Animal Culture Wars

Is culture uniquely human? Or do some nonhuman animals have culture?

How we answer this question is crucial for our understanding of how we fit in the natural world. As I have indicated in Darwinian Natural Right and Darwinian Conservatism, Immanuel Kant originally formulated the modern concept of culture as that uniquely human realm of artifice in which human beings escape their natural animality to express their rational humanity and their freedom from the laws of nature. In doing this, Kant extended Thomas Hobbes' dichotomy of animal nature and human artifice in rejecting Aristotle's biological understanding of human beings as political animals by nature.

In Aristotle's biological studies of social and political behavior, he did not separate animal instinct and human learning. He saw that, in varying degrees, all social animals have natural instincts for social learning. Of course, in the capacity for learning, human beings far surpass other animals. Still, it is as true for nonhuman animals as it is for human beings that social life emerges as a joint product of nature and nurture. So, for example, Aristotle observed that some birds teach their young to sing, and the languages they teach differ in different localities in a manner comparable to the diversity of human languages, which suggests that among birds "language is not natural in the same way as voice but can be trained" (HA, 536b18-20).

Charles Darwin made the same observation about variable dialects of bird songs transmitted as traditions by social learning, and he saw this as showing "that an instinctive tendency to acquire an art is not peculiar to man" (Descent of Man, 2nd ed., Penguin Classics, p. 106-109). More recently, Peter Marler and other biologists studying bird-song dialects have presented this as an example of animal culture--that is, a behavioral tradition transmitted by social learning.

Over the past two decades, there has been an intense controversy over the possibility of animal culture. On the one hand, there has been growing evidence for elaborate cultural traditions among animals--particularly, chimpanzees, orangutans, whales, and dolphins. One highlight of this new research was the report in 1999 in Nature by Andrew Whiten et al. comparing seven chimp study sites in Africa and showing 39 behavior patterns customary in some communities but absent in others, so that each chimp community has its distinctive profile of behavioral patterns, which is comparable to the cultural diversity of human communities. On the other hand, many critics of this research deny that this is really evidence for true culture.

This whole debate is now conveniently surveyed in the new book The Question of Animal Culture, edited by Kevin Laland and Bennett Galef, published last year by Harvard University Press. Laland and Galef have arranged the book's chapters to move from the strongest advocates of animal culture (such as Frans de Waal, William McGrew, and Andrew Whiten) to the strongest skeptics (such as Michael Tomasello, Susan Perry, and Kim Hill).

As the book makes clear, much of the controversy turns on how one defines "culture." As Laland has said, "Culture is as rare or as common as it is defined to be" (101). So, if culture is defined narrowly as requiring uniquely human traits (such as symbolic teaching and moral norms), then nonhuman animals cannot have culture. But if culture is defined broadly as the transmission of behavioral traditions through social learning, then there is plenty of evidence for animal culture.

Under the broad definition of culture, the evidence for animal culture comes both from field studies of wild animals and laboratory experiments with captive animals. So, for instance, the comparative study of wild primate communities does show remarkably complex variation in the behavioral profiles of primate communities, which suggests social traditions initiated by individual innovators that have been passed on by social learning. But the inference that these behavioral profiles really do arise from socially learned traditions is largely circumstantial and based on arguments of plausibility. This inference is reinforced, however, by experimentation with captive primates showing that individual primates can learn a new behavioral pattern and then others in the primate group can learn the new behavior by imitation, which then becomes a group tradition.

But under the narrow definition of culture, we might concede that nonhuman animals show socially learned behavioral traditions, while still denying that this is sufficient to count as "culture" in the full sense. This position is best stated by Kim Hill. Hill is an interesting case. As an anthropologist and behavioral ecologist using models from evolutionary ecology to explain the behavior of hunter-gatherers, Hill has for many years rejected the claim of his colleagues in cultural anthropology that human culture is too unique to be explained as animal behavior. But recently Hill has changed his mind, and now he thinks the behavioral ecologists and evolutionary psychologists are mistaken in not facing up to the uniqueness of culture for human beings. Hill thus belongs to a growing movement among Darwinian social scientists who now see that a Darwinian account of human behavior requires explanation of the co-evolution of genes and culture in human history.

Hill agrees that nonhuman animals are capable of socially learned behavioral traditions. But he argues that such animal traditions differ in kind from human cultures. They differ in two respects. First, human culture shows a ratchet effect in which socially learned information accumulates over time to achieve ever more complex forms, while social learning among animals doesn't show such accumulation. Chimps can learn to crack open nuts with pointed stones used as hammers and flat stones used as anvils, and this technology can become a behavioral tradition in certain chimp communities. But they don't progressively improve on their technology to develop ever more intricate tools.

The second difference is that the content of human culture is unique. Human culture shows three categories of socially transmitted information. Animal traditions show the first but not the other two. The first category is "socially learned techniques, technology, and environmental information (traditions, beliefs)" (275). This can be seen among many animals.

The second category of socially transmitted information is unique to human culture--"regulations of individual behavior enforced by rewards and punishments (norms, conventions, institutions, laws)" (276). Other animals manifest some rules of behavior--for example, rules among chimps about subordinate individuals deferring to dominant individuals. But these rules of behavior are not enforced by third-party punishment and reward. Human beings will punish someone for violating social norms even when the punishment is costly to them and does not directly benefit them. "Morality is demonstrated by altruistic third-party punishment" (286).

Other animals don't show such altruistic punishment. For example, when human beings play the Ultimatum Game, responders will punish proposers who make unfair offers even when the punishment is costly for the responders. But when chimps play the Ultimatum Game, responders will not engage in such costly punishment.

The third category of human culture, which is not seen in other animals, is "symbolic means of reinforcing, and signaling adherence to, a specific rule system" (277). Human beings invest their social rules with such emotional weight that they become moral rules, so that violation of the rules elicits moral emotions of guilt, disgust, and anger. Often these moral rules are reinforced with religious sanctions, including supernatural rewards and punishments. And often human beings use symbolic ethnic markers--for example, dialects, clothing, ritual behavior, and dietary restrictions--to signal their commitment to a distinctive set of social norms, which allows them to interact with those who share the same norms, while avoiding those who don't.

What are these distinctively human social norms? Hill suggests that hunter-gatherer societies show a universal human pattern: "Hunter-gatherers universally regulate access to valuable resources (e.g., food, mates) and regulate how competition for these resources may be legitimately expressed." This would include the following social norms regulating behavior (279-80):

1. Mate access
a. Prohibitions and prescriptions (applied on the basis of age, kin, or ritual-group membership)
b. Polygyny (degree allowed and who may practice it)

2. Food production
a. Land use (territoriality)
b. Specific resource rights (ownership of specific plants or animals)
c. Niche specialization (informal trade unions)

3. Food redistribution
a. Sharing (who receives, how much they receive, and, what body parts of some game species)
b. Consumption taboos (applied on the basis of age, kin, or ritual-group membership)

4. Display rights (ritual participation)
a. Mating (who may participate in organized displays that are important forums for mate choice)
b. Other sociopolitical messages (who has the right to "broadcast" symbolic messages and in what context)

5. Access to kin and other allies
a. Residence rules (who is allowed to reside with close kin)
b. Activity and ritual regulations (who can be a member in some organized activities)

6. Political power
a. Designated positions (reserved for specified age, sex, kin or ritual-group members)
b. Transfer of power (rules of succession, turn taking, or context-specific leadership)

7. Regulation of violent conflict
a. Within-group contests (ritual dueling, divining, and justice)
b. Participation in social group defense (who may or must defend the group and in what contexts)

8. Regulation of life history
a. Age at first reproduction (acceptable age for sexual relations and marriage)
b. Investment in infants and juveniles (who must invest and in what contexts)
c. Age- and sex-specific rights and restrictions (that change during the life course) in competition over resources

Notice how these social norms correspond to my list of 20 natural desires, which I take to be rooted in evolved human nature. Notice also how these social norms underlie the natural moral sense as enforced by our moral emotions.

For example, "prohibitions and prescriptions" for "mate access" would include incest taboos. Other animals have behavioral mechanisms for avoiding incest. But only human beings reinforce incest avoidance with a deeply emotional incest taboo that cannot be violated without feelings of moral repugnance.

Chimps might avoid incest. But they don't have an incest taboo as a moral norm, because they don't have a "conscience" that is expressed as moral emotions such as guilt and anger. Hill concludes from this that chimps are like psychopaths--that is, they lack the moral emotions required for a moral sense or conscience. Human psychopaths apparently are innately psychopathic because they have alleles that incline them to emotional poverty, so that they don't feel the moral emotions that are normal for the rest of us. This suggests that the human moral sense arose from a genetic adaptation that separated normal members of Homo sapiens from our primate ancestors. This innate moral sense is now universally expressed in all human societies, although the presence of psychopaths as a small proportion of each society shows that there is still an adaptive niche for those without any conscience.

This moral sense as expressed in deep moral emotions is the ground for "normativity," for our feeling that one ought to obey our social norms. This normative ought is unique to human culture. But if it has an evolutionary history--the evolution of a propensity for moral emotions in the human brain and nervous system--then we can see it as an emergent property of human evolution. And like every emergent property, it shows an emergent difference in kind with an underlying difference in degree.

We can put this all in the evolutionary framework suggested by Eva Jablonka, distinguishing four dimensions of evolutionary history--genetic, epigenetic, behavioral, and symbolic. Animal traditions belong to behavioral evolution. Human morality belongs to symbolic evolution, which is a system of evolutionary inheritance that is unique to human beings.

Some posts on related themes can be found here, here, here, here, here, here, here, here, here, here, and here.