In his SciAm post addendum (scroll to the bottom), Jesse Bering has been very gracious. This post really isn’t about that now-infamous advice column, but about broader ways to interrogate claims people make.

This post is another way of thinking about Sci and my #scio12 session on “Sex, gender and controversy” (see our other session posts here and here). When do we use evidence? When do we interrogate claims? When should we rile people up and when do we calm them down? Maybe unpacking the good and the bad from our follow-up posts (because Sci has an excellent one up as well) will provide more fodder for conversation Thursday.

As Scicurious has done in her own post, I am pasting Deep Thinking Hebephile’s original letter in the beginning as a reference:

“I am a non-practicing heterosexual hebephile—and I think most men are—and find living in this society particularly difficult given puritanical, feminist, and parental forces against the normal male sex drive. If sex is generally good for both the body and the brain, then how is a teen having sex with an adult (versus another teen) bad for their mind? I feel like the psychological arguments surrounding the present age of consent laws need to be challenged. My focus is on consensual activity being considered always harmful in the first place. Since the legal notions of consent are based on findings from the soft sciences, shouldn’t we be a little more careful about ruining an adult life in these cases?

—Deep-thinking Hebephile”

“— and I think most men are —”

Deep Thinking Hebephile (DTH) makes the point that “most men are” heterosexual hebephiles. But is this consistent with what we know are the most common preferences and actions of heterosexual men? Further, is this behavior within the range of natural sexual preferences, or is it pathological?

Let’s first be clear on definitions: hebephilia is the sexual preference for pubescent children. Not teenagers, but pubescent children. In industrial and post-industrial populations, that means a sexual preference for ten to twelve year olds, in agricultural populations eleven to fourteen year olds, and in forager populations maybe closer to thirteen to fifteen year olds. Scicurious has already alerted us to studies that negate DTH’s claim that most men are hebephiles. Others that assess sexual preference through a measurable penile response (though in some ways a problematic assessment) show not only high agreement between men’s stated preference for pubescent children and their response to images of them, but that these men are quite small in number (Blanchard et al. 2009).

“…the normal male sex drive.”

Embedded within this idea that “most men” practice hebephilia is the assumption that it is a part of the “normal male sex drive.” If DTH is contending that hebephilia is the normal male sex drive, that implies he thinks it is the natural state for men to prefer pubescent girls.

For something to be naturally occurring, it does not have to be practiced by everybody in a population, so the earlier evidence that hebephilia is uncommon doesn’t necessarily negate this next claim. But for that behavior to continue in some frequency in future generations, it needs to be an evolutionarily stable strategy. So let’s go over the main conditions that would convince me hebephilia is an evolutionarily stable strategy:

  1. Hebephilia is an adaptation: sexual preferences would have to be variable and heritable, and hebephilia itself would need to promote reproductive success under plausible conditions. Further, we would need evidence selection is acting on the sexual preference for pubescent children, rather than a correlated response to a different trait.
  2. Hebephilia is at least equivalent and ideally resistant to alternative reproductive strategies – it needs to be successful enough to beat out most other strategies.

Condition 1: Is hebephilia is naturally selected?

I think the claim could be made that sexual preferences are both variable and heritable – these are the first two conditions necessary for a trait to be naturally selected. For instance, despite very poor support for a “gay gene,” there is strong support that sexual preferences are both heritable and influenced by environment (Jeremy Yoder’s SciAm Guest Blog post on the topic explains this very well). So I am okay with extending this claim to hebephilia, that it very well may be part of sexual preference variation and that it may be heritable. Just keep in mind that hebephilia being part of the range of variation of sexual preferences doesn’t necessarily keep it within the range of normal, appropriate, healthy or socially acceptable.

The third part of natural selection – that the trait must promote reproductive success relative to other strategies – is where the claim breaks down. DTH’s first point, that he thinks most men are heterosexual hebephiles, suggests it is an evolutionarily stable strategy that results in enough reproductive success to continue to succeed among other existing strategies (like, say, a sexual preference for adult women). Perhaps hebephilia couldn’t beat out a preference for adult women (though I am being very generous here, since in a way this is exactly what DTH is trying to argue), but can it at least beat out the other sexual preferences?

I’ve talked about this before, but girls just past menarche (that’s her first period) are usually what’s called “subfecund” – this means that fewer of her cycles, when she does cycle, are ovulatory, compared to an adult woman. In fact, the most consistent ovulatory cycles and highest hormone concentrations are found in women 25-35 years old, shattering the myth that younger women are actually the most fertile (Ellison et al. 1993).*

So it is harder to get pregnant if you are just past menarche. And since the definition of hebephilia is attraction to a pubescent child, this includes attraction to and sex with girls who haven’t necessarily even had a period yet – girls who are completely infertile. If you are going to bet your reproductive success on one partner age, pubescent girls are probably the wrong one.

The second issue is that very young teen pregnancies have pretty negative health outcomes. Girls who give birth over sixteen or seventeen don’t experience any more negative birth outcomes than those over eighteen, but girls under fourteen – which, again, fits the hebephilia preference for pubescence – have increased risk of maternal and infant mortality (Kramer 2008). Further, the higher rates of infant mortality in those girls who first give birth under fourteen years of age can therefore expect a lower number of surviving children out of the total number they bear (Kramer 2008, Figure 1).

Now remember, these data come not from the “puritanical, feminist” American culture, but from a rural, traditional Mayan culture. But these data do support those found in industrialized populations found in a simple Google Scholar search. This search revealed a wealth of data showing that very young girls having babies doesn’t happen much, and when it happens it doesn’t often end well (Chen et al. 2007, Duenhoelter et al. 1975, Felice et al. 1981, Fraser et al. 1995, Haiek and Lederman 1989, Merritt et al. 1980, Olausson et al. 1999).

Condition 2: Is hebephilia resistant to alternative strategies?

For hebephilia to be an evolutionarily stable strategy, it needs to beat out other strategies. Already we are in danger of this strategy losing out because young teen pregnancies are far less successful than older teen and adult pregnancies. But let’s put a few more nails in the coffin.

We know of past and current cultures where older men marry pubescent, even pre-pubescent girls. However, hebephilia is defined as an adult who wants to have sex with pubescent children. This is not the same as an adult man who wants to marry a twelve year old girl and not have sex with her until she is older, for the purpose of securing a dowry or piece of land or better relationship with her family. That is common throughout human history. We cannot use as justification the few marriages in the Middle Ages (or partnerships in traditional forager societies) that happen to involve pubescent girls, because they rarely, if ever, involve sex with the child until she is older.

The example I know best is from the classic !Kung ethnography Nisa by Marjorie Shostak (1983). In this book, Nisa narrates how she is forced to marry an older man before she hits menarche. She runs away to her family several times, and her family is very permissive of this behavior. After a while, they demand she grow up and live with him. And she more or less does. Nisa eventually gets her first period and the menstrual celebration commences. It is only after this point that she is pressured to have sex with her husband. And eventually, she does.

Sex with pubescent girls appears to be highly infrequent. In Kramer’s paper on birth outcomes in teen pregnancies in traditional Mayan population, she reviews this exactly literature (2008). She finds:

“In natural fertility populations, the lapse between menarche and exposure to conception is highly variable, and may last one to two years up to over a decade (Whiting et al., 1986; Schlegel, 1995)” (Kramer 2008: 346).

And yes, she’s being nice about it, but by “exposure to conception” Kramer is talking about straight sexual activity. So in traditional, natural fertility populations (where natural fertility generally means non-contracepting) girls tend not to have sex, on the lower end of the spectrum, until a few years after menarche. That is post-pubertal, which means non-hebephilic.

Finally, there are definitely strategies that beat out hebephilia. There are two main mating strategies to secure a high chance for reproductive success if you’re male: to control the fertility of a female starting early, or to find a female who already has demonstrable reproductive success – a mother. Our closest primate relatives generally choose the latter: male chimpanzees don’t salivate over adolescent female chimps, and in fact reject them as sexual partners quite frequently. Instead, male primates and other animals fight over sex with the older females who’ve already borne a kid or two (Anderson 1986, Muller et al. 2006, Nichols et al. 2010, Proctor et al. 2011, Robbins 1999).

In humans we see plenty of individuals choose between either strategy and both can be quite successful. The former strategy is not unlike the one Nisa’s husband employed: marry the girl he wants, but don’t actually have sex with her until she comes of age. Of course, DTH will be sad to know that many consider this strategy to control female fertility part of a suite of behaviors that helps us understand the evolution of patriarchy (Smuts 1995).

At the end of the day, neither condition is supported. Hebephilia is not a direct product of natural selection, nor is it a successful strategy compared to other existing ones. DTH cannot get the satisfaction and validation he so desperately wants, because no matter how much he wants to justify it to himself, it cannot be justified in the context of the scientific evidence. Even if somehow this evidence were overturned by a wealth of opposing data, hebephilia is still not a permissible behavior, and it’s important to remember to make the distinction between what we can observe within human behavior, and what is right.


I would like to thank Scicurious and Charles Roseman for their comments on an earlier draft of this post.


*Subfecund is still fecund, and age-based probabilities are still probabilities. Don’t let these data fool any individual into ever thinking unprotected straight sex when a woman is postmenarcheal and premenopausal has few or no babymaking strings attached!


Anderson, C. 1986. Female age: Male preference and reproductive success in primates. International Journal of Primatology 7:305-326.

Blanchard, R., A. D. Lykins, D. Wherrett, M. E. Kuban, J. M. Cantor, T. Blak, R. Dickey and P. E. Klassen. 2009. Pedophilia, hebephilia, and the DSM-V. Archives of Sexual Behavior 38:335-350.

Chen, X.-K., S. W. Wen, N. Fleming, K. Demissie, G. G. Rhoads and M. Walker. 2007. Teenage pregnancy and adverse birth outcomes: a large population based retrospective cohort study. International Journal of Epidemiology 36:368-373.

Duenhoelter, J. H., J. M. Jimenez and G. BAUMANN. 1975. Pregnancy performance of patients under fifteen years of age. Obstetrics & Gynecology 46:49.

Ellison, P. T., C. Panter-Brick, S. F. Lipson and M. T. O'Rourke. 1993. The ecological context of human ovarian function. Human Reproduction 8:2248-2258.

Felice, M. E., J. L. Granados, I. G. Ances, R. Hebel, L. M. Roeder and F. P. Heald. 1981. The young pregnant teenager*,**:: Impact of comprehensive prenatal care. Journal of Adolescent Health Care 1:193-197.

Fraser, A. M., J. E. Brockert and R. H. Ward. 1995. Association of young maternal age with adverse reproductive outcomes. New England Journal of Medicine 332:1113-1118.

Haiek, L. and S. A. Lederman. 1989. The relationship between maternal weight for height and term birth weight in teens and adult women. Journal of Adolescent Health Care 10:16-22.

Kramer, K. L. 2008. Early sexual maturity among Pume foragers of Venezuela: Fitness implications of teen motherhood. American Journal of Physical Anthropology 136:338-350.

Merritt, T. A., R. A. Lawrence and R. L. Naeye. 1980. The infants of adolescent mothers. Pediatric annals 9:100.

Muller, M. N., M. E. Thompson and R. W. Wrangham. 2006. Male chimpanzees prefer mating with old females. Current Biology 16:2234-2238.

Nichols, H. J., W. Amos, M. A. Cant, M. B. V. Bell and S. J. Hodge. 2010. Top males gain high reproductive success by guarding more successful females in a cooperatively breeding mongoose. Animal Behaviour 80:649-657.

Olausson, P. O., S. Cnattingius and B. Haglund. 1999. Teenage pregnancies and risk of late fetal death and infant mortality. BJOG: An International Journal of Obstetrics & Gynaecology 106:116-121.

Proctor, D. P., S. P. Lambeth, S. J. Schapiro and S. F. Brosnan. 2011. Male chimpanzees' grooming rates vary by female age, parity, and fertility status. American Journal of Primatology

Robbins, M. M. 1999. Male mating patterns in wild multimale mountain gorilla groups. Animal Behaviour 57:1013-1020.

Smuts, B. 1995. The Evolutionary Origins of Patriarchy. Human Nature-an Interdisciplinary Biosocial Perspective 6:1-32.