Peter Richerson

Robert Boyd

Not By Genes Alone

Milk was once marketed in the United States with the slogan, 'Every Body Needs Milk.' Catchy, but it's not true. Most people not only don't need milk, they can't tolerate it. The majority of the world's adults lack the enzyme necessary to digest lactose, the sugar in milk, and if they drink milk, the lactose is fermented by bacteria rather than absorbed by the gut, leading to uncomfortable attacks of flatulence and diarrhea. That we didn't know this until the 1960s is testimony to how scientists are blinkered by their cultural background—most nutritionists came from countries where adult lactose malabsorption is rare. It is also testimony to how small a role evolution plays in biomedical science, because even a little adaptationist thinking would have suggested that it is the ability to digest milk that is abnormal, not the reverse. Milk has always been baby food for mammals, and lactose only occurs in mother's milk. Thus, adult mammals had no need for the enzyme that cleaves lactose. Unsurprisingly, ever frugal natural selection shut down the production of this enzyme after weaning in almost all mammal species. The majority of people exhibit the standard mammalian developmental pattern; they can digest milk as infants but not as adults. The real evolutionary puzzle is why in some human populations most adults can digest lactose.

In the early 1970s, geographer Fredrick Simoons suggested that the ability to digest lactose evolved in response to a history of dairying. The people of northwest Europe have long kept cows and consumed fresh milk. Dairying was carried to India by 'Aryan' invaders, and has been practiced by pastoralists in western Asia and Africa for millennia. In each of these regions, most adults can drink fresh milk. Mediterranean dairying people traditionally consume milk in the form of yogurt, cheese, and other products from which the lactose has been removed. Some adults in these populations can digest lactose while others cannot. Dairying is rare or absent in the rest of the world, and few Native Americans, Pacific Islanders, Far Easterners, and Africans are lactose absorbers. Simoons's hypothesis was controversial at the time, but subsequent genetic data confirm that adult lactose digestion is controlled by a single dominant gene, and careful statistical work indicates that a history of dairying is the best predictor of a high frequency of this gene. Moreover, calculations indicate that there has been plenty of time for this gene to spread since the origin of dairying.

The evolution of adult lactose digestion is an example of 'gene-culture coevolution.' Biologists developed the term coevolution to refer to systems in which two species are important parts of each other's environments so that evolutionary changes in one species induce evolutionary modifications in the other. This can lead to an intricately choreographed coevolutionary dance, often with surprising results. For example, normally predatory ants often tend aphids, protecting them from predators. The aphids reward their ants by exuding sugar-rich honeydew, which the ants collect.

The evolving pools of cultural and genetic information carried by human populations are partners in a similar swirling waltz. Genetic evolution created a psychology that allows the cumulative cultural evolution of complex cultural adaptations. In some environments, this process led to the evolution of the dairying traditions. This new culturally evolved environment then increased the relative fitness of the gene that allows whole-milk consumption by adults. As that gene spread, it in turn may have changed the environment-shaping cultural practices, perhaps favoring more whole-milk consumption, or more serendipitously, giving rise to the evolution of ice cream.

We think that gene-culture coevolution has also played an important role in the generic evolution of human psychology. If genetically maladaptive cultural variants are an inevitable consequence of cumulative cultural adaptation, then the pools of cultural and genetic information carried by human populations each respond to their own evolutionary dynamic. Natural selection, mutation, and drift shape gene frequencies, while natural selection, guided variation, and a variety of transmission biases mold the distribution of cultural variants. However, these two processes are not independent.