30 November 2009

Rape and Evolution IV: A response to a different history professor

Thank you for your thoughtful note.

Here is the Thornhill and Palmer assertion. “The most convincing study of pregnancy and rape in peacetime settings (Holmes et al., 1996) involved a three-year longitudinal study of a representative sample of several thousand American women. Among victims of reproductive age (12-45), the rape-related pregnancy rate was 5% per rape, or 6 percent per victim. … [T]he figures reported by Holmes et al. probably should be corrected to about 2 percent. At this time it is not known whether false rape allegations influence this percentage.”

Thornhill and Palmer (2000), 100 citing M.H. Holmes et al., “Rape-related pregnancy: Estimates and descriptive characteristics from a national sample of women, American Journal of Obstetrics and Gynecology 175 (1996): 320-325. Note Holmes et al. (1996) apparently state that the probability of conception following rape is 5.3% for women 12-17 and 4.7% for those ages 18-45.

You asked about the conception rates for consensual sex. It is 3.1%. A. J. Wilcox et al., “Likelihood of conception with a single act of intercourse: providing benchmark rapes for assessment of post-coital contraceptives,” Contraception 63 (2001): 211-215 cited in Fessler. Note that age differences make the comparison to rape-related frequency tricky.

You asked about the comparative rate of conception. “Moreover, analysis of conception rates reveals that the probability of conception following rape does not differ from that following consensual coitus.” Daniel Fessler, “Rape is not less frequent during the ovulatory phase of the menstrual cycle,” Sexualities, Evolution, & Gender 5.3 (2003): 127-147.

Note that the percentages need not be that high for natural selection to operate. “Natural selection can operate effectively with small reproductive advantages, as little as 1 percent.” Steven Pinker, The Blank Slate (2002): 368.

You ask where Thornhill and Palmer (2000) got their data. They went to the Holmes et al. 1996 study. Where did Holmes et al. go to get their data? I don’t know the answer to this. I am assuming that the American Journal of Obstetrics and Gynecology is a peer-reviewed journal that would have been sensitive to this issue, but this is just an assumption.

You point out that conception is not the same as the production of offspring who will themselves reproduce. This is correct, but all the conception-data was used is to show that there is evidence that some of the known effects of rape are consistent with evolutionary theory. Compare this to predictions made by the feminist theory (rape-is-not-about-sex) and the anti-evolution theory (rape is not an evolution-based adaptation or effect of such an adaptation or adaptations). Is there any evidence for these theories?

I’m curious as to whether you found Johnston-Robledo and McVicker’s arguments #1 through #7 (my numbering) to be weak.

1. If so, I’m not sure why a curiosity about one piece of data would have led you to sign a letter containing such arguments.

2. If not, I’m wondering why neither you nor anyone else has presented a plausible defense of any of these arguments.

Thanks again for the note. I hope your semester is finishing well,
Steve K

29 November 2009

Evolution and Rape III: Responding to a history professor

Here is my response that appeared in proftalk to an attempt by a history professor at Fredonia to defend the Johnston-Robledo and McVicker letter that she signed. Proftalk is a campus discussion forum for faculty and staff.


Thank you for your thoughtful note. Here is Johnston-Robledo and McVicker’s argument.

“To argue that rape is a reproductive strategy or that it is motivated by reproduction continues the illogic of the column. … Further, social science research has not determined that men have a reproductive motivation for rape despite Kershnar's reliance on limited evidence to suggest that they do. When men indicate that they are motivated to rape by sexual desire, this response does not mean that they are motivated by reproduction as opposed to power and control.”

Here is a restatement of their argument.

(P1) If rape is connected to evolution, then rapists are motivated by reproduction.
(P2) It is false that (in general) rapists are motivated by reproduction.
(C1) Hence, rape is not connected to evolution. [(P1), (P2)]

My objection was that premise (P1) is false and rests on a misunderstanding of how evolution works. I’m not sure I see how your points rehabilitate (P1). More generally, I don’t see why anyone would think that this argument is convincing.


You assert that the following statements are true.

1. Rape [often] results from a desire to control others.

2. The desire to control others is the product of evolution.

We thus agree that the evolutionary theory of rape is likely true. Note that even if the control theory (proposition 1) is true, this does not show that other specific theories (e.g., rape-is-about-sex theory) are false. The theories are compatible. So I’m not sure we disagree.


Johnston-Robledo and McVicker make the following claim.

“It is virtually impossible to argue that rape is primarily about sex and/or reproduction. Rape is, arguably, more about sexualized aggression than aggressive sex.”

I don’t see that you or Johnston-Robledo and McVicker have presented any evidence for this claim.

First, I don’t know how “impossible” is being used here.

Second, there is some evidence that rape is about sex. Here are a couple of examples. At least one researcher who interviewed rapists concluded that their actions were explained in part on the basis of their desire for sex. Also, a significant percentage of college men (60% in one study) report having used force to achieve sexual intimacy despite the female’s negative response. It is not implausible to think that this behavior is related to sexual desire.

Hence, even if one thought that rape is not primarily about sex, it is hard to see why it is “impossible” to argue for this claim.

In any case, it would be interesting to see who else signed Johnston-Robledo and McVicker’s letter. Perhaps they have other arguments against the broader or narrower evolutionary theories.

I hope you had an enjoyable Thanksgiving,
Steve K

21 November 2009

Evolution and Rape II: Night of the Living Feminists

The Objectivist
Misunderstanding Evolution
Dunkirk-Fredonia Observer
November 7, 2009

The broader evolutionary explanation of rape asserts evolution either directly or indirectly explains much of the rape-behavior we observe. There are three widely discussed theories as to how this explanation works. First, Randy Thornhill, a University of New Mexico biologist, has put forth a model that asserts that evolution has affected men’s psychology in such a way that in at least some men there is a tendency to think in ways that lead to opportunistic rape. Second, Craig Palmer, an anthropologist at the University of Missouri, argues that rape is an evolutionary by-product of other psychological adaptations that increase male reproductive success. Third, University of Michigan psychologist Barbara Smuts, McMaster University psychologists Margo Wilson and Martin Daly, and others argue that evolution has led to aggression against women that facilitates a long-term reproductive strategy. That is, rape is a biological adaptation by which males maintain control over females and it evolved because such control led to long-term reproductive success. Thus, the three models hold that evolution explains thought and behavior patterns that have in many cases led to rape because rape is connected in some way to greater reproductive success.

Note that the evolutionary theorists do not claim that rape is not bad or wrong, rapists are not blameworthy, rapists should not be punished, the environment plays no role in causing rape, etc.

All three models are consistent with the claim that rape is motivated in part by a desire for sex. The evolutionary theory is a minimal theory in that it is consistent with there being other motivations for rape (for example, control) and other causes (for example, cultural causes). In contrast, the opposing hypothesis is more extensive in that it holds that rape has nothing to do with evolutionary desires. Proponents of this view often view rape as focused solely on power or control and not at all on sex.

Harvard University psychologist Steven Pinker points out just how plausible it is that rape is about sex. He points out that given that men sometimes want to have sex with women who don’t want to have sex with them and men have a tendency to resort to violence to get what they want, it would be extraordinary if some men didn’t use violence to get sex. Because these tendencies have a clear connection to genes and evolution, the broader evolutionary theory should have an obvious feel to it.

SUNY-Fredonia Professors Ingrid Johnson-Robledo and Jeanette McVicker criticize the broader evolutionary theory of rape and narrower Thornhill model. Thirty other SUNY-Fredonia faculty and staff signed on to their findings. Here are some of the reasons the pair provide against the broader and narrower theories.

1. The evolutionary explanation of behavior insults men because it reduces them to biological destiny.

2. If a desire is hard-wired into a man, this implies that he is helpless (to act on it).

3. Rape usually, if not always, leads to gratuitous injury and the injury is psychological.

4. In many cases, rapists do not inflict gratuitous injuries against their victims because they do not use or carry weapons.

5. Some rapists use objects to violate women and some use a condom. Hence, the reproductive theory of rape is refuted.

6. Rapists indicate that they are motivated by sexual desire, not reproduction. Hence, rape is not explained by its connection to reproduction over millions of years of human evolution.

7. Rape has a success rate of 5% and this is quite low.

8. 18% of rape victims are under age 11.

9. There are some rape-free societies.

Most of these are irrelevant. None is anywhere near strong enough to defeat the evolutionary theories.

Consider 1. Johnston-Robledo and McVicker claim that evolutionary explanations reduce men to “biological destiny.” It is not clear what they mean here, but they probably mean that evolutionary theories entail that if men act on genetically induced desires then they are not morally responsible for what they do. This is a mistake. From the fact that a desire is in part the result of genetic factors, it does not follow that a person has no control over whether to act on it. A man can still reason with regard to whether to act on the desire. For example, if I desire to eat my neighbor’s freshly grilled steak and this desire is part innate, it does not follow that I am compelled to eat it.

Consider 2. Johnston-Robledo and McVicker argue that if, as evolutionary theory asserts, a man’s desire (or propensity) is hard-wired into him then he is helpless to act on it. This is a mistake for the reason mentioned above. How could a large flock of faculty and staff make such an obvious error?

Consider 3. Johnston-Robledo and McVicker claim that contrary to the evolutionary theory, rape usually, if not always, leads to gratuitous injuries because it leads to psychological injuries. The proponent of an evolutionary theory of rape would predict that rapists would in general avoid injuring their victims in ways that would prevent them from conceiving and bearing a child. Gratuitous is thus understood in terms of physical injuries that are not needed to carry out the rape or a direct result of it. Psychological harm is irrelevant to the theories because it is a direct result of the rape. The evidence here is striking. According to Center for Policy Research members Patricia Tjaden and Nancy Thoennes, only 5% of rape victims are severely physically injured. Remember the data are being used here to explain how a behavior came about, not whether it is bad, wrong, or harmful.

Consider 4. Johnston-Robledo and McVicker claim that in many cases, rapists do not inflict gratuitous harm because they do not use or carry weapons that would inflict severe injury. Leave aside whether this claim is consistent with the previous one. On their account, attackers who can generate enough force to invade a woman’s body can’t generate enough force to pile on additional severe physical injury (for example, by repeatedly striking them). What?

Consider 5. Johnston-Robledo and McVicker argue that because some rapists use objects to violate women and sometimes they use a condom, rape has nothing to do with reproduction or evolution. First, as Steven Pinker points out, this is true of only a minority of rapes and so it is consistent with the evolutionary theory that most rapes involve reproduction-related acts. Second, Pinker points out, some voluntary sex involves objects and condom use. It doesn’t follow that voluntary sex is unrelated to sex, reproduction, and evolution. If this is true for voluntary sex, then it is true for coerced sex.

Consider 6. Johnston-Robledo and McVicker claim that rapists indicate that they are motivated by sexual desire and not reproduction and that both motivations would apply if evolutionary theory were true. The two assume that evolution never increases reproductive fitness by producing a desire to have sex unconnected to a desire to reproduce. This misunderstands evolution. If a desire leads to increased reproductive success then evolution selects for it, regardless if the organism intends to reproduce. This is true of sexual desire.

Consider 7. The two professors claim that if rape has a 5% rate of conception, this is too low to affect evolution. Again they are wrong. If over millions of years an adaptation increases reproductive success by 5%, this will likely have evolutionary effects.

Consider 8. Johnston-Robledo and McVicker observe that many rapists target girls who are not yet fertile and they claim this weighs against the evolutionary explanation. First, if Thornhill is right, then the majority of rape-victims are fertile. At most the pair’s observation shows that some rapes will have little reproductive success, perhaps because the desire is not sufficiently focused. This is a long way off from showing that over millions of years of evolution on the African plains rape was a losing reproductive strategy. On the first model, it might be that the desire is not a sharply focused one. On the third model, raping pre-fertile girls might be a way of controlling them when they do become fertile. The evolutionary models are thus consistent with this observation.

Consider 9. Johnston-Robledo and McVicker assert that there are rape-free cultures and that this is not what the evolutionary model predicts. They likely have in mind claims by researchers like University of Pennsylvania anthropologist Peggy Sanday who report that 47% of societies are rape-free. But as University of New Mexico anthropologist Melissa Emery Thompson points out, the studies that were not focused on sexual issues, varied substantially in length, and often rested on rape-reports. Because rape is frequently hidden from researchers, especially ones not focused on sexual matters, and because rape is often not reported, the support for there being such rape-free societies is weak. Even if it weren’t weak, many genetically-linked behaviors (for example, homosexuality and aggression) are not expressed in all people and all places. In some cases, this is because environmental factors block the expression of some genes.

We’ve considered the professors’ nine objections. Seven were irrelevant and several involved obvious mistakes. The last two are relevant, but are not strong criticisms of either the broader or narrower theories. There are serious criticisms of both, but such criticisms involve a reexamination of Thornhill’s and other scholars’ data and a discussion of the different types of rape (for example, stranger versus acquaintance rape). None of this showed up in Johnston-Robledo and McVicker’s objections. They simply didn’t do their homework. Mindless adherence to feminist ideology is a poor substitute for rigorous scientific thought.

04 November 2009

Academia: Women, Babies, and Advancement

The Objectivist
Academia and Sex Differences
Dunkirk-Fredonia Observer
November 2, 2009

On average, men and women differ in their preferences with regard to work and family and this leads to workplace differences. An interesting issue is whether the academic workplace should be changed to better accommodate women’s preferences.

In some high-end fields, women have less success at the top of the career ladder. Consider, for example, law and business. Joanne Lipman writing in The New York Times reports that in 2008, women made up almost half of the associate lawyers (those who work for law firms), but only 18.3% of the partners (those who own the firms). Similarly, she reports, only 15 women run Fortune 500 companies.

The same pattern occurs in academia. There women with children or who are likely to have children do worse than their peers. Consider the following research by Mary Ann Mason, Dean of the Graduate Division at the University of California at Berkeley. Women with babies are 28% less likely than women without babies to enter a tenure-track position. A person is on the tenure track when she is on track to have a permanent full-time position in a college or university or has such a position. Married women are 21% less likely than single women to enter a tenure-track position. Women faculty are 27% less likely than men to become an associate professor (mid-level professor) and 20% less likely than men to become a full professor (senior-level professor) within 16 years. A high percentage of mothers end up in second-tier academic positions (specifically, lecturers, part-timers, and adjuncts).

Mason and Marc Goulden (a research analyst at Berkeley) point out that the academia presents family-related problems for women. They point out the following. Only a third of women without children who take a fast-track university job ever become a mother. Tenured women are more than twice as likely as tenured men to be single 12 years out from the Ph.D. If married, tenure-track women faculty are 50% more likely than men to divorce (and twice as likely to divorce as women in second-tier positions).

It is not entirely clear what is going on here. There is no general hindrance to parenting. Mason points out that men who have early babies (babies had less than five years after Ph.D. completion) do better than all other groups, including single men and women.

The problem might be one of spousal selection or timing. Mason and Goulden point out that most women academics are married to people with advanced degrees and most academic men are not. If this results in men being more likely to have a stay-at-home spouse, and I suspect it does, then the differences in academic success might in part reflect this difference. The problem might also be one of timing, specifically having early versus late babies. Mason points out that women with late babies (babies that are not early) do as well as women without children.

The difference in academic success is accompanied by, and probably caused in part by, differences in productivity. Consider Mason’s findings on postdoctorates (Ph.D.s whose job is to do research in a professor’s lab). Married men with children put in 15% more time as a post-doctorate than married women with children and married men without children put in 29% more. Married women with children are 25% less likely to present at a conference in the last year than married men with children and 17% less likely than married men without children. There is some evidence, albeit weak evidence, that this pattern holds for publications. Philosophers Miriam Solomon and John Clarke report that in philosophy, women receive 25-33% of the Ph.D.s, have 21% of the philosophy positions and yet publish only 12.4% of the articles. Note these figures come from different but overlapping periods.

Women professors with children are not lazier than their counterparts. Rather, they put their efforts elsewhere. For example, Mason found that women with children do roughly 16 more hours in housework and caregiving than do men with children and 24 hours more than do men without children.

Proposed explanations for these differences in academic success have included discrimination and genetics. An important factor is probably different preferences. Women attach less weight to their careers and more to family. This can be seen in that male Ph.D.s are more likely to want to be professors at research positions than are women Ph.D.s (32% more likely for University of California Ph.D.s). The latter are also much more likely to shift their career goals away from academia. When they shift their career goals, women are more likely than men to cite as reasons children and spouses. Mason provides an example of this line of thinking is the following quote from a doctoral student, “I feel unwilling to sacrifice a healthy family life and satisfying personal life to succeed in academics, and thus industrial options have become more appealing.”

Women’s preferences are not wrong, bad, or irrational, just different. Professor Michael Argyle of Oxford University points out that marriage is one of the strongest factors to correlate with happiness and the correlation is stronger for women than men. In The Blank Slate, Harvard psychology professor Steven Pinker points out that on average mothers are more attached to their children than are fathers and that this is true in societies all over the world and in our mammal lineage. He notes that on average women are more attentive to their babies’ well-being and place a higher value on spending time with their children. In contrast, Pinker notes, men’s self-esteem is more closely tied to their status, salary, and wealth. He also points out these factors play a larger role in men’s attractiveness as a sexual and marriage partner than it does for women.

One response by some universities such as the prestigious University of California at Berkeley is to excuse parents with substantial caregiving duties for children from having to teach some of their classes. A second response has been to give these parents an extra year or two to earn tenure. Other universities have adopted similar policies. It is not clear, however, whether these policies are justified by either fairness or efficiency.

These policies transfer costs from one person to another. If at a state university, professor Jones get excused from teaching some of her classes, one of three things happens: other professors have to teach her classes, students are offered less classes (or larger classes), or taxpayers have to pay someone to teach a class when they have already paid someone to teach it. By analogy, imagine the Sunshine taxicab business has salaried drivers. If women drivers with babies get excused from their shifts, then consumers pay more money, Sunshine loses money, or other drivers have to work extra shifts. This cost transfer might be good policy, but it is not a requirement of fairness or justice. Furthermore, if the headwind is against people who have early babies and don’t have stay-at-home spouses, then this is arguably a cost of one’s choices.

The efficiency claim is harder to assess. The concern here is whether the policies have positive or negative effects on people other than the professors. Such policies might encourage some of the best and brightest women to go into academia. This is desirable. There might also be eugenics-type reasons that are relevant. On the other hand, if students and others benefit from more productive professors, then there is reason to be weary of these policies. Also, this policy provides an incentive against stay-at-home parenting and if this is bad policy, and I don’t know that it is, this is a reason to avoid it. The effects are speculative and in any case hard to balance against each other. In addition, it is not clear that there is one right answer for every academic sector.

In any case, the issue is an interesting one and one the academic world will have to consider.