Naive sociobiology then gave way to evolutionary psychology, which avoids the danger of making predictions that are too specific and concerns itself with the evolution of underlying behavioral mechanisms of sexual attraction, fear of life-threatening circumstances, group cohesiveness, rationality, and so on. Such explanations, however, do not do the work that historians and sociologists require. For example, evolutionary psychology explains why babies emit piercing howls and wails when they are hungry or uncomfortable. They are helpless, and unless they can distract their parents from other concerns they will not be sure they will be fed or rescued from pain. Natural selection will then favor howling babies, since quiet ones may be malnourished or suffer injuries and so are less likely to survive.
Of course the screams of a baby can be counterproductive since parents have been known, in their frustration, to take drastic measures to quiet crying babies, even to the point of killing them. These are to be seen as pathological exceptions, however, when we take account of natural selection in favor of maternal love, since parents who injure their children will have fewer surviving offspring. While entirely plausible, such a theory does nothing to explain historical and social differences in child-rearing practices. As recently as the middle of the last century the administration of a swat on the buttocks or a rather energetic shaking was an entirely acceptable form of discipline for a recalcitrant child, but such behavior now is grounds for criminal charges of child abuse.
Evolutionary psychology also explains why all spoken languages must have certain phonemic properties in order that hearers can distinguish one word from another. The ability to distinguish similar spoken sounds is clearly of survival value. A confusion between “That animal always calls when cornered” and “That animal always kills when cornered” can lead to injury or death. What evolutionary psychology does not tell us, however, is why some people use clicks, some use rising and falling tones, why the kings of England finally came around to speaking English at home instead of French, or how the use of the periphrastic “do,” as in the replacement of phrases like “I go not” by “I do not go,” grew in the sixteenth and seventeenth centuries. Evolutionary psychology is not a theory applicable to historical change and cultural variation.
As a result, biological models of cultural change and diversity have been replaced by pseudobiological models, using the structure of Darwinian explanation metaphorically rather than literally. Darwinism is a population-based theory consisting of three claims. First, there is variation in some characteristics among individuals in a population. Second, that variation is heritable. That is, offspring tend to resemble their biological parents more than they do unrelated individuals. In modern Darwinism the mechanism of that inheritance is information about development that is contained in the genes that are passed from parent to offspring. Third, there are different survival and reproduction rates among individuals carrying different variants of a characteristic, depending on the environment inhabited by the carriers. That is the principle of natural selection. The consequence of differential reproduction of individuals with different inherited variants is that the population becomes richer over generations in some forms and poorer in others. The population evolves.
A classic case is the evolution of mimicry in butterflies. Some butterflies taste bad to their potential bird predators and the birds quickly learn from a few revolting trials to recognize them by their wing coloration and to avoid trying to eat them. Other species of butterflies that taste good have evolved wing patterns that make them look like the nasty-tasting species, and so are also avoided by their potential predators. This evolution was possible because butterfly wing patterns are genetically variable from individual to individual. In the past, an individual butterfly that tasted good and whose wings somewhat resembled those of the uneatable species would sometimes fool a bird and be spared from predation. The offspring of this survivor would on average resemble it. Some would be lucky enough to have combinations of genes from its two parents that resulted in its looking even more like the nasty species and their lives would be even more likely to be spared. The final result of these repeated generations of selection in favor of the mimics would be the evolution of an essentially perfect mimic.
Metaphorical Darwinian models of cultural and historical behavior do not contain genes, but contain cultural variants that arise like gene mutations and that are somehow differentially propagated over time in human minds and institutions, resulting in cultural evolution. The first, rather simple formulation of such a model in 1982 by Richard Dawkins7 contains elementary particles of culture, memes, playing the role of genes, which are propagated to greater or lesser degrees because they are more or less appealing to people. The memes might be ways of pronouncing the letter r, or whether the color associated with death is white or black, or whether one prefers Luther to the Pope. In this model human beings are the carriers of the cultural particles, the physical propagators of these particles through communication, and they provide the environment that determines which memes are successful.
There have been a number of more or less complex variants on this original elementary metaphor for genetic evolution and it is generally agreed that the most nuanced and sophisticated version is contained in the work of Robert Boyd and Peter Richerson, and laid out in considerable detail in Not By Genes Alone. The title is meant to suggest that cultural evolution is not simply like, but is part of, the entirety of human evolutionary change. The authors begin by asserting, quite correctly, that culture is part of human biology partly because evolved neural structures that underlie psychological states must have some influence on what people believe and perceive and partly because the culture creates an environment in which future physical evolution by natural selection takes place. We could not have our present automotive culture without a certain minimum of depth perception. Moreover, since automobile accidents are the leading peacetime cause of death, by far, among people of reproductive age in technologically advanced countries (about one death per one hundred persons in this age group per generation in the United States), genes that favor short reaction time to perceived danger must be increasing in our population, slowly but inexorably.
Richerson and Boyd reject the simplistic model of gene-like “memes,” but they are rather vague, as they must be, on how to recognize culture or its structure. They are aware that one aspect of culture will change in reaction to and in concert with other aspects of culture, that there is a complex network of causal dependency among parts of culture. Changes in technology, occupation, education, political attitudes, division of household labor and parental responsibility, leisure activities, and styles of speech and dress are connected as both causes and effects within and between generations.
The invention and spread of computers are the direct cause of major changes in patterns of education and leisure as books are replaced by on-line databases and computer games. They are the agents of the creation of new occupations and new methods of work, of changes in vocabulary and in volume and speed of interchange between individuals as well as the possibility that one person can communicate with large populations without the intervention and control of public media. They create the ability to purchase immediately a vast array of goods and services and to have access to a vast quantity of stored information.
All of these changes in turn feed back onto the development of further computer hardware and software, developments that amplify the effects already seen and create new forms of production, commerce, communication, and education. The difficulty that this complexity presents for making models of cultural change and diversification is that it has no clear structure. That structure has to be invented.
In Richerson and Boyd’s formulation, cultural elements, ideas, tastes, languages, and attitudes are properties of individual human carriers who acquire them by a great variety of processes including conscious and unconscious imitation of others, direct teaching by parents, learning in formal educational settings, or by exposure to various forms of communication. Changes in frequency of cultural variants among specific populations occur by two basic mechanisms. First, there are biases in the transmission of cultural elements, some elements being more popular or easier to learn or simply more frequent among those from whom we acquire our culture. That might explain the spread of, for example, hard rock. Second, in a purely Darwinian mode, the carriers of some cultural variants may survive better or have more children. All other things being equal, the religious beliefs of those who oppose contraception on principle ought to be spreading like wildfire. The differential rate of reproduction and the biases in transmission are, of course, dependent on environment, but Boyd and Richerson recognize that the human environment is itself largely a consequence of culture so that cultural change is both the cause and effect of further evolution.
This model has some shortcomings. One is that much of one’s culture is not acquired from other persons. When I walk down the street in Florence I do not have to hear anyone speak or read any sign to know that I am not anywhere in America. Buildings look strange, streets look strange, things have a strange smell, people carry their bodies in an unfamiliar way. I become conscious of a culture different from my own, a culture that I acquired throughout my development simply by walking down the street and being bombarded by sense impressions. Another is that no model of cultural evolution of which I am aware takes account of power. The people of Bavaria are predominantly Catholic while Westphalians are Protestant, not because somehow Lutheranism was more appealing to northerners but because at Augsburg in 1555 the warring German princes and the Holy Roman Emperor made peace using the rule of cuius regio, eius religio, which allowed rulers to enforce their own religion in their own dominions and to expel those who were recalcitrant.
The most important question is why we should use a Darwinian model at all for history and culture. The population model of variation, inheritance, and different rates of reproduction has been specifically designed to explain a particular set of natural phenomena that have a well-known empirical and mechanistic base. Even Darwin, who had no idea of genes or of the rules of inheritance, knew that organisms were reproduced only by other organisms, that offspring resembled their parents more in concrete physical characteristics than they resembled individuals not related to them, and that more organisms were reproduced than could survive to reproductive age. That was no guarantee that his model for evolution would have to be entirely correct because it might have turned out that there was significant inheritance of acquired characters.
Cultural evolutionists have no set of phenomena of comparable concreteness. They can’t even reach an agreement on how to define and describe their objects of interest. The arguments offered by Boyd and Richerson for adopting a Darwinian model of cultural change are all epistemological: they serve an intellectual interest but cannot be said to accord better with the phenomena that they are meant to explain. They say of their arguments, for example, that “they provide islands of conceptual clarity in the midst of otherwise mind-numbing complexity and diversity”; that “they are productive of further work”; that they are “economical” of human intellectual labor; and that they will “increase the chance that we will detect useful generalizations in spite of the complexity and diversity of human behavior.”
That a theoretical formulation is desirable because it makes it easier and more efficient to write more articles and books giving simple explanations for phenomena that are complex and diverse seems a strange justification for work that claims to be scientific. It confuses “understanding” in the weak sense of making coherent and comprehensible statements about the real world with “understanding” that means making correct statements about nature. It makes the investigation of material nature into an intellectual game, disarming us in our struggle to maintain science against mysticism. We would be much more likely to reach a correct theory of cultural change if the attempt to understand the history of human institutions on the cheap, by making analogies with organic evolution, were abandoned. What we need instead is the much more difficult effort to construct a theory of historical causation that flows directly from the phenomena to be explained. That the grand historical theorists of the past tried and failed to do this does not foreclose further efforts. After all, Darwin was preceded by eminent failures and even he did not get it all right.
Richard Dawkins, The Extended Phenotype: The Gene as the Unit of Selection (Freeman, 1982).↩
Darwin & Progress December 15, 2005
Richard Dawkins, The Extended Phenotype: The Gene as the Unit of Selection (Freeman, 1982).↩