As a molecular anthropologist, my research involves using genetic data to address questions of anthropological interest about the origins, history, migration, structure, and relationships of human populations. I frequently am asked to give lectures to nonspecialist audiences on insights from genetics into human evolution, and invariably during the ensuing discussion period the viewpoint will be expressed that while yes, humans may have evolved during the distant past, surely humans have stopped evolving, because of culture: if something changes in our environment, we respond culturally, not biologically. For example, if the ozone layer continues to disappear, and levels of ultraviolet radiation reach life-threatening levels, we will most likely respond by developing protective skin creams and clothing, moving our cities underground, etc., and not by evolving thicker skin or hair. Conversely, many research groups (including my own) have become interested in detecting and analyzing recent biological evolution in humans, which would seem to contradict the above viewpoint. I therefore have been thinking a lot lately about the role of culture in human evolution, and I thought this might make an interesting topic for this forum (at least, I would be interested in the responses I get).
First, some terminology and background, especially for the nonspecialist. “Evolution” has different meanings to different scientists; a population geneticist, for example, views evolution simply as changes in allele frequencies (that is, the frequencies of the variant forms of a gene) over time. Such changes are usually random, reflecting the fact that not everybody leaves offspring, so by chance some alleles increase in frequency and others decrease in frequency over time. These random fluctuations, known as genetic drift, occur more rapidly in small populations than in large ones. Genetic drift results in loss of genetic variation within populations and increases in genetic differences among populations over time, and is countered by migration among populations, which restores genetic variation within populations and decreases genetic differences among populations. Thus, to a population geneticist, since allele frequencies are always changing because of drift and migration, by definition evolution is always happening, and it therefore makes no sense to say that humans are no longer evolving.
But to most people who are not population geneticists, biological evolution means natural selection, in the Darwinian sense: increase in the frequency of an inherited trait which enhances the survival and/or reproductive success of individuals with that trait, also referred to as genetic adaptation. Often, this is expressed as a response to a change in the environment, which in turn leads to a change in those traits that confer enhanced survival/reproduction. Familiar examples of genetic adaptations that resulted in human evolution include bipedality, increased brain size, loss of body hair, and variation in skin pigmentation. To say that humans have stopped evolving, then, is to say that such inherited traits no longer matter when it comes to how humans respond to their environment. This is the view that I often hear: culture acts as a barrier or a buffer between us and the environment, thereby preventing human evolution.
However, if culture is a buffer, it is an imperfect one. For example, humans are plagued by a variety of infectious diseases, and for every success story (e.g., eradication of smallpox and polio) there are diseases that resist our efforts at finding vaccinations or cures (e.g., malaria and AIDS). And you can be sure that if our culture is unable (or unwilling) to do what it takes to prevent or cure a disease, then genetic resistance will indeed occur and will increase in frequency. Some classic examples of natural selection in humans involve genetic variants that increase resistance to malaria, such as sickle-cell anemia. Genetic variants that increase resistance to AIDS have been identified, and it is a safe bet that such variants will increase in frequency if there is no cure/vaccination for AIDS – but such increase comes at the expense of individuals who do not carry such genetic variants. Evolution in response to infectious disease is thus an ongoing story in humans.
But there is an alternative view to that of culture as a (leaky) barrier to human evolution, which can be expressed as follows: humans have been evolving and continue to evolve, not just in spite of culture, but because of culture. That is, cultural practices have actually caused humans to evolve, and a classic example is lactose tolerance. The story goes as follows: lactose is the major sugar present in mammalian milk, and most mammals stop making lactase, the enzyme that digests lactose, shortly after weaning because they are never again exposed to lactose in their diet. This, incidentally, is a nice example of the evolutionary principle of “use it or lose it”: there is no need to continue making lactase if there is no lactose in the diet. Some humans are weird, however, in that they retain the ability to digest lactose into adulthood. It turns out that the frequency of this trait, known as lactose tolerance (or lactase persistence), is highly correlated with milk-drinking populations in Europe and Africa, and was apparently driven to high frequency by natural selection in those populations. Thus, a human cultural trait – domestication of cattle, thereby providing cow’s milk as a new source of nutrition – resulted in human evolution (namely, an increase in lactose tolerance).
Even more provocatively, recently the view has been put forth that not only has culture influenced human evolution, culture has actually increased the rate of human evolution. According to this view, cultural traits such as the invention and spread of agriculture, domestication of animals, increasing population density and urbanization, etc., have influenced recent human evolution much more dramatically than has the environment. The evidence for this view comes largely from studies that find numerous signals of selection in the patterns of genome-wide genetic variation in human populations. That is, we expect that selection for an inherited trait in a particular population will alter patterns of variation at the responsible gene(s): in general, we expect larger than average genetic differences between populations, and unusually long haplotypes (chromosomal segments), to be associated with such genes. There have been numerous studies looking for such signals of selection in genome-wide data from various human populations, and invariably numerous signals of selection are claimed to have been found. Moreover, some analyses indicate that such signals of selection in our genomes have been accumulating recently, indicating that selection has become more prevalent of late, rather than less prevalent as might be expected if culture is increasingly acting as a barrier to human evolution. The most likely explanation for an increase in recent times in genomic signals of selection would appear to be that culture is indeed driving human evolution.
However, some caveats are in order. First and foremost, there is an ongoing controversy over the reliability of genome scan approaches for detecting selection. It turns out that demographic processes – in particular, population growth and geographic expansion, both of which have certainly been important in human history – can mimic the expected genomic signals of selection. Thus, at least some signals of selection are likely to be false positives and not due to selection at all – and there are those who would argue that this holds for the majority of such signals. If the critics are right and the majority of such signals are indeed false positives, then the evidence for culture driving human evolution disappears. My own view is that the role of demographic processes in producing spurious genomic signals of selection certainly deserves more attention. However, I am fairly confident that the genome-wide approaches do provide at least some evidence for selection in humans, for two reasons. Firstly, one of the predictions we would make is that if a signal of selection on a particular gene is real, then there should be a functional difference between the putatively-selected and non-selected variants of that gene. We have tested that prediction in three cases, and in all three cases we do indeed find a functional and/or phenotypic difference (for further details, see Hughes et al. 2008, Bryk et al. 2008, and Ryan et al., 2009) . This is a necessary, albeit not sufficient, indication of selection – it still needs to be demonstrated that the functional difference has resulted in a trait subject to selection – but I emphasize that in all three cases that we tested, the gene was selected for further study solely on the basis of exhibiting a strong signal of selection in a genome-wide study. So this makes me think that some of the candidates identified by genome-wide studies have indeed experienced selection. Secondly, there is good reason to think that there are many more false negatives than false positives in genome-wide studies of selection. We know, from computer simulations, that genome-wide methods only detect very strong selection – weak selection, in which the fitness advantage provided by those with the trait is only slightly larger than for those lacking the trait, will be missed. And since it is quite likely that weak selection is far more prevalent than strong selection, our present genome-wide studies are detecting only the tip of the iceberg. Thus, I do think that that genome-wide studies are, if anything, underestimating the role of selection.
But there is a more important – and less widely-appreciated – caveat to the assertion that culture is driving recent human evolution, and that is that our current methods for detecting signals of selection based on genome-wide studies can only detect recent selection. The genomic signature of selection that has happened in the distant past will be erased by subsequent mutations and recombination, and after some time will no longer be detected by our methods. So a crucial question is: how far back in time can selection be detected reliably? The answer is we don’t know for sure, but a best guess would be on the order of 10,000 – 20,000 years for our current methods of detecting selection, which happens to coincide with the period of time when the rate of selection has supposedly increased during human evolution. Thus, the apparent recent increase in signals of selection that supposedly has been driven by culture could in fact be just an artifact of our methods for detecting selection; maybe there was just as much or even more selection in the distant past, that was driven by the environment and not by culture, but our methods cannot detect the signal of such older selective events in our genome.
To conclude, it is clear that humans have been evolving recently and are continuing to evolve. It is also clear that humans have evolved because of culture, as witness the lactose tolerance trait. However, whether or not culture has been the main driving force in recent human evolution remains to be seen.
Some Selected References:
- Balter, M. (2005). Are humans still evolving? Science 309, 234-237.
- Bryk, J., Hardouin, E., Pugach, I., Hughes, D., Strotmann, R., Stoneking, M., and Myles, S. (2008). Positive selection in East Asians for an EDAR allele that enhances NF-kappaB activation. PLoS One 3, e2209.
- Cochran, G., and Harpending, H. (2009). The 10,000 Year Explosion: How civilization accelerated human evolution. New York: Basic Books.
- Hancock, A., and Di Rienzo, A. (2008). Detecting the genetic signature of natural selection in human populations: models, methods and data. Annu Rev Anthropol 37, 197-217.
- Hawks, J., Wang, E.T., Cochran, G.M., Harpending, H.C., and Moyzis, R.K. (2007). Recent acceleration of human adaptive evolution. Proc Natl Acad Sci U S A 104, 20753-20758.
- Hofer, T., Ray, N., Wegmann, D., and Excoffier, L. (2009). Large allele frequency differences between human continental groups are more likely to have occurred by drift during range expansions than by selection. Ann Hum Genet 73, 95-108.
- Hughes, D.A., Tang, K., Strotmann, R., Schoneberg, T., Prenen, J., Nilius, B., and Stoneking, M. (2008). Parallel selection on TRPV6 in human populations. PLoS One 3, e1686.
- Kelley, J.L., and Swanson, W.J. (2008). Positive selection in the human genome: from genome scans to biological significance. Annu Rev Genomics Hum Genet 9, 143-160.
- Lopez Herraez, D., Bauchet, M., Tang, K., Theunert, C., Pugach, I., Li, J., Nandineni, M.R., Gross, A., Scholz, M., and Stoneking, M. (2009). Genetic variation and recent positive selection in worldwide human populations: evidence from nearly 1 million SNPs. PLoS One 4, e7888.
- Pickrell, J.K., Coop, G., Novembre, J., Kudaravalli, S., Li, J.Z., Absher, D., Srinivasan, B.S., Barsh, G.S., Myers, R.M., Feldman, M.W., et al. (2009). Signals of recent positive selection in a worldwide sample of human populations. Genome Res 19, 826-837.
- Ryan, A.W., Hughes, D.A., Tang, K., Kelleher, D.P., Ryan, T., McManus, R., and Stoneking, M. (2009). Natural selection and the molecular basis of electrophoretic variation at the coagulation F13B locus. Eur J Hum Genet 17, 219-227.
- Sabeti, P.C., Varilly, P., Fry, B., Lohmueller, J., Hostetter, E., Cotsapas, C., Xie, X., Byrne, E.H., McCarroll, S.A., Gaudet, R., et al. (2007). Genome-wide detection and characterization of positive selection in human populations. Nature 449, 913-918.
- Voight, B.F., Kudaravalli, S., Wen, X., and Pritchard, J.K. (2006). A map of recent positive selection in the human genome. PLoS Biol 4, e72.