Towards the end of my academic career, I was appointed to be an academic member of an NIH Special Review Committee tasked with evaluating about a half dozen grant applications submitted in response to a “Request For Applications (RFA)” from the National Institute of Mental Health (NIMH). An official at this agency was interested in the burgeoning field of behavioral genetics, and by issuing the RFA, the NIMH was hoping to stimulate some interesting and useful research in the subject.

I remember being quite unimpressed with any of the proposals, and as it turned out, my views were shared by the rest of the committee. We recommended none of them for funding. I don’t remember the details, but I do recall that all of the proposals seemed very speculative and contained some major unproven assumptions, not much supporting preliminary data, and very weak hypotheses. In all my years as a reviewer, and my subsequent six years as Division Director for Physiological and Pathological Sciences  at the NIH Center for Scientific Review, where thousands of grant applications passed through each year, I have never experienced such a dismal result in the peer-review process.

At the end of the meeting, the official in charge of the (now defunct) program asked us what the problem was. I was the first to respond, and I said that none of the applications were meritorious enough to be considered for funding, because the field was just too immature and the data already gathered were not at all convincing that there was in fact any real connection between any gene or gene variant and human behavior. Another issue brought up by several members of the committee was the implicit racial bias related to the search for genes that “make people violent.”

Of course, that was many years ago, and in the interim, more research has been done. I am sure if the meeting were to be held now, there would be some worthwhile projects proposed. In fact, while I am not at all in the loop anymore, I fully expect that many grants in behavioral genetics have been funded. The field is certainly not dead, and papers are being continually published. But I remain skeptical in general, and especially with respect to the issue of genetics and moral values. Here’s why.

I still think that the connection between genetics and moral behavior is less than straightforward to say the least. There is of course, no question that genes play a role in behavior throughout evolutionary history. Evolution is all about population survival, and the beneficial effects of certain behaviors like individual sacrifice for the sake of a community cannot be denied. But for most animals where individual sacrifice is known, such as with bees, this has nothing to do with a high moral value. A bee’s self-sacrifice for the good of the hive might appear to be result of a laudatory impulse on the part of individual bees, but it is nothing of the sort. No bee decides, in a moment of supreme valor to give her life for the good of the hive. There is no moral imperative acting here—only the evolutionary pressure that produces an appearance of high altruism.

The same is obviously true for the evolution of viruses, bacteria, and various animal and plant predators and parasites. No viruses (such as COVID) decided to be “good” and target the upper respiratory tract rather than the lungs. Increased transmissibility and decreased host mortality is a well-known and very frequent evolutionary strategy for infectious, lethal viruses, and the resulting good news for humans is simply an accident.

While these are extreme examples, the same sort of thing applies to a large extent (though admittedly not entirely) to behavioral traits coded for by human (as well as other animal) genes. For example, increased empathy might be expected to result in behaviors that are more “morally good” according to many human cultures. But a careful examination of the literature shows this is not always the case. Genetically determined reactions to a scene of violence, such as various autonomic reactions, as well as changes in empathy, may lead to various behavioral outcomes.

It is very well known that both emotions and behaviors are multigenic and often involve highly complex gene-environment interactions. All of this makes prediction of anything to do with moral actions based on genotypes of one or a few genes extremely difficult. One particular genetic variant in the oxytocin receptor gene (OXTR) is an example of this complexity. One paper describes a behavioral phenotype (tendency to post-partum depression) in which the A allele appears to be dominant, whereas another paper, which discusses a different behavioral phenotype (increased empathy) for the same gene, finds the G allele to be dominant. The two alleles at this locus have also been associated with a variety of other individually tested behaviors. All the studies concur that this polymorphism is in Hardy-Weinberg equilibrium, meaning that selection pressure for either allele is currently absent, and the gene probably was never a target for natural selection.

Returning to the issue of moral choices, the very fact that there is no single objective morality for all human cultures is strong evidence against an evolutionary genetic connection. There are only very small differences in allele frequencies of some of these genes in different ethnic groups, and these differences are generally not meaningful. During the grant review meeting I mentioned above, we examined one application that hypothesized that different frequencies of a particular polymorphism in a neurotransmitter gene could make African Americans more violent than European Americans. Preliminary evidence was nonexistent and the proposal was considered by the whole committee to verge on outright racism..

This does not mean that there are no genetic correlates with human acts of moral good or evil. All humans (except for rare variants like psychopaths) recognize the inherent good in caring for children, in love, and in doing kindness to kin and relatives. And all recognize that doing harm to these people is bad. There is no question that these moral understandings and resulting behaviors are of evolutionary origin, since the same behaviors are seen in many mammalian and other animal species.

But we only need to think of moral values in modern cultures that extol the killing of “enemies,” the mistreatment of people with sexual or other non-conforming phenotypes, the subjugation of women, and so on, to see that once we go deeper than fundamental common values, we find purely cultural determinants of morality, with no genetic influence at all.

Cultural evolution (which includes good explanations for different human moral codes) is fundamentally different from biological evolution, which depends on genetic variation. For morality, it is memes, not genes, that count. The uniquely human brand of cultural evolution has given us fire, shelter, the internet, and everything else that makes us more than just another ape, including the moral choices we make.