parrying blows, 60; of radius, 61; rar- Giles, E., 417
ity in children, 59; of rib, 61; of skull, 57, Ginouves, R., 374
58; of spinal column, 62; of temporal Glaser, S., 359
bone, 57; of ulna, 60; of vertebral bodies, Glotz, G., 381, 383
62; well-healed, 60, 61 Gluck, L., 153, 394, 396
Frambesia. See Yaws Goff, C. W., 393
Frankel, H., 367 Gaiter, exophthalmic, 12-13
Fraser, G. R., 255, 258,415,422 Golden Age, 38
Fraser, J., 404 Goldman, I., 398
Fraser, R., 418 Goldman, L., 358
Frazer, J. G., 416 Gomme, A. W., 383
Frazer, R. M., 358 Gonococcus, 144
Fredrich, C. J., 289 Gonorrhea, 123, 144; evidence for in Hippo-
Freedman, M. L., 420 cratic writings, 145
Fribourg-Blanc, A., 392 gonorrhoia)145
Fricke, R., 375, 425 Goodall, E.W., 327, 434, 438
Friedrich, W. H., 361 Gorgias of Leontini, age of, 109
Frisk, H., 403 Gorjanovic-Kramberger, D., 80
Froe, A. de, 393 Gourevitch, D., 275, 323, 357, 375, 407, 434
Frohlich, H., 27, 364 Gourevitch, M., 357, 375
Frumin, A. M., 228, 414 Gout, 72-73
Fuchs, C. F., 429 Gouttas, A., 422
Fuchs, R., 427, 431 Grapow, H., 398
Fuld, E., 366 Grassi, G., 438
Furuncles, 129 Graves' disease, 12-13
Furst, C. M., 55, 79, 81, 103, 117, 371, 387, 388 Graziosi, P., 268, 424
Fusion: of cervical vertebrae, 53; of spinal Greco, E., 377
vertebrae, 62 Gregory ofNazianzus, 172
Grensemann, H., 315, 368
Gagniere, S., 393 Greppi, E., 249
Galen, 167, 287, 309, 427 Grimm, H., 377, 378
Galeone, A., 376 Grmek, M. D., 357, 358, 360, 370, 375, 378,
Gallo, P., 358 379, 382, 384, 387, 391, 408, 417, 427, 436
Gallstones, 7 4 Grober, J., 207, 409
Gangrene, osseous, 352 Gruber, G. P., 357
Ganter, R., 366 Gruby, D., 148
Garnham, P. C., 425
INDEX
Gruner, C. G., 357 Heraclitus of Ephesus, 41

Gubler, A., 321, 434 Herdan, G., 358
Guerra, F., 391 Hermaphroditism, 75
Guiard, E., 374 Hernia, 43
Guillain- Barre syndrome, 324 Herodotus, 41, 46, 76, m8, 126, 127, 164;
Guillin, A., 430 medical terms in, 351
Guinot, D., 409 Herpes simplex, 149
guioo) 347 Herpesvirus)149, 335
Gundel, 432 Herpes zoster, in Hippocrates, 335-36
Gurgjian, E. S., 373 Herrick, J. B., 251, 420
Gurlt, E. J., 373 Herrmann, B., 382
Gurrarino, G. , 421 Herzog, R., 406
Gwei-djen, L, 398 Hesiod, 5, 38
Hesse-Turner, D., 423
Haas, N., 421 Hieronymus of Cardia, age of, m9
Habrich, C., 358 Hippocrates, 319
Hackett, C. J., 137, 392, 393 Hippocratic Epidemics)284-304; composi-
Hackett, L. W., 427 tion, 317-19, 352, 355; value of, 340, 352
Hadjimarkos, D. M., 56, 387 Hippocratic medicine: aesthetics of, 347;
Haemophilus ducreyi)150 aphorisms, 187, 190, 295, 329, 332; climate
Haemosporidia, 277 and character in, 94; clinical and philolog-
Hafliger, H., 359 ical study of, 340; Cnidian vs. Coan com-
Hagedorn, B. 0., 436 munities, 189; conception of disease, and
Haldane, J.B. S., 232,265,423 environment, 15; concept of epidemic,
Hall, H. J., 370 320; concept of disease, and individual
Hamburger, 0., 436 habitus, 190; conceptualization process,
Hammond, D., 419 332; diagnosis, 289, 292-95, 349; environ-
Hammond, N. G. L., 381 ment and, 2m; exercise and, 2m; function
Hamper!, H., 419 of katastasis)329; immunity in, 312; kausos)
Hansen, A., 152 290; limits of its value, 241; massage in,
Hansen, M. H., 383 313; muscles in, 345; no dissection, 354,
Hare, R., 200, 388, 407 pharynx and larynx in, 337; prediction in,
Harper, R. M. J., 358 351; prognosis, 187, 190, 292-95; pulmonary
Harris, H. A., 370 tuberculosis and angular kyphosis related
Harris' lines, 49, 113 by, 194; quality of observation in, 187, 188;
Hart, G. D., 377 reasoning process, 294; religious thought
Harter, P., 4m, 408 and, 220; its rise and pathocoenotic dis-
Healing lines, on female pubic symphysis, equilibrium, 92; stylistic concision of, 224,
49 3m, 349; taxonomy, 293; terminology vs.
Heart failure, 42 modern, 345; truncated observations in,
Hebenstreit, J. E., 360 314, 342, 352-54
Heiberg, J. L., 366 Hipponax of Ephesus, 45
Heichelheim, F., 380 Hirsch, A. , 357
Heidenreich, R., 363 Hirschberg, J., 394, 434
Hein, W. H., 360 Histiocytosis X, 86,
helkos)33, 125 Histocompatibility of ancient tissues, 50
Belmont, J. B. van, 287 HLA (human leukocyte antigen) system, 50
Hemoglobinosis E, 265 Hoggan, F. E., 397
Hemophilia: geographic distribution, 11; Hollander, D. H., 392
silence about, n Hollander, E., 376
Hemorrhage, intracranial, 352 Holk, C., 416
Hemorrhoids, 12 Hombert, M., 384
Hengen. 0. P., 421 Homeric world: absence of chronic disease
Hennig, W., 141, 409 in, 33, 36; accidental death in, 21; blood in,
Henry, L., m7, 384 23; color perception in, 364; concept of
Henschen, F., 56, 70, 142, 357, 375, 393, 394, death in, 20, 21, 22; consumption in, 36;
434 health and environment in, 24; health in,
Heracles, 41 23; old age in, 24; pathology of, 20, 21, 22;
Heraclides Ponticus, 26 wounds in, 27-33
INDEX 449
Hooton, E. A., 247,418 John of Alexandria, 308

Hopkins, K., 384 Joly, R., 301, 359, 382, 430
Horace, 42 Jonckheere, F., 402
Houdry, R., 438 Jones, A. H. M., 384
Howe, M., 359 Jones, E., 416
Hrdlicka, A., 181, 246, 418 Jones, R., 95, 382
Huard, P., 398 Jones, W., 73, 80, 85, 248, 376, 396, 418
Huber, J.C., 413 Jones, W. H. S., 39, 279-83, 285, 291, 301,
Huck, J. G., 420 302, 367, 426, 427, 429
hudur (hudrops),41 Jonquieres, E., 396
Hudson, E. H., 135, 392 Jopling, W. H., 408
Hufeland, C. W., 321 Josephus, 151, 162
Hughes, D. R., 396 Jouanna, J., 147, 395, 400, 412, 428, 438
Hull, D. L., 409 Jurmain, R. D., 378
Hulse, E., 161, 398 Justinus, 163
Hunain ibn Ishaq, 310, 312-13
Huntington, E., 427 Kahil, L., 374
Huntsman, R. G., 420 Kail, F., 393
houskuamos)213 Kapferer, R., 392, 428, 431
hupopofkila) 3n Kaplan, J. C., 414
Huxley, G., 377 Kaplan, M. M., 435
Hydrocephalus, 75 Kaposi's sarcoma, 151
Hypertension, intracranial, 122 Karamitsas, G., 298, 430
karkfnoma)350
Iamblichus, 213, 216 Karolyi, L. von, 370
Idiopathic bladder stone disease, n2 katastasis)189, 288, 305, 323, 325, 329-32
ikh7Jr,351 Kattamis, C. A., 231, 415, 421
Ilvento, A., 327, 434 Kattamis, M. D. C., 415, 422
Inborn error of metabolism, 9, 10, 12, 72, kausos)289-92
229, 249, 265 Katz, S., 242, 417
Infantile spastic paralysis, 69 Keaveny, A., 396
Infant mortality, 99, 100; Mycenaean, 102 Kellett, C. E., 390
Infection. SeeFormation of pus; Purulent Kelley, M.A., 401, 402
inflammation Kelley, W. N., 378
Infectious diseases, 93; and demographic Kerenyi, K., 4n
thresholds, 98 Kerkhoff, A. H. M., 361
Influenza, 333-34 khimetla) 4 5
Ingram, G. I. C., 359 kholira) 7
Injo, L. E. L., 421 Ved. kiliisa) 157
Intoxication, alcoholic, 44 Kind, 427
Irigoin, J., 437 Kirkman, H. N., 414
Isocrates, age of, 109 Klebs, E., 337
isrubu) 159 Klebsiella)120
Iversen, I., 373 Klepinger, L. L., 84, 380
Klippel-Feil syndrome, 70
Jacobsen, T. W., 56, 57, 372 Koch, H. T., 394
Jacqueline, F., 378 Koch, R., 177, 298
Jalongo, A., 384 Koel bing, H., 378
Jamalian, J., 414 Kohn, K., 359
James, D. C., 370 Ko Hung, 158
Jansen, H. H., 359 koma)348
Jansen, J., 370 Kondi, A., 302-3, 430
Jaquez, J. A., 358 Korbler, J. , 438
Jarcho, S., 383, 393, 418 Korner, 0., 33, 37, 361
Jarde, A., n3, 385 Kosa, F., 267, 424
Jeanselme, E., 161, 392, 396, 398, 400 Koumaris, I. G., 55, 371
Jedlicka, L., 401 Krause, A., 182, 403
Jerphagon, L., 384 Krische, A. B., 233, 416
Jimenez, C. V., 380 Krogman, W. M., 78, 378
450 INDEX
kuamon apekhesthai)211 Hippocratic writings, 157; not lep_rosyi?

kuamos) 212-13 Greek, 163; in Theophrastus, a mmor ail-
kuamos Aiguptios) 212 ment, 166; treated in sulfur springs, 167;
kutimosHellenikos)212 treatment of in Hippocratic writings, 165
Kudlien, F., 183, 365, 366, 389, 403 Leprosaria (from sixth century A.D.), 171
Kuehlewein, H., 285, 427 Leprosy, 152-76; absence of bon~ evidence
Kuhn, T., 241 in Palestine, 162; absence of evidence for
kundngkhai) 307 in New World, 155; absence in Greece be-
kundngkhe) 337 fore first century, 168; in ancient China,
Kurth, G., 400 158; in ancient Egypt, 154, 159; in ancient
Kussmaul's dyspnea, 287 India, 157; in animals, 2m; artistic evidence
Ved. kustha) true leprosy, 157 of, 397 n. 12; artistic representations of in
Kyphosis, 83 Greece, 155; BCH vaccination against, 408
n. 17; in the Bible, 160; in Bronze Age Ly-
Labarbe, J., 374 cia, 155; brought by Alexander, 175;
Labat, R., 394 brought by Roman legions, 176; complex
Laboulbene, A., 434 relation to poverty, 2m; endemic in Mid-
Lacroix, L., 381 dle Ages, 171; endemic in third world, 173;
Laderman, C., 427 evidence in Confucius's Conversations)158;
Laennec, R. T. H., 405 first medical description of in Greek, 169;
Lagier, R., 378 four clinical forms of, 152-53; idiosyncracies
Laignel-Lavastine, M., 397 of among mycobacteria, 200; immunol-
Lain Entralgo, P., 367, 429 ogy of, 202; limited contagiousness of,
Lallo, J., 421 202; in medieval Europe, 172; in medieval
Lana ta, G., 367 France, 154; in Mesopotamia, 159; a new
Landau, 0., 361 arrival in Greece, 169; a new arrival in
Landre-Beauvais, A., 83 Italy, 169; origins of, 174-76; osteoarcheo-
Landry, A., 383 logical evidence before 100, 154; osteoar-
Langholf, V., 432 cheological signs of, 153-54; in Persia, 164;
Lanzkowsky, P., 421, 425 a pre-Hebraic disease, 162; relation to tu-
Larguier de Bancels, J., 416 berculosis, 153; social consequences of, 161;
Larizza, P., 414 social prerequisites of endemic, 175; social
Lasserre, F., 42, 43 repression by Pippin and Charlemagne,
Lathyrism, in Hippocrates, 223 172; sporadic in ancient Greece, 155, 173;
Laver, W. G., 435 sporadic in ancient Italy, 173; spread in an-
Lead detection, in ancient bones, 51 tiquity, 175; terminology of in Greek and
Lebeau, J., 297 Latin, 168-69; transmission of, 152; trans-
Lebeaupin, A., 410, 416 mitted from rodents in Neolithic, 207
Leca, A. P., 402 Leprosy and tuberculosis compared, 198-
Lederer, R., 229, 414 209; in animals, 2m; common bacillus,
Lee, P., 249, 419 198; contagiousness, 202, 205; history of
Lee, R. B., 381 microbes, 209; immunological competi-
Lee, T., 402 tion between, 203; Mantoux reaction,
Leeper, G. W., 383 203; Mitsuda reaction, 203
Lefebvre, G., 363 Lesky, A., 433
Lehmann, H., 264, 265, 419, 420, 423 Lesky, E., 385
Lehmann, K. B., 198 leuke) 143, 165-67
Leibowitz, J. 0., 369 Leukorrhea, in Hippocratic writings, 146
leikhin) 164 Leveque, P., 380
Leininger, J.R., 408 Levi-Strauss, C., 238
Lemaire, V., 379 Levy, I. , 410
Lenormant, F., 234, 416 Lieber, E., 417
Leon, E. F., 406 Lien-teh, W., 398
Lepers: Jews slandered as, 163; social Liener, I. E., 413
rejection of, 161 Lin, J. Y., 414
Lepine, P., 327, 434 Lincoln, B., 366
lepra:in Hippocratic writings, 165, 166; legal Linear B, pathological terms in, 19
term for leprosy in A.D. 549; meaning in Ling, K. M., 414
INDEX 451
Lipschutz, H., 365 Malgaigne, J. F., 361

Little' s disease, 337 Mallegni, F., 270, 421, 424
Littre, E., 285, 291, 302, 315, 319-24, 331, 341- Malnutrition of children, rr2
42, 360, 427, 431, 434, 436 Manetho, 162
Livadas, G. A., 426 Mani, N., 434
Livingstone, F. B., 423 Mantoux reaction, 203
Lodge, T., 424 Maratchev, A. G., 422
loinws)362 Marchiafava, E., 299, 430
Longevity, ro7-8 Marcovich, M., 368
Lorenz, R., 362 Marcsik, A., 267, 424
Loret, V., 4rr Marganne, M. H., 364
Love, as pathological, 43-44 Marguth, F., 358
Lowe, J., 408 Marinaras, S., 376
Lozo ff, B., 424 marzsca)150
Lucot-Branlard, R., 420, 421 Marquet, M. R., 413
Lucretius, r68 Martin, R., 381
Ludwig's angina, 130 Martini, E., 382
Lugassy, G., 415 Martiny, M., 427
Lumpy jaw, 86 Maspero, G., 384
Lupus polyarthritis, 83 Massa, E., 413
Lurie, M., 409 Masser, C., 380
lusa) 34 Masson, 0., 45
Luzzatto, L., 232 Mastoiditis, 121, 130
Lymphogranuloma venereum (LGV), 149 Masure!, N., 435
Lynch, K. A., 380 Matsaniotis, N., 421
Lysimachus, 162 Mattes, J., 366
Mattingly, P. F., 277, 425
MacArthur, W., 291, 429 May, J.M. F., 412
McCardle, A., 383 Mayer, F. J. K., So
McCartney, J. J., 357 Mazars, G., 413
McCurdy, P. R., 414 Measles, 336
McDonald, M., 361 Medical geography, of Greece, 54
McDonald, W. A., 382 Mediterranean, eastern, as disease pool, 92
McEvedy, C., 95, 382 Medosades, 316
MacKie, A., 370 Meillier, C., 369
McKusick, V. A., 414 Meinecke, B., 183, 367, 403, 404
McNeill, W. H., 381 rrzile)344
McNulty, F., 408 Mellaart, J., 56, ror, 381, 425
Madden, J., 396 Mellink, M. J., 56, 425
Maegraith, B., 430 Menander, 68
Maingault, V. P.A., 433 Mendoza, D., 402
Maj no, G., 366, 387 Menetrier, P., 438
Major, R. H., 406, 413, 430 Meningitis, 122, 130, 131; epidemic, 123
Makler, P. T., 385 Meningoencephalitis, 131
Mal de pinto. SeePinta Menini, C., 266
Malamos, B., 421 Menke, C. T., 415
Malaria, 37, 39, 89; arrival of in Hellenistic Mensforth, R. P., 421
period, 280; distribution vs. thalassemia, Merke, F., 359
265; falciparum, and porotic hyperostosis, Messina, P., 225
278; frequency of in relation to porotic hy- Mestler, G. E., 360
perostosis, 275; hemolysis in, 297; hyper- Mesulam, M. M., 369
endemic of in Greece, 282; literary Metabolic disease, 74
evidence, 280; origins, 277; prehistory in Metazoa, 120, 208
Greece, 278; retreat in early Greece, 279; Meyer-Steineg, T., 376
revival in Hellenistic period, 281; social Meyerhof, M., 363, 429
change and, 282-83; varieties known to Micheli, F., 249
Hippocrates, 281 Michler, M., 358, 375, 406, 436
Malarial hemoglobinuria. See Blackwater Micrococcaceae, rr9-21
fever
452 INDEX
M icrococcusJn9 Nabothian follicles, 149

Microradiography, of ancient bones and Nardi, F., 358
teeth, 49 Nathan, H., 421
Microserology, of ancient blood, 50 nautfa) 35
Mildner, T., 364 Necrosis, 128
Milian, G., 392 Needham, J., 158, 398
Milne, J. S., 437 Nee.I, J. V., 419, 420
Miltiades, death of, 127 Nei-Ching) 158
Mina La Grua, A., 225, 413 N eisseriaceae, 122, 123
Mira Franco, M. P. de, 224 N eisseria gonorrhoeae, 144
Mitropoulos, K., 361 Nemeskeri, J., 376
Mitsuda reaction, 202 Neumann, C., 381
Moissides, M., 359 Neumann, R. 0., 198
Mollaret, H. H. , 392 Nicias, in old age, m8
Ms6ller-Christensen, V., 153, 154, 162, 371, Nicolle, C., 359
393, 396, 402, 408 Nicolucci, G., 54
Montalenti, G., 423 Night blindness, 75, 327, 338
Montano, G., 225-26, 413 Nikitas, A. A., 318, 331, 433
Moodie, R. L., 48, 120, 181, 247, 370, 388, notiasphth(sisJ 145
418 nukmlopes)308
Moore, S., 247, 418 Bab. nuqduJ 159, 164
Moreau, J., 383 Nyqvist, E., 387
Morelli, F., 120 Nystagmus, 349
Moreux, B., 361
Morse, D., 181, 4m, 402 Oberhummer, E., 317, 433
Morsier, G. de, 371 Oculo-urethro-synovial syndrome, 83
Moseley, J. E., 420 Odysseus, 33-34, 36
Moser, M., 359 Oedipus, 363
Motulsky, A. G., 265, 300, 414, 415, 423 Old age, m8
Maule, L., 362 Oldelehr, H., 367
Mueller, H., 247, 418 omikhefnJ 45
Mueri, W., 428 Ongaro, G., 437
Mugler, C., 361, 400 opisthot6nosJ351-52
Mule Bertolo, 413 Oppenheim, A. L., 398
Muller, K. E., 390 Oppert, J., 159
Mumps, 336 Opthalmia, 26, 73
Munck, V., 358 Oribasius, 168
Munter, L., 65, 374 Oriel, J. D., 144, 392, 394, 395
Munz, F. R., 385 6robos)222
Muratova, V. S., 4IO Orsini, A., 258, 414, 422
Mycobacteria, 198; in animals, 199; dado- Ortner, D. J., 370, 4m, 421
gram of, 208; evolution of, 206-9; evolu- 6spria)222
tion parallel to hosts', 199; free-living, Osteitis deformans, 70
commensals, and parasites, 199; paratuber- Osteoarchaeology, neglect of Greece in, 55
culous in humans, 199 Osteoarthritis, 77; improved living condi-
M ycobactmum aquaeJ198 tions and, 82; oflimbs and joints, 79; of
M ycobactmum bmnsJ198 spine, 80-82; of temporo-mandibular
Mycobactmum lepra.eJ152, 198; history of, joint, 78
209; specialization of, 200; specialization Osteomalacia, 75
of and newness, 207-8 Osteomyelitis, 121; on bones of prehistoric
MycobacteriumphleiJ 198 animals, 120; cranial, 122; of long bones,
M ycobactmum smegmatisJ198 128; of skull, 128
M ycobactmum terra.eJ198 Osteoperiosteitis, 121
Mycobactmum tuberculosis)177, 198; history Osteophytosis, 78; of the spine, 82-83
of, 209 Osteoporosis symmetrica, 246
Myllias, 217 Osteosarcoma 4 7, 72
Mylonas, G. E., 64, 384, 388 Otanes, 151
M yxedema, 148 Otitis media, suppurative, 130
M yxmnrus influenza.eJ333 Over, H.J., 370
INDEX 453
Overdetermination, in mythical thought, Peritoneal tuberculosis, 42

238-40 Peritonitis, 129, 148; tuberculous, 195
Overpopulation, and rise of epidemics, 90 Pernicious anemia, 148
Pertussis, 335
Pagel, W., 404 Perzigian, A. J., 409
Pager's disease, 70 Petrequin, J. E., 373, 374
Paleo-immunology, 50 Pezzia, A., 402
Paleo-odontology, n3 Pfaff, F. , 431
Paleodemography, 56, 87; and daily life in Phaedrus, 69
antiquity, 51 Phagedenic ulcer, 151
Paleopathology: of bones from Troy, 53; Phankim-Chapuis, M., 380
early history of, 47-48; of Egypt, 52; of phannakon) 33
Greece and Italy, 49; modern, 49; and os- Pharyngitis, membranous, 130
teoarchaeology, 52; and race, 53, 54, 55, Phenylketonuria, ro
246; and statistical analysis, 52 Pherekles, 32
Pales, L., 48, 247, 248, 370, 388, 4or, 418 Philemon, age of, ro9
Palladius, 308, 3n Philiscus' disease, 284-304
Panayotatou, A., 156, 326, 397, 434 Phineas, 363
Pandora's jar, 5 Phlebitis, 130
Panessa, G., 427 Phlegmon of the foot, 129
Pao Hsien, 158 Phoenician disease, true leprosy in Greece,
Papanicolaou, A., 439 167
PapyrusEbers) 159 phofdes)45
Paracentesis, thoracic, 131 Phokas, G., 325, 434
paraleluminon) 344-45 phrenftis) 359
parekrouse)286 phthfsis) 183
Parrot, J., 247 phth6e) 18
Partsch, J., 381 phurna) 42, 350-51; 'tubercle,' 188
Pa thocoenosis phuton) 434
-ancient, 4 Physicians, blindness of, 2ro, 241
-with broken equilibrium, 88 Pian. SeeYaws
-concept of, 2-3, 8 Piery, J., 4or
-dynamics of, 88 Piga, A. M., 413
-Greek: and Archaic Age, 91; in Bronze Pike, A. W., 370
Age, 90; and Dark Ages, 91; in equilib- Pindar, 46
rium, 91; and lifespan, 99; and malaria, Pinel, P., 437
281-83; out of equilibrium, 92; and over- Piniatoglou, L., 416
population, 90; pre-Classical, 91; and tran- Pinta, 134; as prototypical treponematosis,
sition to Neolithic, 89 137
Pathocoenotic equilbrium, 338-39; and Pirquet, C. von, 227
mycobacteria, 204 Pisa, W., 321
Pathocoenotic symbiosis, 205 Pittacus, 44
Paul, J. R., 327, 337, 434 Plagiocephaly, 66
Pauling, L., 420 Plague of Athens, 92
Paul of Aegina, 172 Plasnwdiumfalciparum) 277
Pausanias, 26 Plasnwdium rnalariae)277
Pays, J.-F., 261,422 Plasnwdium ovate)277
Pazzini, A., 144, 357, 394 Plasnwdium vivax) 277
Pea intoxication, 223 Plato, 73, 94, ro8
Peiper, A., 395 Pleurisy, 187; serofibrinous, 131
Pellagra, 75 . pleuritis) 131
Pelvispondylitis ossificans. SeeAnkylosmg Pliny the Elder, 169, 193, 22u
spondylitis Pliny the Younger, 194
pepasnws)124 Plutarch, on leprosy and new diseases, 170
pepsis)124 Pneunwcoccus)n9
Perdrizet, P., 406 Pneumonia, lobar, 121, 131
Pericles: head of, 67; in old age, ro8 Poirier, J., 408
Perinthus, 314 Poliomyelitis, 86, 336-37
peripleumonfe)131
454 INDEX
Polyneuritis, acute idiopathic, 324 Puschmann, T., 3n, 325, 387, 431, 434
Pomtow, H., 406 Bab. pusu) 159, 164
Pontani, F. M., 361 Putschar, W. G. J., 370, 4or, 418
Porotic hyperostosis: causes of, 248; de- Pyelonephritis, 145
creases in Bronze Age, 90; differentiating Pyogenic bacteria, prehistory of, 120, 121
etiologies, 253; distribution of, 267-68; dis- Pyorrhea, alveolar, 121, 122
tribution of in relation to malaria, 275; fre- Pythagoras: death of, 216; as healer, 220; the
quency in Greek wetlands, 272; in Greece, rooster and, 218
247; hereditary anemia and, 248; lesions Pythagoreanism: broad beans in (seeBroad
of, 245-46; multiple etiologies of, 252-53; beans); food tabus of, 213-14; magical pan-
nonanemic in America, 268; nutritional theism of, 214
deficiency and, 267; osteoarchaeological
distribution, 267-68; osteoarchaeology in Rabello, F. E., 391
Italy, 268-70; paleopathology, 266-68; sta- rakhis) 343
tistical analysis of Greek cases, 273, 276; Ramazzini, B., 223
thalassemia and, 251; thalassemia as cause Ramsdale, C. D., 278, 426
in Greece, 271; worldwide distribution, Rathmann, W., 4n
246
Raven, H. C., 392
Pott's disease, 178, 195 Reader, R., 396
Potter, P., 336, 391, 436 Reddy, D. V. S., 402
Pouilloux, J., 289, 428 Reen, R. Van, 385
Poynter, F. N. L., 358 Regnault, F., 156, 397, 427
Poynton, H. G., 421 Regoly-Merei, G., 76, 377, 380
Preaux, C., 384 Reinach, S., 235, 416
Preiser, G., 358, 367, 369, 391 Reinmuth, 0. W., 384
Preuss, J., r6r, 398 Reiter's syndrome (Fiessinger-Leroy-Reiter
Prien, E. L., 385 syndrome or RS), 83, 85
Prim, E., 364 Renfrew, C., 95, 382
Primiero, M., 385 Resistance to infection, rn
Pringle, J., 291 Resnick, D., 379
Pritchet, C. D., 431 Retrospective diagnosis, 6-7; from artistic
Probe, ancient medical tool, 344 representations, 155; evolutionary change
Proclus, 26 and, ro; frailty of, 7; immunological fac-
Proksch, J. K., 391 tors in, 13; literary evidence for, 17; pre-
Prolapsed uterus, 43 suppositions of, 8-9
prophasis)3ro, 347-48 Rhazes, 292
prorrhisis) 351 Rheumatic fever (Bouillaud's disease), 83,
Proteus)n9 130
Protospatharius, T., 297 Rheumatism. SeeOsteoarthritis
Prusias Monodus, 74 Rheumatoid arthritis: absence of in antiq-
Ps.-Aeschines, 168 uity, 83, 84; genetic factors in, 84; possible
ps~llos)344 ancestor of, 84
psora)163 ,, Richardson, B. E., ro5, 384
Psoriatic rheumatism, 83 Richer, P., 377
Pucci, P., 227 Richter, G. M.A., 406
Pulse, food in famine, 222 Rickets, 75-76; absence of in Greece, 76; in
puon)123 northern climates, 76; in Sicily, 76
puret6s) 37 Riddle, J.M., 379
Purulent inflammation: antiquity of, 121; Riese, W., 357
death by, 126, 129; evidence of, 121; Hip- Rietti, A., 85
pocratic knowledge of, 128; Hippocratic Rietti, F., 249
views on, 124; of internal organs, 130; Robert, F., 38, 317, 390, 405, 429, 431, 433,
nontraumatic, 131, 132 436, 437, 438
Pus: ancient concepts of, 124; formation of, Robert, L., 433
120; "good" and "bad," 126; Hippocratic Robinson, D. M., 56, 383
terminology for, 124; Inda-European ter- Robinson, H. S., 69
minology of, 123 Rodden, R. J., 56, 380
pus) 123 Rohde, E., 416
Rokhlin, D. G., 267, 388, 423
INDEX 455
Rolleston, J. D., 357 Schilling, E. D., 413

Roney, J. G., 402 Schistosomiasis, in Egypt, 86
Roselli, A., 373 Schliemann, H., 52-53, 371
Rosen, G.~ 434 Schmeideberg, 0., 366
Rosenbaum, J., 143, 391 Schmidt-Ries, H., 381
Roshem, J., 40I Schmorl's nodules, 83
Rosner, E., 360 Schneider, L. A., 68, 374
Ross, R., 279-80, 426 Schnurrer, F., 434
Rotberg, A., 408 Schonbauer, L., 365
Rotes-Querol, J., 378 Schopf, J. W., 120, 388
Roth, K. L., 228, 414 Schroader, L. von, 411
Rothari, King of the Lombards, 171 Schuchardt, B., 413
Rothschild, N., 359 Schultz, D., 403
Rubens, P. P., 84 Schultz, W., 364
Rucknagel, D., 419, 420 Schultze, F. , 381
Ruffer, M. A., 47, 48, 85, 121, 179, 369, 380, Schumacher, J., 238, 416
388, 402 Schumann, H.-J. von, 364, 396
Ruffie, J., 264, 415, 421, 423 Schwartz, J., 417
Rufus of Ephesus, 168 Schwengler, K., 399
Ruiz Moreno, A., 379 Scott, H., 303, 430
Russell, J. C., 384 Scrofula, 178; in Hippocratic texts, 196
Russell, P. F., 425 Scurvy, 75, 77, 377 n. 117; absence of in
Ryan, D. J., 424 Greece, 77
Seasickness, 35
Sacko, B., 413 See, G., 405
Sacred disease, 40 Segni, G., 413, 414
Sacrifice, human, 63 Semelaigne, A., 366
Sager, P., 402 Senyiirek, M. S., ·56, m3, 384
saharsubbu) 159 sepsis)124
Sahlins, M., 381 Serjeant, G. R., 420
Saimot, G., 430 Seze, A. de, 378, 380
Saint-Perier, J., 246, 418 Sharpe, W. D., 404
Sakellarakis, Y., 63, 374 Short, C. L., 84, 379
Salmon, P., 97, 382 Shufeldt, R. W., 369
Salmonellatyphi) 348 Shute, P. G., 278, 426
Salmonelloses, 38 Sickle cell anemia, 248, 251; abnormal hemo-
Salmonellosis, 89, 348 globins in, 251; bone lesions of, 252; d~stri-
Sandison, A. T., 370, 396, 438 bution of, 264; distribution vs. malana,
Sansone, G., 413, 414 265; first mutation locus, 264; in Hippoc-
Sapouna-Sakellaraki, E., 63, 374 rates, 274-75; homo- vs. heterozygous, 252
Sappho, 43 Siegel, R. E., 303-4, 388, 391, 395, 403, 429,
Sarcoma of eye-socket, 72
43 1
Sarton, G., 416 Sigerist, H. E., 248, 361, 370, 382, 388, 393,
Sartori, E., 414 398, 416, 419, 427
Sauvy, A., 384 Silvestroni, E., 256, 259, 422, 423
Savicki, A., 370 Simon, B., 361
scabies)163 Simon, I., 398
Scamuzzi, U., 406 Simpson, R. H., 382
Scarlet fever, 337 Sinanthro-puslantianiensis)121
Schachter, M., 359 Singer, R., 423
Schadewaldt, H., 357, 359, 366, 4m Siniscalco, M., 266, 415
Schalimtzek, M., 402 Sinusitis, 122
Schapira, G., 414, 420 Skinsnes, 0. K., 396
Schar-Send, M., 397 Skoog, T., 375
Schaudinn, F., 134 Skull deformity, consequences of, 67
Scherlievo, 134, 139, 142 Smallpox, 86; absence of, 89
Scheuermann 's disease, 83 Smith, E. G., 73, 84, 85
Schiller, W., 367 Smith, G. E., 179, 376, 396, 402
456 INDEX
Smith, W. D., 362 Suppuration. See Purulent inflammation

Snell, B., 361 Sussmann, M., 390
Snodgrass, A. M., 381 Svejgaard, A., 370
Sokoloff, L., 380 Swinton, W. E., 370
Sokolowski, F., 416 Sydenham, T., 84
Solente, G., 394 Syncope, 21
Sophocles: age of, ro9; skull of, 54, 65 Syphilis, 133-51; absence of osteoarchaeologi-
Sorre, M., 296, 382, 430 cal evidence of in Europe, 133; in animals,
Soulangas, M., 387 138; in bone traces, 130-40; endemic, 134,
Souques, A., 324, 391, 434 139, 142; imported by Columbus, 134; ori-
spasmos)390 gin of, 134; presence of bone traces in
spasthenai)390 New World, 140; "venereal," 134
Spastic hemiplegia, 68 Syphilitic gummas, 157
Spastic paralysis, 337 Systemic disease, 74
Spencer, D. G., 380 Szeinberg. A., 414
Spermatorrhea, 145
Sperry, R. W., 67, 374 tabes)192
sphtikelos)sphakelismos)128, 352 Tacitus, and the Exodus, 163
Sphangos, J. C., 426 Taillardat, J., 400
Spina bifida, 70 Taliaferro, W. H., 420
Spine, biological history of, 81 Talipes valgus, 71
Spirochaetapallida. See Treponemapallidum Teet~, Athenian vs. Spartan, u5
Spiropoulos, N., 421 tekedon)37
Spitery, E., 393 Temkin, 0., 292, 368, 387, 428, 438
Spoehr, 0., 393 teredfm)390 n. 53
Spondylitis, 80, 81 Terminology of disease: Inda-European, 18;
Spondylolisthesis, 82 problems in reconstructing, 6-7
Sprengel, K., 233, 416 titanos) 342
Springer, B., 392 Tetanus, 351; chronic, 342
Stamatoyannopoulos, G., 414, 415, 423 Thalassemia: artistic evidence of in Greece
Staphylococcus) u9 275; bone lesions of, 252; current distribu-'
Staphylococcus aureus) 128 tion, 254-57; discovered, 249; distribution
Stastny, P., 370, 379 vs. malaria, 265; and Greek colonies in
Steinbock, R. T., 370, 4or Italy, 256; Greek origin theory, 259-60;
hemoglobin metabolism in, 250; heterozy-
Stephanos, C., 54, 298, 302, 371, 399, 4or,
408, 430
gous, 250; homozygous, 250; Mongolian
Stephens, J. W.W., 430 origin theory, 260-63; multiple origins of,
Stephenson, F. R., 432 262-64; mutation witnessed, 263; named,
249; origin and spread, 257-64; osteoar-
Stewart, T. D., 393, 409
Sticker, G., 291, 303, 388, 427, 428 chaeology of in Greece, 270-74; and po-
Stillwell, A. N., 375 rotic hyperostosis, 251; radiology of, 250;
Stine, W. A., 385 same distribution as favism, 263
Stinking bean-trefoil, hemolytic anemias Thamyris, 362
and, 228 Theognis of Megara, 44, 45
Strabo, W alafrid, 171 Thersites, 25, 71
Straton of Alexandria, 168 Thillaud, P. L., 38b
Streifler, M., 413 Thompson, E. S., 435
Streptococcus)u9 Thompson, H. A., 56
Streptococcuspyogenes)129 Thompson, R. C., 402
Stromberg, R., 360, 428 Thrombosis of meningeal sinus, 122
Strong, F. M., 413 Thrush, 148
Striimpell-Marie's disease. SeeAnkylosing Thucydides, ro7, ro8
spondylitis thumia) 150
Stubbings, F. H., 381 thumos) 22
Studniczka, F., 374 Tierney, M., 416
Sturrock, R. G., 380 Timken-Zinkann, R. F., 431
Succussion, 192 Timycha, 217
Sudhoff, K., 140, 398 Toldt, C., 248, 417
Superstition, value of, 236 Tommaselli, S., 298
INDEX 457
Tonsillitis, streptococcal, 83 185-88; terminology of in Greek, 183-85;
Tonz, 0., 263, 423 third world disease, 205; tracheobronchial,
Tooth decay. See Dental caries 187~ urogenital, 150
T oscanelli, N., 426 Tuberculosis and leprosy compared. See
Toti, A., 269, 387, 424 Leprosy and tuberculosis compared
Townsend, J. F. , 435 Tuberculous osteoarthritis, 178
Trachoma, 26 Tuberculous spondylodiscitis. See Pott's
trdkhews) 343 disease
trauMs)344 tuphos) 437 n. 21
trauma) 33 Turchetti, A., 413
Trepanation, cranial, 63-65; magical origins Turner, T. B., 392
of, 64; technique of, 64 Typhoid fever, 89, 348
Treponarid, 134, 139, 142 Typhus, exanthematic, 304, 348
Treponemacarateum) 135, 137, 139
Treponemacuniculi) 139 Ucko, P. J., 382, 385
Treponemapallidum) 134, 136, 139, 141 Ulcer, 45
Treponemapertenue) 134 Ulcers, puerperal, 150
Treponematoses Urethritis, nonspecific, 149
-absence of in Greece 142 Urinary lithiasis, 145
-and osteoarchaeology, 139 Urinary stones, 3, 45
-diversity: biochemical, 137; biological and Uta, 157
environmental, 141; environmental, 135 Vachon, F., 430, 431
-historical hypotheses of, 135 Valentine, W. N., 419
-one or many, 135 Valeri, C. M., 419
-origin and cladistic rules, 141 Valerius Maxim us, 44
Treponemazuelzerae) 138 Vallaoras, V. G., 56, 384
Trichonwnavagina/is)148 Vallois, H.-V., mo, 259, 383, 387
Trichomoniasis, 148 Van der Waerden, B. L., 4m
Tringham, R., 381 Van Gerven, D., 247, 418
Trismus, 341 Vassal, P., 397
Tsouros, A. D., 56, 374 Vegetti, M., 288, 428
Tuberculosis, 89, 177-97; absence of Near Veith, I., 435
Eastern evidence of, 180; absence on pre- Venereal disease: in Bible, 142; in Greece,
historic animals, 179; accounts of death 144-51; and morality in nineteenth cen-
by, 193; artistic evidence of in early Egypt, tury, 143
180; Celsus on, 192; as a collective disease Veras, S., 238, 416-17
in Hippocratic texts, 190; common after Vermeule, E., 373, 381
Hippocrates, 193; complex symptomatol- Verrettas, S., 298
ogy of, 177; contagiousness known to Ro- Vertue, H. St. H., 394
mans, 194; "cures" for, 193; derived from Vidal-Naquet, P., 381
bovine tuberculosis in Neolithic, 200; epi- Villard, F. , 381
demiology of, 204; ethnic variation in re- Vintr6, E., 429
sistance, 206; evidence of in ancient Virchow, R., 48, 53, 54, 66, 157, 371, 374,
China, 181; evidence of in Hippocratic 417-18
texts, 183-92; extrapulmonary, 194-97; fin- Viruses, origins of, 383 n. 39
brocaseous, 178; and habitus phthisicus, Vitamin deficiencies. SeeDeficiencies
191; Hippocratic etiology of, 188; immuno- wtiligo) 163
logical competition with leprosy, 203; in- Vogel, C. W. de, 4m
adequate evidence for frequency statistics, Vore, I. de, 381
182, 197; literary evidence of in ancient In-
Vulvo-vaginal diphtheria, 149
dia, 180; literary evidence of in Greece, Vust-Mussard, M., 292, 429
182; on Neolithic spine, 179; osseous, Ne-
olithic evidence of in Europe and Egypt, Wace, A. J. B., 373, 381, 388
179; osteoarcheological evidence of, in Ward, R. H., 382
Greece, 182; osteoarcheological evidence Warren, C. P. W., 361, 381
of, on pre-Columbian bones, 181; osteoar- Weatherall, D. J., 419
ticular, 195; osteoarticular, in children, Webster, R. G., 435
196; pulmonary, Hippocratic accounts of, Weiss, D. L, 408
INDEX
Weiss, K. M., 382 Wound, ulcer as, in Hippocrates, 125

Weiss, P., 419 Wright, D. J., 393
Welcker, H., 246, 417 Wright, J., 392
Wellman, M., 399 Wunderlich, C. A., 291, 302, 429
Wells, C., 49, 73, 267, 370, 376, 384, 385,388,
396, 402, 424 Xenophilos, age of, ro9
Wenkebach, E., 431 Xenophon,46
Wheat, production and consumption, n3 Xerophthalmia, 75, 3n
Whipple, G. H., 249, 419 Yamamoto, A. S., 432
Whitlow, 129 Yaws, 134; as prototypical treponematosis,
Whooping cough, 335
136
Widmer, L., 409 Yellow fever, absence of, 89
Wiesenfeld, S. L., 423
Yoeli, M., 397
Wilamowitz-Moellendorf, U., 367
Yoshida, A., 414
Wilke, P. J., 370
Wilkie, G., 370 Sic. zafara) 255
Williams, C. B., 358 Zaino, E., 258, 422
Williams, H. U., 179, 246, 248, 393, 418 Zambacho Pacha, D., 173,4or
Willigan, J. D., 380 Zannos-Marioulea, L., 415, 422
Wilson, J. V. K., 159, 398 Heb. Zara cat) 160-64
Windle-Taylor, P. C., 375 Zenner, A. A., 362
Wine, effects of, 44 Zimmerman, M. R., 402
Winkle, S., 438 Zinkham, W. H., 414
Wittern, R., 290,. 358,389, 391, 403, 429 Zorab, P. A., 380
Wolf, J. H., 358 Zuelzer, W.W., 419
Wong, K. C., 398 Zulueta, J. de, 278, 424, 425, 426
Woo, Ju-kang, 388

Mirko D. Grmek - Diseases in The Ancient Greek World-Johns Hopkins University Press (1991)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mirko D. Grmek - Diseases in The Ancient Greek World-Johns Hopkins University Press (1991)

Uploaded by

Copyright:

Available Formats

MirkoD.

Translated byMireille Muellner

THE JOHNS HOPI{INS UNIVERSITY PRESS

Baltimore and London

Originally published as Les Maladies a l)aube de la civilisationoccidentale.

English translation © 1989The Johns Hopkins University Press

The paper used in this publication meets the minimum requirements of

LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA

Grmek, Mirko Drazen.

Introduction: THE CONCEPTUALIZATION OF PATHOLOGICAL EVENTS I

The Conceptof Pathocoenosis 2

Chapter One: LITERARY REFLECTIONS OF PATHOLOGICAL REALITY 17

Chapter Two: PALEOPATHOLOGY: Evidencefrom Ancient Bones

Chapter Three: PALEO DEMOGRAPHY: Evidencefrom Ancient Bones

Chapter Iour: COMMON PURULENT INFLAMMATIONS 119

Chapter Five: THE ORIGIN AND SPREAD OF SYPHILIS 133

Chapter Six: LEPROSY: The Gradual Spreadof an Endemic Disease 152

Chapter Seven:TUBERCULOSIS: A Great I(iller 177

Extrapulmonary Tuberculoses 194

Chapter Eight: LEPROSY AND TUBERCULOSIS: Their Biological

Chapter Nine: THE HARM IN BROAD BEANS: Legendand Reali-ty 210

Chapter Ten: POROTIC HYPEROSTOSIS, HEREDITARY ANEMIAS,

Chapter Eleven: THE HIPPOCRATIC CONCEPTION OF DISEASE:

Chapter Twelve: THE CONSTITUTION OF A WINTER IN THRACE:

Chapter Thirteen: A DIALOGUE BETWEEN A PHILOLOGIST

This book is a history of diseases, not a history of medicine. It is not,

It is impossible to apprehend correctly the significance of an ancient text

from malaria or pneumonia. Bouts of tertian or quartan fever succeed each

The Conceptof Pathocoenosis

Three propositions can serve to define this neologism: (1)Pathocoenosis

Semantic Constants and the

not the only imaginable pathogenic reality. It is not impossible that in

The Constants of Human Geqgraphy

I hesitate to supply figures for diseases caused by an inborn error of

chains, one genetic and the other external. Consider hypertension as an

did in the previous three millennia. 27 A modern medical inquiry, such as

The GreekPathocoenosisin the ClassicalPeriod

If we are not distracted by the essentially false etiological discussions of the

diseases. The most awful endemic diseases were, in modern terminology,

The HistoricalImpact of Disease

The Hippocratic corpus is a collection of about sixty fairly heterogeneous

Language as Historical Evidenceand the Linear B Tablets

to the administration of the palaces; unfortunately, none concerns medi-

Life and Death in the Homeric World

some aspects of daily life in Greece, it is incumbent on me to make plain

That is the law of mortal men, when they die:

The psukh~ of a man, declares Achilles in his passionate speech to Odysseus,

Health and Sickness in Heroic Times

Thersites the Insolent is turricephalous, almost bald, stooped, bow-legged,

Trojan War TV<Junds

TABLE I. Statistical Tableof vVtJundsin the Iliad

is only applicable to the poetic discourse by analogy; consequently, I refuse

gasped as an ox does when among the mountains the herdsmen

True Diseasesin Homer

The poet gives the epithet stuger6s'hateful, abominable, odious' to the

idemiology of infections due to salmonellae is cautionary in this regard:

From Hesiod to the R-esocratics

The clinical descriptions found in this treatise and in other Hippocratic

forms of epilepsy, sonvulsions caused by encephalitis, severe spasmophilia,

It would be pointless to hold forth on the possible causes of Heraclitus's

case of a 60-year-old leading the life of an anchorite, a very serious ailment

Lyric Poets of the Seventh and Sixth Centuries

matic state perceived to be abnormal. Medical literature and countless