Mixed Messages

Cultural and Genetic Inheritance in the Constitution of Human Society

The University of Chicago Press

Contents

A Brief History and Outline of Dual Inheritance Theory

In this chapter I offer a survey of the most important ideas that have led to the emergence of the contemporary theory of dual inheritance. Among the authors to be considered are Luca Cavalli-Sforza and Marcus Feldman, Richard Dawkins, Donald T. Campbell, E. O. Wilson, David Sloan Wilson, William Durham, and Robert Boyd and Peter Richerson. I make no claim to completeness, but intend rather to touch on those key contributions that allow me to trace and explain the basic assumptions of the theory. As I hope to show, taken together, many of these assumptions are far enough away from the most dogmatic and reductionist versions of evolutionism (in relation to the study of culture), and compatible enough with ideas in socio-cultural anthropology, to make a rapprochement between the two streams of thought possible. It is my project in this book to contribute to that rapprochement.

It is helpful to state right at the outset that the phrase "dual inheritance" is in some ways a misleading one. On the one hand, there is the genetic code, but on the other side of the duality there are several other ways in which instructions and information are transmitted among people and across successive generations. So, for example, Evelyn Fox Keller (2010) writes at the conclusion of her critique of the binarism of "nature and nurture": "Let us acknowledge that ... almost all human traits are transmitted from one generation to the next; but at the same time, let us also accept the fact that the mechanisms of transmission are very varied. They may be genetic, epigenetic, cultural, or even linguistic" (80). Or to take another example, Eva Jablonka and Marion Lamb (2005) encapsulate the whole argument of their valuable book in its title and subtitle: Evolution in Four Dimensions: Genetic, Epigenetic, Behavioral, and Symbolic Variation in the History of Life. Here genetic evolution is contrasted with not one but three other modes of information transmission across generations. In short, the "duality" in the dual inheritance model is based on the binary opposition between the genetic program and everything else that is not genetic. This latter category includes a congeries of phenomena which may loosely be categorized as "culture," in that they are acquired by a new organism from its predecessors by some form of "learning" or "experience" undergone by the phenotype itself rather than given by means of sexually transmitted genetic information.

The emerging field of epigenesis studies how at the molecular level the genetic code itself can be manipulated by the organism, and indeed how some "acquired characteristics" may be transmitted to a future generation in apparent defiance of evolution's "central dogma" that rules out such so- called "Lamarckian" inheritance. This phenomenon is extremely interesting and has potentially important implications for the understanding of genetics and of the transmission of information across generations more generally; however, it is beyond the scope of my present project, as well as of my competence, and so I will not consider it further here.

Behavioral learning, or social learning, whereby information is passed from a phenotypic "teacher" to a "learner," is found among many species as well as in humans, and involves the imitation of observed or shown behavior. In humans, this form of information transmission can be quite intentional; to what extent that is so among animals such as some primates—one thinks here of chimpanzee tool use, for example—is open to debate. Thus if a child picks up its parents' accent, gait, or bodily comportment, or learns from an elder how to string a bow or make a blow-pipe by watching and imitating, these are examples of social learning in humans that need not involve much or any explicit encoding in symbolic form.

It is only among humans that there also exists an extrinsic system of encoded information comparable in complexity and in its mode of operation to the genetic code. That is the cultural code, which in my view should be understood to include language (in contrast to Keller who, in the passage quoted above, seems to view language as if it were something other than culture). Language, like DNA, depends on the creation of significant differences in a material vehicle—molecules forming proteins in the one case, sound waves shaped by the human larynx in the other—in the form of primary binary oppositions, on the basis of which complex systems of representation can be built. Language is not, however, the only form of symbolic information by any means, and human culture is rich in all sorts of vehicles for the conveyance of meaning from one person to others in what Susanne Langer (1957) called "significant form." Langer's concept includes not only discursive forms, such as language, but also presentational symbols, such as are found in art styles, ritual objects, spatial arrangements, and the like. Human "cultures" as actual existing entities are best understood as socially constructed symbolic worlds within which (reasonably) coherent human lives are led by individuals in social groups acting in meaningful patterned relationship with each other. What both social learning and symbolic systems have in common, in opposition to genetic information, is that they are both transmitted via sensory perception in a social arena.

There is also in the determination of human action the external factor referred to as "the environment": many aspects of the wider non-human world impinge on the organism and affect the organism's form and behavior. While the environment does not usually "intend" to teach, since non-human nature does not (so we Westerners assume) have the capacity to act on such "intentions," nonetheless environmental factors instruct learners in the sense that the latter draw lessons from their experiences with the environment.

Of these nongenetic factors in the inheritance of information and instructions by humans from one another, the one on which I will concentrate is the symbolic dimension, both because it so closely parallels the genetic channel of information transmission (albeit with highly significant differences that I will specify more fully later) and because it is so salient in informing actual human existence, in individuals and collectively in society. Since the form in which much extrinsic information flows between and among people, including from parent to child, is along the symbolic channel, I will for convenience often refer to the symbolic code as if it were the same as "culture," but it will be understood that other nonsymbolic factors of the sort I have just described are in play as well. What all these forms of cultural information share, however, is that they are made available and transmitted in the perceptual sensorium of phenotypic organisms, either fully formed or in the process of development, rather than via the gametes. It is this feature that plays a key role in my theoretical formulations, so that the question of whether a particular piece of behavior is acquired by the transmission of symbolic form or by extra-symbolic, imitative social learning, while interesting, is not of crucial importance in this context. I will expand on how I understand cultural symbolism in chapter 3.

Cavalli-Sforza and Feldman

The origins of dual inheritance theory can be charted beginning in the 1970s with the work of Luca Cavalli-Sforza and Marcus Feldman. While various earlier authors had seen a parallel of some sort between genetic and cultural evolution in an often vague or intuitive sense, it was the development of sophisticated mathematical operations that could be used to model processes in population genetics and their application to cultural ideas by Cavalli-Sforza and Feldman (1973) that paved the way for the future development of a "dual inheritance theory" that was scientifically sound enough to provide the basis for a substantial research paradigm along well-established Darwinian lines. Evolutionary theory had made a great stride forward when it was established as a principle, in the years during which the neo- Darwinian synthesis was being forged, that what Darwin had called "species" (as in "The Origin" thereof) can usefully be thought of as collections or "populations" of individual organisms capable of interbreeding and thus transmitting their genetic material to offspring. Evolution in this view proceeds by the differential rates of reproduction and mortality among these individuals, affecting the changing form of the gene pool of the population over time. These differential rates, in turn, are a result of the fact that individuals within a population exhibit variations, and that there is inherent competition when, as is typically the case, there are resource limitations. Some traits allow the phenotypes exhibiting them to be more successful in reproducing offspring in future generations than others, and this process leads to evolution and, in some cases under particular conditions, speciation.

Traits, including behavioral traits, that are more or less adaptive in the sense of enhancing reproductive "fitness," are in turn the expression in the organic form and behavior of the individual phenotypic organism of instructions encoded in the genome about how to develop ontogenetically. For the most part, genetic instructions are stable, but occasional random or blind changes, or "mutations," in the genes introduce a new trait, which, when expressed in the phenotype, is usually deleterious but in a few instances proves to be advantageous to the individuals instructed by them. This advantage, understood in the specific sense of enhancing reproductive success, would then spread in the population since the phenotype exhibiting the new adaptive trait (and the genetic material it passes on to its offspring) provides an advantage over other phenotypes in the population. One can construct sophisticated and complex mathematical models of how percentages of genes might wax or wane in a population and thus make predictions that can be tested against empirical observation; this is the basis of the science of population genetics.

Cavalli-Sforza and Feldman realized that the same method could be used to study the ebb and flow over time of "ideas" or "beliefs" within a population of humans if these ideas or beliefs were treated as heritable "traits" that could be passed from one generation to the next; and they were able to model the way new inventions or ideas might spread in a given population using the concepts already in use in population genetics. At the same time, they were well aware of the limitations of the analogy of socially transmitted ideas to genes in a gene pool and noted some obvious differences. One of these is that the reproduction of ideas or innovations in a human population does not entail differences in shifting patterns of the birth, survival, reproduction, and death of numbers of individual organisms with or without a certain trait. Such "cultural" information can be transmitted directly in much less than the lifetime of a human generation, which is the key unit required for the blind "trial and error" method of natural selection. The information is acquired directly, sometimes in a matter of minutes or seconds, and its spread has nothing necessarily to do with the genetic reproductive rates of the individuals who have learned the new ideas (though of course it may). In this sense, as others before Cavalli-Sforza and Feldman had already noted, cultural evolution is "Lamarckian" rather than strictly "Darwinian," in the sense that innovations acquired during the lifetime of the phenotype can be inherited by future generations. In the sexual transmission of DNA, by contrast, it is assumed that the DNA in the gametes changes only by random mutation; it does not alter itself in response to lived experience that the phenotype bearing the DNA has undergone.

Cultural inheritance is likewise not Darwinian in that it could be said to involve "artificial" rather than "natural" selection. The latter—Darwin's great contribution to science—operates blindly and automatically as phenotypes carrying genetic instructions or "alleles" for various traits are simply observed to die out or spread in a population without any intent to do so. Indeed, the value of Darwin's formulation is that it eliminates the need for anything resembling foresight, intention, will, purpose, choice, value, or any related phenomena from every stage of evolutionary process, from "random" mutation to the shifts in the gene pool of a population that result from the accumulation of more adaptive traits as new environmental pressures shape differential survival and mortality within the population. This elimination of "anthropomorphic" factors enabled evolutionary biology to become congruent with the natural sciences of physics and chemistry, which had long since dispensed with anything but physical matter whose properties and changes over time could be described in mathematical terms. The elimination of intention from evolutionary process did not, to be sure, negate the observable fact that individual organisms produced by natural selection can and do exhibit purposive behavior, but the process itself is understood to be teleologically blind.

But "artificial" selection, with a discussion of which Darwin opens his master work, does involve choice, specifically choice on the part of humans, such as pigeon fanciers who can, through selective breeding, produce all sorts of phenotypic variation in their pigeon sub-populations. If breeders want pigeons with fluffier feathers on the crest or tail, they can mate those among their flocks who already have something of those features over several generations. Darwinian evolution involves substituting involuntary environmental pressure for the human breeder who acts with a conscious goal in mind, thereby eliminating the factors of choice, intention, or foresight from the process.

But it would seem to follow, if we grant that humans are capable of making choices guided by preferences, intentions, purposes, and so on when it comes to selecting the kind of pigeon they want to breed, that the same would go for any human invention. People are self-evidently capable of envisioning a desired outcome, figuring out what needs to be done to get there, and doing it; so that when a cultural "innovation," whether transmitted through symbolic means such as via language or imitated in pure "social learning," is adopted by individuals, it is with the hope, wish, or expectation that it will in fact provide some benefit to themselves. Thus, not only does cultural evolution appear to be "Lamarckian" rather than "Darwinian," it also seems capable of being the result of "artificial" (that is human, and hence intentional) selection rather than of "natural" selection, as Christopher Boehm (1978) argued many years ago. It must be added that if one understands human thought, including the forming of intentions, as something that can occur unconsciously, as contemporary cognitive science recognizes, then one need not suppose that all such artificial selection is made by conscious deliberation.

Finally, Cavalli-Sforza and Feldman's analogy of cultural to genetic evolution in populations implies that whereas genetic material can only be transmitted from biological parents to their own offspring, that is "vertically," cultural information can also be passed indirectly or horizontally just as well. Thus, for example, American children might absorb English from their biological parents, be taught Spanish by a genetically unrelated school teacher, and learn how to play games in the playground from other (unrelated) children. The effect of this would be perhaps the most dramatic of all in differentiating a cultural evolutionary process from the genetic one, because it would mean that cultural evolution could proceed to some degree independently of considerations of genetic fitness. If cultural instructions can be transmitted between or among individuals with no genetic relation to each other, but nonetheless are to the benefit of those individuals, then there is no inherent reason to suppose that the cultural traits selected for will necessarily confer any benefit to those individuals in terms of genetic as opposed to cultural fitness.