Rape as an adaptation

Is this contentious hypothesis advocacy, not science?

A Natural History of Rape: Biological Bases of Sexual Coercion

MIT Press: 2000. 272 pp. $28.95, £17.95

In A Natural History of Rape, Randy Thornhill and Craig Palmer argue that rape is an adaptation — that it has evolved to increase the reproductive success of men who would otherwise have little sexual access to women. Their analysis of rape then forms the basis of a protracted sales pitch for evolutionary psychology, the latest incarnation of sociobiology: not only do the authors believe that this should be the explanatory model of choice in the human behavioural sciences, but they also want to see its insights incorporated into social policy. Thus, in a single, slim volume, Thornhill and Palmer give us both an inflammatory analysis of a sensitive topic, and a manifesto outlining evolutionary biology's future conquest of the social sciences.

In the furore that has greeted the book's publication, the scientific evidence for the authors' arguments has been largely ignored. However, it is here that we must start. If their specific claims about rape are not scientifically sound, then the authors' grand vision of the centrality of natural selection to every aspect of our behaviour collapses as well. In their media appearances, Thornhill and Palmer cloak themselves in the authority of science, implying that the controversy over their ideas is purely political, and that the underlying biology is unimpeachable. This is a serious misrepresentation.

What persuasiveness the book does possess rests on an ingenious rhetorical trick. The authors lay out two alternative evolutionary hypotheses: rape is either a ‘specific adaptation’ (that is, natural selection explicitly promoted the act) or a ‘by-product of evolution’ (there was no direct selection for rape; rather it is an accidental product of selection for, say, male promiscuity and aggression). Readers unconvinced by the specific-adaptation argument therefore find themselves embracing by default the by-product alternative. Either way, Thornhill and Palmer claim a convert.

But what, in behavioural terms, is an evolutionary by-product? Everything that is not a specific adaptation. Thus, playing the piano — an activity unlikely to have been instrumental in the evolution of the brain — is an evolutionary by-product, because it depends on a brain that was itself produced by natural selection. If every human behaviour can be seen as a by-product of evolution, then the by-product idea is useless, for a theory that explains everything is merely a truism. The claims that rape and playing the piano are by-products of evolution are claims without content.

It is not surprising, then, that A Natural History of Rape is largely an argument for the specific-adaptation theory. The authors' evidence, however, either fails to support their case, is presented in a misleading and/or biased way, or equally supports alternative explanations. The following examples illustrate each of these failings.

First, Thornhill and Palmer make much of the claim that rape victims tend to be in their prime reproductive years, suggesting that reproduction is indeed a central part of the rapist's agenda. But the data they present contradict this claim. In a 1992 survey that attempted to deal with the substantial statistical problem of unreported rape, 29% of US rape victims were under the age of 11. As that age group comprises approximately 15% of the female population, under-11s were over-represented among rape victims by a factor of two. So invested are the authors in their specific-adaptation hypothesis that they try to explain this non-adaptive anomaly by noting that the data do not indicate the “proportion of the victims under 11 who were exhibiting secondary sexual traits”. Further, “the increasingly early age of menarche in Western females contributes to the enhanced sexual attractiveness of some females under 12”. In the end, the hopelessness of this special pleading merely draws attention to the failure of the data to support the authors' hypothesis.

Second, the authors contend that, based on sociological studies, female rape victims of reproductive age are more traumatized by the experience than are women either too old or too young to reproduce. The rationale is that reproductive-age women are in effect mourning the lost opportunity for mate choice which rape, in the world view of evolutionary psychology, represents to them. The authors see this apparent heterogeneity in the reaction to rape as supporting their claims about the reproductive essence of the act.

In checking the cited reference (one of its authors is Thornhill himself), we find that the original work's conclusions differ critically from those given in the book. According to Thornhill and Palmer, the cited study shows post-rape trauma to be higher in reproductive-age women (age 12–44) than in the two other age classes (under 12 and over 44). In fact, the data show that the only heterogeneity in response to rape comes from the under-12 class: the over-44 class is just as traumatized as the 12–44 one. However, when the over-44 and under-12 classes are pooled, the under-12 effect of less trauma makes this combined ‘non-reproductive’ class significantly different from the 12–44 one. The authors have used statistical sleight of hand to buttress their argument. And we need hardly point out that the relative lack of trauma in the youngest age group may be unrelated to sexual immaturity: rather, children may be less able to express their feelings. Furthermore, the original study's data are questionable because much of the assessment of trauma in the under-12 class was necessarily based on reports from the child's care-givers rather than from the child herself. Direct comparison of observer-reported and self-reported data on such a subjective issue is extremely problematic.

Finally, the fact that women of reproductive age experience more violence during rape than do older women or children — suggesting that they fight back harder — is taken by the authors as evidence that they have more to defend. There is, they contend, more at stake — reproduction, no less — for reproductive-age women. While it is true that women of reproductive age who resist rape may be partly motivated by the fear of unwanted conception, it is also true that such women, at the peak of their bodily strength, are most physically capable of fighting back. Children cannot fight off a full-grown man, and older women may also find resistance beyond them. In exclusively championing their preferred explanation of a phenomenon, even when it is less plausible than alternatives, the authors reveal their true colours. A Natural History of Rape is advocacy, not science.

We have highlighted just three examples of the book's flawed arguments. There are many more. The evidence that rape is a specific adaptation is weak at best. In keeping with the traditions established early in the evolution of sociobiology, Thornhill and Palmer's evidence comes down to a series of untestable ‘just-so’ stories.

Sociobiological approaches to human behaviour may yield interesting insights. But it is disciplinary hubris — a long-standing feature of evolutionary psychology — to suppose that natural selection underlies our every action. Because of the central role of reproduction in Darwin's theory, sexual behaviour is, in principle, a good candidate for fruitful sociobiological study, but even here it usually fails dismally. The most imaginative and committed sociobiologist would be hard-pressed to show that masturbation, sadomasochism, bestiality, and pornography's enthusiasm for high heels are all direct adaptations. In its insistence on forcing everything into the strait-jacket of adaptation, evolutionary psychology offers a woefully incomplete perspective on human behaviour. Thornhill and Palmer have inadvertently revealed just how deficient that perspective is.

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Coyne, J., Berry, A. Rape as an adaptation. Nature 404, 121–122 (2000). https://doi.org/10.1038/35004636

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing