Evolutionary psychology: gender “construction”

Whenever I post on evolutionary psychology, lurkers come out of the woodwork, enraged that I’ve criticized their own discipline and vociferously defending the crappy papers that I sometimes highlight.  Recently one (or more) of the lurkers asked me to provide an example of an evo-psycho paper that I considered good.   Well, that made me think a bit.  Many of the papers are tolerable rather than outright lousy, but in my opinion good ones are rare.  But I’ve just read one that I’d put in the “pretty good” category.  Some might not consider the subject to be evolutionary psychology, but I think it is, for it’s about whether behavioral differences between males and females have a genetic basis.  If they do, then one can begin to uncover their evolutionary roots.  If they don’t, then we need not engage in adaptive storytelling about why males are aggressive and females are nurturing.

It’s a paper on “gender construction” by two doctors, William Reiner and John Gearhart, and it appeared in 2004 in the New England Journal of Medicine.  The jaw-breaking title is “Discordant sexual identity in some genetic males with cloacal exstrophy assigned to female sex at birth”. (The pdf appears to be free.) I found the reference to this paper in the “gender” chapter of Steve Pinker’s The Blank Slate.

The topic is one that used to exercise academics, especially feminist ones:  is gender-specific behavior and identity specified at birth by biological factors like genes and hormones, or is gender “constructed” by the way a child is raised?  I well remember feminists saying that if you could somehow raise a male child as a girl, or vice versa, they would show behaviors and personalities opposite to that of their “normal gender.”  Or, if you could raise children in a “gender neutral” environment, then males and females would grow up not differing in gender-specific behaviors (i.e., girls would play with trucks, boys with dolls, and, when adult, males would not be more aggressive nor females more “nurturing.”)  As the authors put it very dryly in their paper,

The concept of neonatal sexual neutrality subsequently developed, emphasizing postnatal, nonhormonal influences.

Now perhaps most of my readers already reject the concept of “gender neutrality” at birth, a neutrality suborned by rearing practices.  We know that injecting testosterone into females, for example, increases some “male-typical” behaviors (like aggression) and physical traits.  It’s clear that hormones have an influence on behavior, and in precisely the way predicted by which sex has the highest testosterone/estrogen ratio and vice versa.  But what about at birth?  Are newborns already wired up to show sex-specific behavior?

Reiner and Gearhart tested this by doing psychological studies on children born with cloacal exstrophy, a severe condition in which (in males) the genitals are malformed (but not the testes), and there are defects of the bowels and bladder.  It used to be fatal, but now can often be fixed by surgical intervention. (Often, however, medical problems remain, like the need for a colostomy or  the later appearance of sciolosis).  If you have a strong stomach, click on this link to see what a newborn with the syndrome looks like (go halfway down the page).

Reiner and Gearhart’s idea was this: if gender is “constructed” by socialization at birth, newborns who are raised as members of the opposite sex from birth should show behaviors characteristic of their “socialized” sex rather than their biological sex.  Cloacal exstrophy gave them a chance to do this, because males born with the syndrome sometimes have their penises and testes removed, a vulva constructed instead, and are raised as girls.  If the “socialization” hypothesis is correct, these males should show female-typical behaviors when older; if the biological hypothesis is correct, they should lean towards male behaviors.

The authors had a sample of fourteen newborn males with cloacal exstrophy whose parents agreed to participate in the study.  The babies were surgically constructed to have female genitalia, and parents agreed to raise the boys as girls, never telling the children of their biological gender.  (Two other males with the syndrome were raised as males even though they had the surgery.)  Several of the parents were raising “normal” girls at the same time.

At ages ranging from 5 to 16, the female-raised males were given psychological tests that explored their interests in toys, dolls, and clothes, the time spent playing various games, athleticism, aggressive behavior, career and sexual interests, sex of friends, etc.  They were also asked to declare their gender. The parents were also given questionnaires on their child’s behavior and relationships with other children.

The upshot:  all 16 subjects, including those with female genitals raised as males, “revealed moderate-to-marked male-typical behaviors” compared to the scores of children raised according to their biological sex at birth. (The paper reports the scores for each child on a number of scales.)  As for the parents, here’s what the authors report:

The parents of all 14 subjects assigned to female sex stated that they had reared their child as a female. Twelve of these subjects have sisters: parents described equivalent child-rearing approaches and attitudes toward the subjects and their sisters. However, parents described a moderate-to-pronounced unfolding of male-typical behaviors and attitudes over time in these subjects — but not in their sisters. Parents reported that the subjects typically resisted attempts to encourage play with female-typical toys or with female playmates or to behave as parents thought typical girls might behave. These 14 subjects expressed difficulties fitting in with girls. All but one played primarily or exclusively with male-typical toys. Only one played with dolls; the others did so almost never or never. Only one ever played house. Each of the three exceptions represents a different subject. Parents noted substantial difficulty attempting to dress the subjects — but not their sisters — in clearly feminine attire after about four years of age.

And, tellingly, of the 14 subjects, four of them declared themselves as “males” even though they had female-type genitalia, had been raised as girls, and had never been told of their birth sex.  Four more were actually told of their birth sex by parents who abrogated the agreement, and all four of them declared themselves males.  At the last follow-up, two more of the children were “unclear” about their sex, and another one refused to discuss it.  (I believe, but am not sure, that the initial assessment of self-declared sex, and the children’s psychological tests, were performed before those four had been told that they were born male.)

At the end of the study, all eight of the male-declarers used male names and male restrooms, and all eight wanted surgical reconstruction of a penis.  The other six still living as females all reported difficulty fitting in with female peers, a result not seen at all in cases of genetic females with cloacal exstrophy).

The conclusion:  babies are born with brains already wired up in a gender-specific way. The authors theorize that this is due to pre-natal hormonal influences on the fetal brain that affect subsequent behavior. And although the study has weaknesses (see below), I agree with this conclusion.  The results show that, to a large extent, the roots of gender-specific behavior are biologically rather than socially based, and are present at birth.  This jibes with the experience of many of my friends who are parents, who report that despite their efforts to raise kids in gender-neutral ways, male toddlers go for trucks and females for dolls.  The fact that rearing genetic males as females does not much affect their behavior indicates that socialization, at least via parenthood, plays at best a small role in the development of behavior.

Now the study is not perfect, and I’m sure readers (especially the evolutionary psychologists) will tell me that this study is just as flawed as those I’ve criticized previously.  There is no explicit comparison with the psychology of genetic females with cloacal exstrophy reared as females (though the authors do cite their impressions), the psychological scores of the children were not compared statistically to those of “normal” males and females of the same age (though one can certainly do this given the data given by the authors), and four of the subjects were told that they were born as males (at ages 5, 7, 7, and 18 respectively).  But the psychological profiles of the children were taken before any were told of their birth sex (and most were not), as well as the self-declaration by four subjects that they were “males”, is pretty telling.

This is not my field, and there may have been followup work confirming or refuting these findings since they were published 7 years ago. (If you know of any, do post below; Pinker quotes some other studies that support these results.)  But if the results hold up, they show pretty clearly that gender-specific behavior has strong biological roots, and may not be much changed by how a child is raised.  In other words, even if you give your male toddlers dolls and your girls toy soldiers, they’re not going to become gender-neutral.

Why is this study better than many run-of-the-mill evolutionary psychology papers?  First, because the authors took advantage of developmental accidents to do a study that normally would be either prohibited or impossible: psychologists rearing babies in either a totally gender-neutral environment or in which they do not know their sex. Many evo-psycho hypotheses simply can’t be tested properly because they require manipulations of adults or infants that are considered unethical.

Second, the study comprises a simple test of a simple hypothesis, and does not involve telling tortuous adaptive “stories” to rationalize the results.

Finally, the results are pretty clean cut:  amazing, in fact.  Given the limitations of the study—the small sample size, the parents who abrogated their promises, and so on—the results still stand out clearly.  Whatever biological gender a child has at birth will condition its later behavior in a strong way.

______

Reiner, W. G. and J. P. Gearhart.  2004.  Discordant sexual identity in some genetic males with cloacal exstrophy assigned to female sex at birth. New Engl. J. Med. 350:333-341.

104 Comments

  1. Tulse
    Posted January 31, 2011 at 8:07 am | Permalink

    What’s the evolutionary bit?

    • JBlilie
      Posted January 31, 2011 at 9:10 am | Permalink

      I think: Genetic control.

      • Tulse
        Posted January 31, 2011 at 9:15 am | Permalink

        But where is the demonstration that such control is actually adaptive? This tends to be the central problem of evo-psych, which is to say that if some feature shows the least evidence of genetically control, it must be adaptive, even though we can’t be bothered to show the relevant fitness benefits (but hey, how about this cool story instead?).

        • Dan Gaston
          Posted January 31, 2011 at 9:44 am | Permalink

          Actually, and I think Jerry would agree with me here, one of the key problems of Evo-Psych is the “Just So” Storytelling tendencies and rampant adaptationism. Evolution is NOT just natural selection and adaptation.

        • Posted January 31, 2011 at 9:06 pm | Permalink

          Seconding Dan’s comment. Evolution DOES NOT equal adaptation. You can probably make a small [academic] fortune going into evol psych and ling and cultural evol with a proper understanding of evolutionary theory and actually apply, you know, all this other work that has been done in the field. Why adaptationism bothers us is because it is such a painfully impoverished view of evolution that you might as well not bother at all. Drift, mutation and [in some species] recombination are the other prominent ‘forces’ of evolution, and diversity and complexity have an intrinsic property to increase *despite* adaptation/selection through their own awesome mechanisms (eg. constructive neutral evolution).

          There’s a whole science out there, and very few have made it past the first lecture, sadly.
          /rant

          • JdL
            Posted February 1, 2011 at 11:28 am | Permalink

            The adaptationism of sociobiology & EP & evolutionary medicine & behavioral ecology, etc. -is METHODOLOGICAL adaptationism. (Look it up on Google scholar, see Godfrey-Smith 2001).

            Its validity as a research strategy has been defended by numerous evolutionary biologists who did make it past the first lecture (to put it mildly).

            See, for example, Ernst Mayr’s (1983): How to Carry Out An Adaptationist Program, or George C. Williams’ writings on evolutionary medicine.

            All I see here is a mindless paroting of Gould & Lewontin 1979.

            • Dan Gaston
              Posted February 1, 2011 at 12:17 pm | Permalink

              Not mindless parroting of Gould and Lewontin. But I come from a molecular evolution/molecular phylogenetics background. Neutrality is the Null Hypothesis (and an appropriate one for evolutionary scenarios). Selection is a hypothesis that must be rigorously supported by the data and not merely evoked as the causal explanation for a trait. Otherwise it is just story-telling, plain and simple.

              It seems to me this should be especially considered when talking about human behavior, much of which is likely a byproduct of selection on other cognitive functions.

              • JdL
                Posted February 1, 2011 at 2:28 pm | Permalink

                Thanks for that civil reply to my admittedly somewhat intemperate remarks (apologies).

                I do stand by my point, however. Yes, neutrality is the null hypothesis in your field. But the assumption that a trait X is an aspect of an adaptation (itself an adaptation, or a by-product of some specific sort) has proven heuristic value in many fields of biology (this is the point made by people like Mayr and Williams). Of course, the evidence has to support the resulting hypotheses (but that should go without saying).

                If you are suggesting that molecular genetic evidence (signatures of selection) is a necessary condition for a thorough testing of each and every adaptation hypothesis, then I cannot follow you but nor can the evolutionary biological literature.

              • JdL
                Posted February 1, 2011 at 2:35 pm | Permalink

                I should have made this explicit: Mayr 1983 considers the ‘alternatives to selection’ in G&L 1979 to be applications of the AP properly conceived.

                Williams’ views are similar. His advice in EM is to assume that a trait that makes us vulnerable to disease is an aspect of an adaptation (you will be familiar with the link between malaria and some hemoglobinopathies.In the case of EM, the methodological adaptationist is interested in explaining diseases, but will never argue that the disease itself is an adaptation.

              • Dan Gaston
                Posted February 1, 2011 at 4:08 pm | Permalink


                Williams’ views are similar. His advice in EM is to assume that a trait that makes us vulnerable to disease is an aspect of an adaptation (you will be familiar with the link between malaria and some hemoglobinopathies.In the case of EM, the methodological adaptationist is interested in explaining diseases, but will never argue that the disease itself is an adaptation.

                Well if anyone ever described the disease itself as an adaptation I would wonder if they understood evolution at all of course, but this clarification doesn’t address my point. While a Methodological Adaptationist is willing to concede that Selection may not be the best explanation for a trait, the trend is that many seek to explain any and all traits, a priori, as the product of selection as opposed to the product of evolution.

                In the case of many hemoglobinopathies we have good solid evidence for selection and can realistically measure fitness differences between heterozygotes and either of the two homozygous alternatives. That’s fine, I have no problem with that.

                But without such (or similar) evidence, a paradigm of seeking to explain everything as the product of selection first can lead in the wrong direction entirely.

                Stating that a trait is under selection is a hypothesis that needs to be tested, as rigorously as possible, in every case, by some appropriate methodology.

              • JdL
                Posted February 1, 2011 at 4:11 pm | Permalink

                A final remark. ‘Null hypothesis’ is a statistical- technical concept. It does not mean ‘most likely hypothesis’ or ‘default hypothesis’.

                As I understand it (but correct me if I’m wrong), the reason that neutral evolution is typically the null hypothesis in test of selection in population genetics is that neutrality expectations are relatively easy to define (so that statistically significant deviations from neutrality epectations are evidence of selection), NOT because neutral evolution is considered a priori more likely (even though it may be for randomly selected genes).

              • JdL
                Posted February 1, 2011 at 4:30 pm | Permalink

                A reply to your second reply:
                “the trend is that many seek to explain any and all traits, a priori, as the product of selection as opposed to the product of evolution.”

                This is an a priori *methodological* choice. The reason for doing this is that ‘aspect of adaptation’ hypotheses give rise to testable predictions in a way that neutrality does not *in the absence of knowledge of the genetic basis of a trait*. More often than not in biology the genetic basis (if any) of a trait is not known and so the neutrality tests you are familiar with cannot be used.

              • Dan Gaston
                Posted February 1, 2011 at 4:41 pm | Permalink


                A final remark. ‘Null hypothesis’ is a statistical- technical concept. It does not mean ‘most likely hypothesis’ or ‘default hypothesis’.

                As I understand it (but correct me if I’m wrong), the reason that neutral evolution is typically the null hypothesis in test of selection in population genetics is that neutrality expectations are relatively easy to define (so that statistically significant deviations from neutrality epectations are evidence of selection), NOT because neutral evolution is considered a priori more likely (even though it may be for randomly selected genes).

                Yes that’s true. I would point out that I didn’t imply that it is the most likely hypothesis or the default, it is merely the Null.


                This is an a priori *methodological* choice. The reason for doing this is that ‘aspect of adaptation’ hypotheses give rise to testable predictions in a way that neutrality does not *in the absence of knowledge of the genetic basis of a trait*. More often than not in biology the genetic basis (if any) of a trait is not known and so the neutrality tests you are familiar with cannot be used.

                Note that I am not saying “use molecular tests of selection” here, just use some “appropriate” test for selection, for traits where the genetic components are either too complex or unknown obviously this needs to be some sort of other test.

                In addition, if ‘aspects of adaptation’ give rise to testable hypotheses… then by testing those hypotheses you are doing exactly what I am advocating. The problem is the intellectual framework, and the mindset it tends to create, as opposed any sort of methodological problem.

              • Dan Gaston
                Posted February 2, 2011 at 8:28 am | Permalink


                If you are suggesting that molecular genetic evidence (signatures of selection) is a necessary condition for a thorough testing of each and every adaptation hypothesis, then I cannot follow you but nor can the evolutionary biological literature.

                As I said, I advocate an appropriate test of selection, not specifically those used in molecular evolution (although when appropriate obviously these should be used). Although I would weaken my remark somewhat and say instead that appropriate evidence for a trait having been selected for should be sought. Observed fitness effects for instance.

                The issue is that while a Methodological Adaptationist, in principle, searches for evidence of selection they really only consider non-adaptive explanations in the case of failure to find such evidence. In practice this leads to an over reliance on adaptation as the explanatory concept for evolutionary biology (it’s easier to explain) a prevalence for “just-so” stories. This latter drawback is something that afflicts EP quite heavily IMHO. This, in effect uses adaptation as a sort of null hypothesis.

                If, on the other hand, we come at the question from the other point of view, then what we are doing is looking for evidence of adaptation/selection and if we haven’t found it yet we have simply failed to reject neutrality. It also puts us in a starting position where we are really considering the plurality of possible evolutionary hypotheses. I think this mindset would be of great utility in all of evolutionary biology, it forces us to really consider what we want to see as good and persuasive evidence of adaptation.

            • Posted February 2, 2011 at 3:31 am | Permalink

              Yes, but plenty of well-stablished evolutionary biologists also employ a non-adaptationist framework quite successfully. Like, assuming a priori that things aren’t adaptations, and that if something can be explained perfectly fine through non-adaptive means, adaptation need not be evoked at all. Just because ‘most’ evolutionary biologists subscribe to a priori adaptationism does not make them right. It’s a tradition that may do for evolutionary ecologists, maybe, but does not work for the rest of us.

              And I can’t be mindlessly parroting Gould because I’ve barely read him (which I must fix eventually, I know). In fact, I’m not too interested in the philosophy of evolution, and will stand by my views until someone can point out how, specifically, they are contradicted by hard data. Until then, I’m right.

              Intradisciplinary wars aside, it would be nice if the dominating adaptationists at least let the public (and students) be aware of other mechanisms in evolutionary biology. While we may argue to death over their relative importance, there is absolutely no doubt that adaptation alone does not explain evolution. Thus, equating adaptation with evolution is not only philosophically questionable, but factually dead WRONG.

              • Dan Gaston
                Posted February 2, 2011 at 8:29 am | Permalink

                And let them be aware of it in more then just a cursory dismissive fashion.

    • J.J.E.
      Posted January 31, 2011 at 11:49 am | Permalink

      Like many systems in evolutionary research, the potential strength of this publication is likely to come from laying the groundwork for understanding a potentially important phenotype. While this study may not evaluate evolution of this trait, it does establish the role of genetics in the trait. Insofar as possible and ethical, the study is a natural experiment that mimics an experiment that would test the importance of genetics versus culture/socialization. If we take Jerry’s summary at face value, it appears that the biggest difference between the children in this study and “normal” girls is merely their karyotype, i.e. whether they are genetically male or genetically female. That is very important, for without a genetic difference underlying a phenotypic difference, differential survival has no lasting influence, and thus isn’t natural selection in the biological/Darwinian sense.

  2. Becca Stareyes
    Posted January 31, 2011 at 8:09 am | Permalink

    The question that comes to mind here is how much of this is from the kids having gender identities set at birth and then mimicking social indicators of gendered behavior observed. In other words, do the kids raised as girls but identifying as boys pick up ‘boy things’ from some hardwiring of what ‘boy things’ are or because they learn what things are ‘boy things’ and feel like ‘I am a boy, so I should act like one’? Or, more probably, to what extent does culture and to what extent biological factors play a role in setting gendered behavior once you establish that gender identity has a biological basis?

    Which I guess is follow-up work, and I’m not sure how to test it short of similar studies done in different cultures/environments. (Even then, it’s not perfect, as you can’t exactly set up a culture where kids see, for instance, an exact swap of male and female roles from American culture.)

    • theshortearedowl
      Posted January 31, 2011 at 10:12 pm | Permalink

      Very good point – it can still be the gender identity that is innate (or determined in utero) rather than the behaviour that accompanies it.

  3. MosesZD
    Posted January 31, 2011 at 8:13 am | Permalink

    Interesting. I tried to raise both my daughters as gender neutral as possible. Especially in light I was a stay-at-home dad and my wife, the scientist, was a full-time worker.

    And our affects are pretty much reversed from what you’d call ‘typical.’ I’m much more empathetic than my wife. I talk far more, I’m far more emotionally accessible, etc. I mean, basically, we have a fairly strong role-reversal.

    And both my girls are girly-girls.

    Which is nothing more than anecdote. But I know I’m not the only one in my social circle who has tried the ‘gender neutral’ parenting style and been completely sabotaged by the children’s marked preferences.

    So when I read the paper… Yeah, I get it. The womb, and our hormones, control far more than we’d like to believe.

    • Posted January 31, 2011 at 8:22 am | Permalink

      The “control” of nature doesn’t seem all that bad as long as the child is not pressured into roles that run contrary to her/his nature. You took pains to avoid those pressures.

      No doubt some boys prefer the dolls and some girls prefer the trucks. Just not usually.

      • Dan Gaston
        Posted January 31, 2011 at 9:47 am | Permalink

        Its one thing often overlooked by both sides of the Biological versus Social Construct argument, although one I think generally at least implicitly acknowledged by biologists generally. When we talk about differences between the sexes for various traits or preferences we are describing in effect statistically different distributions of those traits between the sexes. It is perfectly natural to have outliers of either gender who overlap with the “other” distribution.

        • Sven DiMilo
          Posted January 31, 2011 at 12:03 pm | Permalink

          A key point, and one not grasped by anybody who uses the phrase (gender essentialism”.

          • Dan Gaston
            Posted January 31, 2011 at 1:14 pm | Permalink

            I think it is also overlooked by those who see Gender as just a social construct as well. One group (mostly) ignores that we are dealing with populations and distributions, the other ignores the importance of biology and genetics to behavioral tendencies.

    • Donovan
      Posted January 31, 2011 at 9:48 am | Permalink

      I have no children, but often visit a friend who has a young son (2yo). He and his wife allow their son to pick his clothes, toys, movies, and so on. Gifts from friends and family tend to be “boy” items, but at 2, it seems unlikely it should sway him too much. He chooses Thomas the Train movies and clothes. He like trucks, cars, and train tracks. He does, however, have a very large doll house and one doll. Every time I see him play with the house, though, he just moves the walls and stairs around.

      It is anecdotal, but his parents (both PhDs) claim to have seriously tried to allow him to define his self.

      • Carl Troein
        Posted January 31, 2011 at 3:00 pm | Permalink

        I’m hoping that if we can pile enough anecdotes here, they’ll make data.

        As I recall, my father and his wife were a bit surprised when my (half-)brother wanted work gloves, shovels, trucks and such. His two older sisters were never into that sort of thing.

        My two daughters play with all sorts of toys, but put a real baby in the room and they’ll forget about everything else. It’s been like that since they were little more than babies themselves. For all the soft toys (and a doll or two) I had as a child, I never cared much for real babies.

  4. Bill Gilliland
    Posted January 31, 2011 at 8:15 am | Permalink

    This study should put a full stop to any doctors offering gender reassignment as an option to parents (either after cloacal exstrophy, or even more heartbreaking, a botched circumcision.) The more ethically sound route would seem to be: leave as much function intact as possible, and let the child decide for themselves when they are old enough to make an informed choice.

    • Sal Bro
      Posted January 31, 2011 at 9:02 am | Permalink

      I agree wholeheartedly. Sadly, many parents feel they’re protecting their children by assigning them a clear gender that is in agreement with their genitalia, or by simplifying the genitalia to more clearly conform to male/female standards. Research has not borne out the helpfulness of these approaches, though. Such surgeries should be severely regulated and restricted.

      • Posted January 31, 2011 at 1:48 pm | Permalink

        I agree, too, and I think it did lead to rethinking of the practice. I wish the authors had stuck to the identity aspect (and the psychological and emotional consequences for the kids) and left out the weak stuff about interests/behaviors.

      • Posted January 31, 2011 at 8:20 pm | Permalink

        Actually, it’s often not the parents who rush to have these surgeries performed on their intersex children, but the doctors who pressure the parents to do so when they are in a vulnerable state:
        http://www.fathermag.com/206/intersex/

  5. Sigmund
    Posted January 31, 2011 at 8:19 am | Permalink

    The results in that study sound rather like the famous Bruce/Brenda gender reassignment case.

    • Posted February 1, 2011 at 1:11 pm | Permalink

      This is an excellent book! Came here to suggest it if no one else had already.

  6. Alex
    Posted January 31, 2011 at 8:23 am | Permalink

    If people are interested in other studies on when the brain becomes gendered, this series of articles (I’ve linked to #4 of 4 because it has links to the other 3) on American Trans Man looks at gender identity in people with a couple of hormone conditions – namely, Complete Androgen Insensitivity in XY individuals and Congenital Adrenal Hyperplasia in XX individuals.

    • AR
      Posted January 31, 2011 at 9:35 am | Permalink

      Thanks for the link, Alex. The links in the CAIS article will be useful in my own research. :)

    • Marlene Zuk
      Posted February 1, 2011 at 8:59 pm | Permalink

      One of the books I mention in a later comment, Brainstorm: The Flaws in the Science of Sex Differences, by Rebecca Jordan-Young, also discusses the use of CAH and other genetic/hormonal disorders to understand gender. The analysis is very thorough and comes out quite skeptical about the conclusions. One issue is that people’s definitions of what is appropriate female and male behavior has changed a great deal, but no one has bothered to re-examine the studies that concluded, for example, that girls with CAH are “more masculine”. That term has meant a lot of different things over the decades.

  7. Posted January 31, 2011 at 8:34 am | Permalink

    I remember that part of Blank Slate–I agreed when I read and I still agree with it. I would be one of those parents (boy-boy-girl, 10, 8, and 7, respectively) who would say boys and girls are different *from birth*. They even breastfeed differently.

    • AR
      Posted January 31, 2011 at 9:37 am | Permalink

      “They even breastfeed differently.”

      Interesting. I’ve never heard this. (And, not being a mother even once over, never experienced it either!)

  8. JBlilie
    Posted January 31, 2011 at 9:08 am | Permalink

    This fits well with my wife’s and my observations of babies: They usually exhibit their basic “personality” right from the start. Crabby people are crabby as babies, etc.

    • Carl Troein
      Posted January 31, 2011 at 3:23 pm | Permalink

      It’s funny… my 3½-year-old is every bit as strong-willed, fearless and determined to do everything herself as when she was just a baby. On the other hand, her 2-year-old sister has in the last few months gone from generally contented to the more hellish dimensions of the terrible twos. I have no idea what she’ll be like when that’s all over, but I get a feeling that our parenting will affect the outcome.

  9. K.D. O'Brien
    Posted January 31, 2011 at 9:35 am | Permalink

    Okay, I know I’m one of the weird outliers, but I was one of those crazy tomboy girls that loved trucks, hated dresses, thought dolls were only cool with their heads popped off, and never felt like I fit in with other girls (unless they too were tomboys). And, if you think about it, if this kind of behavior was so abnormal, we wouldn’t have a name for it (tomboy). And I am fully female identified and am also heterosexual.

    Sorry, I think there is a small problem with this type of research. I think that we define gender too much as a dichotomy. Maybe there are more than 2? (Not talking biological sex here, just gender. Two different things.)

    • Dan Gaston
      Posted January 31, 2011 at 9:50 am | Permalink

      See my comment above. What we are really talking about are distributions of traits, not binary operators. The distinction will likely hold true most of the time but there will always be outliers (like tomboys) who overlap with the “other” gender’s distribution of traits for one or more traits.

    • S.K.Graham
      Posted January 31, 2011 at 11:44 am | Permalink

      K.D. I think the problem you refer to is one of societal prejudice, not a problem in this research. If anything, this research supports the idea that it is perfectly natural for you or any girl to be a “tomboy”. In other words a girl who is a tomboy gravitates to more “typically boyish” behaviors because, despite being female, she has something in her genetic makeup that give her these tendencies.

      This and related research teaches us that we should not try to force people into narrowly defined roles and behaviors based on a few superficial attributes, such as genitalia.

  10. Posted January 31, 2011 at 10:01 am | Permalink

    Huh?

    It’s a mess.

    First, sexual identity is not the same thing as behavior, and they shouldn’t be conflated. The issue Becca Stareyes raises above is valid. Moreover, which behaviors are associated with which gender is highly variable over time and across cultures. Many of the activities they discuss didn’t even exist until very recently. To try to assert an evolutionary basis for a sex difference in interest in reading or fantasizing about weddings doesn’t make much sense.

    Second, not only is it small-N but it’s not double blind. The parents knew. Their saying they raised these children as girls (again, as this is defined at this moment in our culture)* is about as convincing as doctors saying they knew who was in the placebo group but didn’t let on to the patients in any way. It’s why double blind studies are done in the first place. The authors acknowledge later on that they really can’t know that the parents raised them fully as girls despite their knowledge. Then they claim:

    That 12 subjects had genetically female sisters who did not have atypical sexual behaviors offers some evidence that parents did use female sex-of-rearing practices.

    But it doesn’t offer any evidence of what they did with regard to these children. (It also, funnily enough, undercuts their argument in that it implies that sex-of-rearing practices are indeed determining of behaviors.)

    Also, it’s worthwhile to point out, since it’s purportedly about identity and the behavior stuff is pretty much garbage, that five of the kids identified “unwaveringly” as females.

    *By the way, parents who claim that they raised their children gender-neutrally should realize that this is pretty much impossible even when they’re really trying and that they’re not the only agents of childhood socialization.

    • Kevin
      Posted January 31, 2011 at 10:13 am | Permalink

      That’s just pure unadulterated nonsense. How in the world would you go about “blinding” this? Raise the children as wolves? Not let the doctor see what surgery they were performing? Blindfold the parents during diapering and bathing?

      The doctors knew what gender they were “assigning” the children. The parents knew what gender the children were being assigned to. They were being raised as girls, because that’s what the parents and surgeons thought best.

      And yet — they were boys.

      I’m as big a fan of double-blind, placebo-controlled trials as anyone. For those issues where such a trial design is possible.

      This is not one of those areas. Seriously. Reconsider your critique.

      • Posted January 31, 2011 at 10:34 am | Permalink

        That’s just pure unadulterated nonsense. How in the world would you go about “blinding” this? Raise the children as wolves? Not let the doctor see what surgery they were performing? Blindfold the parents during diapering and bathing?

        You couldn’t, ethically. I’m not suggesting anyone does. My point is that in order to make the claim that how they were raised was controlled for it would have to be double blind. The parents can say they raised them fully as girls despite their knowledge, but it’s doubtful and we have only their assurance. The authors explicitly acknowledge that this is a problem.

        The doctors knew what gender they were “assigning” the children. The parents knew what gender the children were being assigned to. They were being raised as girls,

        Again, we have only the parents’ claims (which I think are honest, just doubtfully true) that their knowledge that these were genetically males did not affect the extent to which they raised them as girls.

        And yet — they were boys.

        Actually, five of them self-identified as girls, and others are unsure about their identity. But my critique isn’t about the identity portion in any case, but the behavioral portion.

        Seriously. Reconsider your critique.

        Seriously. Try to read my critique for comprehension.

        • Travis
          Posted January 31, 2011 at 12:40 pm | Permalink

          I’ll have to *agree* with your critique. A double blind study would be needed to get the real facts re: nature vs nurture…yet a double blind study in this case is ethically impossible. It’s bullshit “studies” like this that confound our ideas of gender in the first place. All I can be sure of after reading I’d that there’s

        • Travis
          Posted January 31, 2011 at 12:46 pm | Permalink

          *All that can be sure after reading is that there’s still a great deal to be unsure about.

        • S.K.Graham
          Posted January 31, 2011 at 6:06 pm | Permalink

          Since the authors provide full disclosure of their methods, I think to call the paper a “mess” is unfair. You might want to argue that their conclusions are too strong due to the absence of any double-blind controls and dependency on self-reporting.

          However, all of the subjects tested and demonstrated typical male-type behaviors. You mention 5 still identifying as females — but what would you expect? They have no knowledge of their birth condition, have female body parts, and have been told they are female all their lives. What is extraordinary is the 4 who were not told about their condition, have female parts, have been told they are female all their lives, and yet they insist that they are male. It is the sheer magnitude of the result which is convincing, despite the potential for bias due to lack of controls.

          It would be extraordinary for such strong results to come about as a result of bias, unless the bias itself was the result of extreme incompetence or fraud on the part of the researchers and/or deliberate deception on the part of *all* the parents — and why on earth would parents do such a thing (agree to the female genitalia then raise as boy?). Do you really think mere subconscious bias on the part of honest, competent researchers, could explain these results?

          Regarding your point about evolutionary basis for gender preferences for modern behaviors of reading or fantasizing about weddings. Or course these specific behaviors could not have been selected for. What would be expected is genetic predispositions which are then given specific culturally dependent forms. The wedding fantasy could (emphasis *could*) be the culturally shaped manifestation of a genetic predisposition — a predisposition more common among girls than boys.

          • whyevolutionistrue
            Posted January 31, 2011 at 6:28 pm | Permalink

            Good comment (by which I mean I agree with all of it!).

            • Posted January 31, 2011 at 8:46 pm | Permalink

              Oh, FFS.

              Since the authors provide full disclosure of their methods, I think to call the paper a “mess” is unfair.

              What? If you disclose your methods your paper can’t be a mess?

              You might want to argue that their conclusions are too strong due to the absence of any double-blind controls and dependency on self-reporting.

              I did argue that their claims about a biological basis for interests/behaviors were unfounded. Wasn’t about self-reporting. Read my critique.

              However, all of the subjects tested and demonstrated typical male-type behaviors.

              Read my critique.

              You mention 5 still identifying as females — but what would you expect?

              What part of identity is not behavior do you not understand?

              Regarding your point about evolutionary basis for gender preferences for modern behaviors of reading or fantasizing about weddings. Or course these specific behaviors could not have been selected for. What would be expected is genetic predispositions which are then given specific culturally dependent forms. The wedding fantasy could (emphasis *could*) be the culturally shaped manifestation of a genetic predisposition — a predisposition more common among girls than boys.

              Yes, the specific interests and behaviors associated with males and females in a specific culture at a specific historical moment could have an evolutionary basis, as could any others. The role of evo psych, in any culture at any moment, is ideological: to make up stories to explain how the culturally prevalent ones do. It’s garbage, and it’s a shame Dr. Coyne can’t read these studies with a fully critical eye.

              • Sven DiMilo
                Posted February 1, 2011 at 9:39 am | Permalink

                The role of evo psych, in any culture at any moment, is ideological: to make up stories to explain how the culturally prevalent ones do.

                This is precisely the presupposition that makes talking about the subject online so immensely frustrating. Just because everything you say is filtered through your ideology does not mean that it’s true of everybody else.
                Is there any possibility in your mind that the “role of evolutionary psychology” might actually be an honest attempt to apply knowledge of the evolution of other animals’ behavior to our own species in order to, you know, better scientifically understand our behavior?

                Oh but wait: intent is not magic, and only social effects matter. I forgot.

          • Posted January 31, 2011 at 9:10 pm | Permalink

            However, all of the subjects tested and demonstrated typical male-type behaviors.

            It’s like my words are invisible.

            • S.K.Graham
              Posted January 31, 2011 at 10:19 pm | Permalink

              “It’s like my words are invisible.”

              Perhaps you over-estimate the clarity of your words. Oh, wait….

              Yes, yes, “behavior is not identity” or whatever you said. Since neither I, nor Coyne, nor the paper’s authors, have equated behavior with identity, this part of your “critique” is utterly pointless.

              I rebutted the portion of your critique which, if valid, would have had merit. You ignored my actual rebuttal. Perhaps my language is too opaque?

              There are strong correlations among how people self-identify, how people identify others, genitalia, and how people behave. Under question here are the underlying causes (or lack thereof) of said correlations — genetics vs. cultural environment.

              Despite the methodological shortcomings (which could not be avoided), the paper provides good evidence in favor of a significant genetic component to the underlying causes of the aforementioned correlations. It is not as if we are talking about some sort of “small but statistically significant” non-null result. The result is so large, that even with small sample size and potential subconscious bias, it cannot be dismissed.

              • Posted January 31, 2011 at 10:54 pm | Permalink

                Yes, yes, “behavior is not identity” or whatever you said. Since neither I, nor Coyne, nor the paper’s authors, have equated behavior with identity, this part of your “critique” is utterly pointless.

                The paper’s authors and Coyne of course did. So did you when you responded to my critique with a comment about identity, which I pointed out in passing but which was not the subject of my critique.

                I rebutted the portion of your critique which, if valid, would have had merit.

                You rebutted nothing.

                There are strong correlations among how people self-identify, how people identify others, genitalia, and how people behave.

                You’re very confused about the claims the article is making re behaviors and my critique.

                Under question here are the underlying causes (or lack thereof) of said correlations — genetics vs. cultural environment.

                Under question here is an article that claims evolved sex differences in interests and behaviors, but the study on which these claims are based is a mess. (It isn’t a mess with regard to a biological basis for sex identity, though it perhaps overstates its case…or at least others who talk about it tend to.)

                Despite the methodological shortcomings (which could not be avoided),

                What can be avoided is making unfounded claims.

                It is not as if we are talking about some sort of “small but statistically significant” non-null result. The result is so large, that even with small sample size and potential subconscious bias, it cannot be dismissed.

                Again, you’re very confused. Why don’t you try reading and responding to each sentence of my original post.

      • winwar
        Posted January 31, 2011 at 12:52 pm | Permalink

        His critique is perfectly valid. This study is interesting but is hardly conclusive. And part of the reason is precisely what you state-an RCT isn’t going to happen. Instead we have an N=16 study single blinded study, which probably is only an N=12 study which may not even be single blinded. For pretty much anything else, we would call that “interesting, might be something there, but needs far more study”.

        • Carl Troein
          Posted January 31, 2011 at 4:47 pm | Permalink

          N=1 is acceptable if your null hypothesis is about non-existence. In this case, surely the outcome should be compared with the probability that half or so of 16 (or 12) subjects would view themselves as male if gender identity is not biologically determined.

          This, of course, is irrelevant to SC’s argument that the parents may have been biased into making the children identify as male. I’ll counter with this argument: if gender identity is _that_ sensitive to what the parents do, the prevalence of transsexualism would likely be much higher than one in a few thousand (or so).

          • Posted January 31, 2011 at 8:49 pm | Permalink

            This, of course, is irrelevant to SC’s argument that the parents may have been biased into making the children identify as male.

            Wasn’t my argument. My argument wasn’t about identity, but interests/behaviors.

            • Carl Troein
              Posted February 1, 2011 at 4:00 am | Permalink

              So it was. Very sloppy reading and/or brain-use on my part. :-/

  11. Posted January 31, 2011 at 10:06 am | Permalink

    As a semi-lurker who twice requested that you provide an example of research in evolutionary psychology that you consider good, I’d like to thank you for responding. I’d also like to clarify a bit: not all requests for you to post “good” research were coming from people defending the papers you attacked. In my case, I’m a layman who is interested in many aspects of evolution and who happens to respect your thinking a great deal; I asked you to share more of your thinking out of genuine curiosity and interest in your thoughts on the matter. Evolutionary psychology is uniquely confusing (to my mind, at least) and this blog is a great place to read good arguments and counter-arguments. Again, thanks for responding to reader requests, and keep the dialogue going!

    Also, I wonder if there are any studies
    regarding children born as biological females, but raised as boys. I can’t imagine any circumstances in which that would arise, but then, I’d never heard of cloacal exstrophy ’till now either.

  12. Posted January 31, 2011 at 10:12 am | Permalink

    Shorter me: The behavior part isn’t as facepalmingly stupid as claiming biological gender differences on the basis of observing whether female vervets play with cooking pots or showing 6-month-olds pictures of ovens, but it’s still bad.

    • Sven DiMilo
      Posted January 31, 2011 at 12:38 pm | Permalink

      It’s evidence, SC, part of an incremental amassing of evidence that gradually reveals robust patterns. You just can’t treat every study as if it’s supposed to be a knock-down smoking-gun proof. SImilarly, a problem you might have with the logic of the hypothesis being tested (putative bathwater) does not permit the trashing of the empirical data (teh baby).

      Oh, and it’s not just vervets… here is a corroboratory study in another monkey genus, and here is a very interesting-looking comment (which I have not yet read carefully) that at least attempts to address some of your questions about what toys are (empirically) “gendered” (for Old World primates generally, it’s increasingly looking like) and includes some plausible-looking speculation about why .

      • Posted January 31, 2011 at 1:03 pm | Permalink

        It’s evidence, SC, part of an incremental amassing of evidence that gradually reveals robust patterns.

        No, the only thing the vervet and infant-picture studies were evidence of is the stupidity of the researchers and the journals that published them. I am gobsmacked at your continuing failure to recognize that. And the comment you link to actually cites that ridiculous vervet study. Gah! I can already see numerous fatal problems with the studies referenced in that comment, but if you still can’t see why the vervet study isn’t evidence of anything at all, it’s probably pointless to try to explain them to you. Bad research, in evo psych or homeopathy, doesn’t corroborate other bad research – it just adds to the pile of bad research.

        • Sven DiMilo
          Posted January 31, 2011 at 1:14 pm | Permalink

          I’m just going to gently suggest one more time that you don’t get it.

        • Posted January 31, 2011 at 1:26 pm | Permalink

          Oh – the vervet study in question, for anyone interested, is this hilarious thing (I think the full text is available free):

          http://www.eurekalert.org/pub_releases/2002-12/tau-tca121002.php

          • Torbjörn Larsson, OM
            Posted February 1, 2011 at 1:07 am | Permalink

            Perhaps you could walk me through this paper, because I don’t get it.

            First, a seemingly necessary disclaimer. I have no political/not political gender axe to heave. I _am_ skeptic towards evo psych, mainly because its apparent lack of results which is under discussion here. (But also because of its preferred methods of story telling and greedy selection as theory instead of as research strategy, as recounted in comments above. And what Zuk says in a later comment – no double blinding is no double blinding, which should be blindingly obvious.)

            With all respects to the specific area and its experts, I also don’t subscribe to the “incremental amassing of evidence” thing, because a) it is quasi-induction; been there, done that, doesn’t work – but makes for good pseudoscience/religious “science explanation” b) a real effect has to show itself in physics or it tends to not be there.

            Now, a brief scan of the paper seems to show sex based differences in toy selection in vervets.

            So far so good, but a later paper on rhesus monkeys makes the valid (or not) criticism that a) it isn’t an actual selection experiment b) male sex vervets doesn’t display differences in time spent c) the sex differences is exactly opposite of those human babies displays. Their conclusion, which seems as overly pessimistic as the vervet paper was overly optimistic is that there simply isn’t any actual sex differences, only individual differences. (And of course that _their_ paper uses the proper methods and get good results. (O.o))

            To me there is a difference in the data and it is correlated with sex. But the aggregate of data over related species seems to indicate that whatever innate sexual differences there could be they don’t acquiesce to what the researchers tend to believe. In a way that is promising.

            But also it seems to me it could simply be not individual preferences but individual species differences, i.e. the usual contingency of evolution and not some strong selective effect. It would be odd if male and female mammal brains are wired differently as a result of sex hormones and differences doesn’t show up.

            • Torbjörn Larsson, OM
              Posted February 1, 2011 at 1:14 am | Permalink

              Also, I should add that despite it would be odd if actual physiological differences _didn’t_ show up, it could well happen.

              Doesn’t Coyne tend to speak of plasticism in biology, and how there is no purpose. It could well be that those differences are too weak/coincidental/compensated for. I’m remembering those newts that had 20 % individuals with differences from the basic bone structure in their feet and managed. At the same time not _all_ sexual differences are “too weak/coincidental/compensated for”, obviously. :-D

            • Posted February 1, 2011 at 4:40 am | Permalink

              Cooking pots. Vervets.

              Sex-segregated hunting. Vervets.

  13. marcel
    Posted January 31, 2011 at 11:07 am | Permalink

    Is there a “not” missing from this sentence?

    “This jibes with the experience of many of my friends who are parents, who report that despite their efforts to raise kids in gender-specific ways, male toddlers go for trucks and females for dolls.”

    • whyevolutionistrue
      Posted January 31, 2011 at 12:43 pm | Permalink

      Yes, good catch. What I meant was “raise kids in gender-NEUTRAL ways.” I’ve fixed it now–thanks!

  14. Marlene Zuk
    Posted January 31, 2011 at 6:26 pm | Permalink

    Please, before anyone moves another step, PLEASE read Cordelia Fine’s 2010 book Delusions of Gender: How Our Minds, Society, and Neurosexism Create Difference. She eviscerates a lot of the literature on “innate” sex differences, and has a brilliant chapter on how even the most avid attempts at gender-neutral parenting are a drop in the bucket in the face of societal progrmming. We simply can’t manufacture the appropriate controls, as some commenters have noted. We don’t mean to — it just happens. Mothers, for example, will describe the in utero movements of boys as more “vigorous” than those of girls, but only when they know the sex of the fetus. When they don’t, boys and girls are described the same.

    I just finished reviewing that book, as well as another somewhat more technical one, Brainstorm: the flaws in the science of sex differences, by Rebecca Jordan-Young, for the journal Quarterly Review of Biology. Highly recommended, both of them. Neither one denies the possibility of such “hard-wired” differences between the sexes, but they both point out how flawed the evidence used to support such claims has been.

    Another example that is often used by the “gender is hard-wired” crowd is congenital adrenal hyperplasia, CAH. This genetic disorder causes the fetus to be exposed to elevated levels of androgens during development, which presumably should masculinize females if the idea that early hormones shape gender-typical behavior is correct. And indeed, starting in the 1950s and continuing up to the present, girls with CAH have been scrutinized and their hormone levels measured. Early studies seemed to point unequivocally to a higher degree of “tomboyism” and other masculine traits in girls with CAH, including higher IQ.

    But as Jordan-Young points out, measuring toy preferences or teen sexual activity in people who were born with ambiguous genitalia, often subjected to repeated medical tests and surgeries during childhood, and raised by parents who understandably harbored anxiety about the gender role and appropriate behavior of their daughter, is not exactly a well-controlled experiment. So many confounding factors occur in addition to the altered hormone exposure that it is difficult to conclude that the elevated androgens themselves caused whatever differences between CAH girls or women and their counterparts exist.

    • Posted January 31, 2011 at 7:47 pm | Permalink

      Thank you! That’s exactly the kind of review I’d like to see (if I can get a hold of the books…).

    • Posted January 31, 2011 at 8:55 pm | Permalink

      Thanks! I’ll check them out.

      • Posted January 31, 2011 at 9:23 pm | Permalink

        Though I doubt many others here will. They’ll likely prefer to continue to offer their alleged gender-neutral parenting anecdotes and believe that vervets cook and 6-month-olds know what an oven is from a picture.

        Alas.

    • Sven DiMilo
      Posted February 1, 2011 at 9:31 am | Permalink

      Looking forward to reading this review.

    • AR
      Posted February 1, 2011 at 11:56 am | Permalink

      Awesome! Thanks for the recommendations.

  15. Posted January 31, 2011 at 8:36 pm | Permalink

    I was a show on FOX 11 called Mental Summer of 09 that a story where the girl was raised as a track star. In high school she had a boyfriend who she was very close too. She had a condition where every time she looked in the mirror she could not see her face, just a blank blur. Turns out that the girl was born a boy with a birth defection. I guess it was a mental issue dealing with the fact. Also the drama was incredible when her boyfriend found out that she was really a boy under all that make up.

  16. Tim Harris
    Posted February 1, 2011 at 5:58 am | Permalink

    And in the meanwhile what do we do, apart from engaging in intellectual arguments about scientific method, in the cases of children such as those out of whom Dr John Money made, and lost, his reputation? For he was regarded as a reputable scientist at one time, and that reputation caused a great deal of suffering. I realise that the issue has been raised above, but it is an important one.

  17. Posted February 1, 2011 at 7:27 am | Permalink

    The conclusion: babies are born with brains already wired up in a gender-specific way. The authors theorize that this is due to pre-natal hormonal influences on the fetal brain that affect subsequent behavior. And although the study has weaknesses (see below), I agree with this conclusion. The results show that, to a large extent, the roots of gender-specific behavior are biologically rather than socially based, and are present at birth. This jibes with the experience of many of my friends who are parents, who report that despite their efforts to raise kids in gender-neutral ways, male toddlers go for trucks and females for dolls. The fact that rearing genetic males as females does not much affect their behavior indicates that socialization, at least via parenthood, plays at best a small role in the development of behavior.

    This is the part that’s the problem, and it’s not a problem with the data itself but with your, and to a lesser extent the authors’, interpretation of the data. To say that this study shows that “to a large extent, the roots of gender-specific behavior are biologically rather than socially based” is to make a huge and unfounded leap. As Becca Stareyes points out above, the tendencies toward certain behaviors and activities could be the result of identifying as/with boys and gravitating towards those activities associated (in this particular cultural moment) with boys. All of the data, including toy choice, are consistent with this hypothesis, and there are good indications that this is in fact the case.

    First, they’re resistant to wearing girls’ clothing. This can’t indicate some evolved preference for certain clothing which just happens to be associated with males in this culture. It’s choosing clothing based on identification.

    Second, regarding the favorite activities (with some interesting differences between the parents’ reports and the children’s): How typical they are of boys and girls in general terms I don’t know (certainly the “typical female response,” whatever that is, wouldn’t be no interest in sports*), and they don’t say. I was struck by the references to sports that are almost exclusively male – baseball, football, hockey. It’s unusual for athletic people to lean toward sports in which they have few opportunities when they could choose others that are far more welcoming (tennis, soccer, etc.). (And it can’t be because these sports are rougher, since baseball, the one listed most, isn’t.) It’s hard not to see this as being drawn to the things associated with the sex you identify with/as.

    The other responses they provide are totally consistent with identity and don’t really say anything about behavior. It’s totally reasonable to conclude from this research that sexual identity has a significant biological basis. It’s improper to make the assertion on the basis of it that sex differences in behavior or interests are innate.

    *But would have been, say, 100 years ago, and would be in some cultures (just like the gendering of reading, art, biology, animals, cooking,… will change and flip), which is why this evolved male/female-typical idea is so dumb.

    • Posted February 1, 2011 at 9:12 am | Permalink

      How typical they are of boys and girls in general terms I don’t know

      The more I think about this, the more it appears that the very extreme findings (to the extent that these aren’t selected or exaggerated by the parents) suggest something very different from what the authors think. The picture we’re given is of these boys, raised as girls but whose innate boyish interests and behaviors prevailed over socialization.

      But they seem like extreme-stereotypical (in our culture) males.

      All but one played primarily or exclusively with male-typical toys. Only one played with dolls; the others did so almost never or never. Only one ever played house.

      They don’t feel comfortable playing with girls even at a young age. They’re not just all into sports, but, as I pointed out above, often into male-exclusive sports. But in a randomly-selected small group of boys raised as boys, you would expect to have a few/several who played with girls and “girls’” toys and who weren’t interested in sports at all. The more you look at it, the more it appears like kids feeling an identity that isn’t in accordance with their sex and asserting those behaviors and interests they (and their culture) see as fitting that identity.

    • S.K.Graham
      Posted February 1, 2011 at 1:27 pm | Permalink

      Your writing here is more transparent, so your words are less invisible… or something.

      “This is the part that’s the problem, and it’s not a problem with the data itself but with your, and to a lesser extent the authors’, interpretation of the data.”

      Your original critique had two main points, the second, which you labeled “second” was problems with the data (low N, not double blind, implying bias or potential for bias in the results). Now you seem to have withdrawn this point.

      What is left is your concern with conflating behavior and identity. For your critique on this point to have merit, you must be supposing that the 5 who identify as female yet still exhibit strong male-typical preferences and behaviors have identified as male subconsciously. Despite consciously believing (learned from their environment) that they are female, they must subconsciously identify as male and so emulate what they have learned are male-typical behaviors.

      So genes have the power to establish gender-identity in the face of environmental influences expressly teaching the opposite, and yet genes lack the power to generate gender-specific behaviors and preferences (or predispositions to behaviors and preferences), all of which must be learned from the cultural environment?

      The authors, nor Coyne, nor I have conflated behavior and identity. Nor have we claimed that culture does not play a significant part in determining what behaviors and preferences are associated with each gender. What I do (and I believe they do) is recognize the absurdity of the preceding paragraph. If genes play such a significant role in determining gender-identity, it is ridiculous to suppose that genes play an insignificant role in determining gender-typical behaviors preferences. No doubt, the cultural environment also plays a very significant role in shaping gender-specific behaviors and preferences.

      Is baseball, in all its details, a genetically determined gender-typical sports preference? Not likely. Throwing/running games in general, or competitive athletic competitions in general? Maybe. Is the male-bias for baseball and basketball as opposed to volleyball and tennis culturally determined? Most likely. How about football verses figure skating? If we ever have the data, my bet is on a strong genetic component to that preference independent of cultureal influence.

      Is the relationship between gender identity, gender behaviors, genes, and culture very complicated? Undoubtedly.

      Have I, Coyne, or the authors done more than speculate on what aspects of gender-typical behaviors and preferences are determined by genes vs. culture?

      Is it absurd to suppose that genes play an insignificant role in determining gender-typical “behaviors” (but not gender identity)?. Yes, given all that we know about genes, biology, animal and human behavior, that would be an extraordinary hypothesis requiring extraordinary proof.

      • whyevolutionistrue
        Posted February 2, 2011 at 10:29 am | Permalink

        Salty Current has asked me to post his/her comment as for some reason it didn’t make it to the site. Here it is:

        __________________________________
        Your original critique had two main points, the second, which you labeled “second” was problems with the data (low N, not double blind, implying bias or potential for bias in the results). Now you seem to have withdrawn this point.

        Of course I haven’t. I’ve mentioned it several times. I believe the parents’ knowledge that these were genetically boys influenced how they raised them and also the interests and behaviors they focused on and reported. We know that this happens generally, and there would be no reason to assume it’s not the case here. (I see some further evidence in the differences between the children’s reporting of their activities and that of the parents.) I haven’t withdrawn this point, and you haven’t rebutted it.

        What is left is your concern with conflating behavior and identity.

        The other point hasn’t been rebutted, and this is a significant “concern.” The data do not support the claims being made.

        For your critique on this point to have merit, you must be supposing that the 5 who identify as female yet still exhibit strong male-typical preferences and behaviors have identified as male subconsciously.

        Despite consciously believing (learned from their environment) that they are female, they must subconsciously identify as male and so emulate what they have learned are male-typical behaviors.

        Yes, I think this is what’s going on (and upon reading the article a second time I had more of a sense of it), in addition to the parental issues described above. Anyway, I haven’t looked closely enough to see which kids had which “scores,” but we’re talking about a handful of kids; a few people could have those interests and it wouldn’t mean anything about biology.

        So genes have the power to establish gender-identity in the face of environmental influences expressly teaching the opposite, and yet genes lack the power to generate gender-specific behaviors and preferences (or predispositions to behaviors and preferences), all of which must be learned from the cultural environment?

        Why not? Genes affect some things and not others. I’m not saying they must all be learned, but that there isn’t good evidence to the contrary. Just a lot of bad and misinterpreted research.

        The authors, nor Coyne, nor I have conflated behavior and identity.

        Of course you all have.

        Nor have we claimed that culture does not play a significant part in determining what behaviors and preferences are associated with each gender.

        It doesn’t make any difference what you’ve claimed about this, only what claims can and should be made on the basis of this study.

        What I do (and I believe they do) is recognize the absurdity of the preceding paragraph. If genes play such a significant role in determining gender-identity, it is ridiculous to suppose that genes play an insignificant role in determining gender-typical behaviors preferences.

        Of course it isn’t. They’re not coupled in the sense that because I enjoy male activities, even if I enjoy a lot of them, I think I’m male, feel any sense that I’m a boy or wish I was (in this sense – it’s of course possible for a girl to wish she were a boy at some point because males are privileged or out of curiosity). And the spatiotemporal variation in “gender-typical behaviors and preferences” (which have at each moment been assumed to be “natural”) should give anyone pause in making biological claims.

        No doubt, the cultural environment also plays a very significant role in shaping gender-specific behaviors and preferences.

        You haven’t established that biology does, so saying culture “also” does is stupid.

        Is baseball, in all its details, a genetically determined gender-typical sports preference? Not likely. Throwing/running games in general, or competitive athletic competitions in general? Maybe.

        And maybe not. We’ve seen huge changes in women’s involvement with competitive sports over the past several decades that have tracked changes in policy and opportunities, to the extent that the suggestion that interest in sports is somehow female-atypical doesn’t seem to make much sense. It would have made sense a century ago, though, which is a major part of my point. The behaviors associated with males vary extensively, and researchers and others arbitrarily assign them based on cultural prevalence and then make up stories for how that must have evolved. (I suspect that if these kids were in an intellectual or arts-oriented culture in which males were celebrated as scholars or painters, they and the parents would have asserted a strong interest in those things.)

        Is the male-bias for baseball and basketball as opposed to volleyball and tennis culturally determined? Most likely.

        Do you understand what I’m saying about the preference for baseball in this context? It’s widely acceptable for girls to play softball, and they have opportunities to play it if they enjoy sports, but these kids express a preference for baseball. It’s telling. As is the fact that sports (and especially virtually-male-exclusive sports regardless of their particular qualities) seem to be so predominant in the reports, which is probably not likely in a random sample of boys raised as boys.

        How about football verses figure skating?

        How about biology, animals, secretarial work? Re figure skating, from Wikipedia:

        The next step in the development of ice skating came in 1742, when the first ice skating association was formed, the Edinburgh Skating Club. The first instructional book concerning ice skating was published in London in 1772. The book, written by a British artillery lieutenant, Robert Jones, describes basic figure skating forms such as circles and figure eights. The book was written solely for men, as women did not normally ice skate in the late 18th century.

        Which I’m sure was considered to be based on biology at the time. Girls are at present largely excluded from football.

        If we ever have the data, my bet is on a strong genetic component to that preference independent of cultureal influence.

        But no one should care what your bet is, because such bets are bound to be culturally influenced and are based on nothing.

        Is the relationship between gender identity, gender behaviors, genes, and culture very complicated? Undoubtedly.

        Vapid.

        Have I, Coyne, or the authors done more than speculate on what aspects of gender-typical behaviors and preferences are determined by genes vs. culture?

        The authors claim that behavioral differences seen here are genetically based from the same sources as identity, such that certain interests and behaviors are in essence aspects of identity.

        Is it absurd to suppose that genes play an insignificant role in determining gender-typical “behaviors” (but not gender identity)?. Yes, given all that we know about genes, biology, animal and human behavior, that would be an extraordinary hypothesis requiring extraordinary proof.

        It most certainly would not. There’s a sample from one of the bpoks recommended above on Amazon. You should read it. You haven’t demonstrated you know anything about the subject, and the “research” on which claims are made ranges from extraordinarily weak to laughable (see: vervets). Moreover, what we know about variation suggests otherwise.

        • S.K.Graham
          Posted February 2, 2011 at 1:41 pm | Permalink

          You: “…and it’s not a problem with the data itself but…”

          Me: “…[your second point was problems with data]… Now you seem to have withdrawn this point.”

          You: “Of course I haven’t.”

          ’nuff said.

          Further point by point discussion here will be fruitless.

          • Posted February 2, 2011 at 8:04 pm | Permalink

            You: “…and it’s not a problem with the data itself but…”

            Me: “…[your second point was problems with data]… Now you seem to have withdrawn this point.”

            You: “Of course I haven’t.”

            ’nuff said.

            Further point by point discussion here will be fruitless.

            I said this particular point was related (primarily – OK?)to interpretations. I didn’t withdraw anything about any other problems. It’s all related, as I’ve noted and described, but I was in this post introducing another major aspect. In any event, you haven’t rebutted any of it.

            I’ve offered a detailed criticism and explanation of the interpretation argument, and repeated my earlier points about cultural variation and the problem it causes for the assumptions underlying this study and its analysis. I can list and explain the relationships among the various problems if you’re unable to follow blog comments. If you can’t address the many substantive points and prefer instead to make it about my presentation and writing, it’s pretty obvious why. Sad.

            • Posted February 2, 2011 at 8:23 pm | Permalink

              (Oh, and thanks to Jerry Coyne for posting my comment!)

            • S.K.Graham
              Posted February 3, 2011 at 12:20 am | Permalink

              SC, you repeatedly say that I have not rebutted you, simply ignoring my original rebuttal rather than offering a counter-rebuttal. Yet you complain that I and others have not taken the time to answer you point for point.

              Just for grins, I’ll have another go on a couple points that happen to stand out to my mind.

              Re: conflation of gender-ID and behavior. You ignore the fact the self-identification is, in and of itself, a behavior. What about obvious behaviors/preferences such as gender-typical sexual attraction for the opposite sex? Do you propose that this is not in large part genetically driven? And if not, exactly where is the line between gender-typed behaviors and preferences which are/are not genetically driven? Does this line somehow magically coincide with politically correct boundaries? And even if we all are sub-consciously wired to know “I am male” or “I am female”… how are we to know who is “like me” other than to observe behaviors? Appearance wouldn’t have worked for our ancestors who did not have mirrors and cameras. Matching genitalia might account for it, except this paper flies in the face of such an assumption. Maybe the subjects had a subconscious “I am male” identity… but how did they learn “who else is male” so as to emulate the typical behaviors of “the one’s like me”. Pheromones? Maybe, but then you are layering another unsupported assumption onto your subconscious gender-id hypothesis. And what about other species? Is it learned cultural norms that cause male bison to leave the herd as young adults to return at full maturity to battle for mating privileges? [repeat ad nauseum for countless species and behaviors] What magical assumptions must be made to explain that somehow human evolution resulted in a gender-agnostic genome when it comes to psychological traits (or a politically correct subset of psychological traits)? Occam’s razor is not a friend to your position.

              Oh, and unless you find flaw with their data-gathering methods (which is possible, as they do not go into enough detail for me to be absolutely confident they rule out observer bias), what exactly is so “laughable” about the vervet paper? If you accept the data, surely you do not suggest that gender aligned toy preferences of the vervets were learned from their culture. At a minimumm, the paper strongly suggests that there exists a significant genetic component to gender-aligned toy preferences among primates. By itself, I would agree that it is at best weakly suggestive that the *specific* gender-aligned toy preferences of modern human children in a modern cultural context are genetically determined. In other words, that vervet males have a genetic preference for trucks over dolls does not imply that the same bias among human children is genetically determined. But it *does* imply that genetically modulated biases for toy preference exist in primates, and it is a big stretch to suppose that humans are somehow an exception. As to what aspects of said preferences are influenced genetically verses culturally, remains a wide-open question. But if the vervet experiment were repeated across a range of primate species, with some preferences consistent among most species and consistent with human preferences, then that would indeed be strong evidence for specific genetic determination of specific human gender-aligned toy preferences.

              On to the other point I will address: male preference for football vs. figure skating. You bring up 18th century bias that figure skating was a male sport as if it is somehow damning to my thesis. But I was referring to the here-and-now. Within the current cultural context, there is a strong male preference for football over figure skating. Proposing a significant genetic component to this preference, as opposed to merely conforming to cultural norms post-gender-identification, does not rule out such learned cultural norms also being a factor. Furthermore, even if two centuries ago figure skating was “just for men” (as with pretty much *all* sports at the time), it would be interesting to look at what men’s *preferences* were at that time. I rather suspect that most men of the era *preferred* the more more aggressive sports which were precursors to games like modern football. My point here is that you bring in an irrelevancy, the time dependency of cultural norms for gender-typed preferences, when I have not disputed that cultural norms are a significant factor. If you could somehow magically eliminate the effects of culture, I would expect, on average, the male preference for football vs. figure skating to significantly diminish. But I would also expect a significant preference to remain which would reflect underlying genetic predispositions (most likely having to do with physically aggressive, directly adversarial competition).

              If we look at, say baseball vs. softball, in the magical zero-cultural-effect universe, I would expect little discernible genetic effect, as there is so little difference between the games… except for the following. The zero-cultural-affect universe is not absent of culture, merely absent of impact on gender-typed preferences. The cultural perceptions of the nature of the two games would remain intact. So, softball may retain a general perception of being a less competitive more “for fun” sport than baseball (event he names, “soft-” vs. “hard-” have this implication), and to the extent that this were true, I would expect a slight genetically driven preference for baseball among males.

              I raise the preceding discussion of sports preferences with attendant hypotheticals and thought experiments, not because I expect you or anyone to take them as “true” because I say so or because it is what I would bet on. The points is (1) to illustrate the irrelevancy of some of your counter-argument such as time-dependent cultural norms, as well as to illustrate some of the possible nuances I expect exist in the interaction of genetic and cultural factors on gender-typed preferences. Even in situations where culture shapes perceptions of somethings in a completely gender-neutral way, there could be gender-typed genetic predispositions which shape our response to those gender-neutral perceptions. And even though cultural norms play a huge role, there can still be gender-typed predispositions some of which reinforce those norms and some of which may go against those norms.

              • Posted February 3, 2011 at 12:19 pm | Permalink

                I’ll start with the vervets:

                Oh, and unless you find flaw with their data-gathering methods (which is possible,

                I have several (and I’m not going to get sucked into your nitpicking game about which issues involve design, data, methodology, analysis), but they’re not even worth noting here as the research itself is so profoundly stupid that specific data-gathering issues are minor in comparison. Even if they found strong sex differences in object “preference,” which they didn’t, all they could say is “We gave these vervets some random objects and seemed to find some sex differences in the objects they touched. We have no clue why.”

                as they do not go into enough detail for me to be absolutely confident they rule out observer bias), what exactly is so “laughable” about the vervet paper? If you accept the data, surely you do not suggest that gender aligned toy preferences of the vervets were learned from their culture. At a minimumm, the paper strongly suggests that there exists a significant genetic component to gender-aligned toy preferences among primates.

                I’ll try this one more time.

                1) They are vervets. If you are designing a study about sex differences in object preferences in vervets, it has to be based on what vervets do.

                a) Vervets do not cook. To call cooking pots female-oriented “toys” in a study of vervets and expect any sex differences in preference for cooking pots is ludicrous. All cooking pots are to vervets is objects with a certain shape, color, and texture. (It is possible to study sex differences in preferences for objects of different shapes, colors, and textures in different species, but 1) the expectations and study design should be based on the lives of that species, including sex differences in activities if they exist, and 2) if you’re testing for color, shape, texture, size, whatever, other factors have to be controlled for or your results are uninterpretable. Of course, they were not doing this.)

                b) Vervets do not hunt. They are, like most primates, herbivores, though they seem to eat insects and grubs. They’re little monkeys known for emitting a high-pitched squeal at the approach of a predator. There’s no hunting, much less sex-segregated hunting. To call certain objects male-oriented “toys” based on researchers’ notions of toys related to hunting, expecting male vervets to show a preference for them, is dumb. If you’re suggesting that a common ancestor of humans and vervets had cooking and sex-segregated hunting, show it.

                c) The only conceivable ones of these objects that could be expected to show gender preferences for females are those “representing” vervet infants, and establishing this is very difficult. It’s extremely questionable that a human doll would represent a vervet infant to a vervet, or that it could be distinguished from a stuffed dog, which they included as a neutral “toy.”

                d) To call this “play” or these “toys” for these animals is not really defensible. There really isn’t evidence that they “played” with these objects to any meaningful extent. The researchers recorded “touches,” which doesn’t make much sense if you’re arguing that they were drawn to different object for nurturing or throwing or rolling or whatever.

                2) Studies involving 6-month-old human infants looking at pictures of objects like ovens, cars, pots, footballs, or whatever are almost as painfully stupid. They fail to grasp that sex differences in preference for these objects could only be symbolic.* An infant this old has no clue what an oven is or how it’s different in use from a radio. If you think they’re not stupid, complete this sentence: “6-month-old infant boys would be expected to show a preference for a picture of a car over one of an oven without knowing what these objects are because they’ve evolved to __________ and these objects _______.” Vervets and human infants can of course engage in pre-symbolic play (though humans have already had massive cultural exposure, so impossible to rule that out; once you’re talking about symbolic play, of course, it’s obvious that they’re highly socialized), but then it’s the play (what are they doing with the objects?) that has to be studied. Again, you can do studies of sex differences in preferences (with the caveat that there’s been cultural exposure) for shapes, textures, colors,…, and those that have been done claiming to show them with whatever just-so explanation have sucked to the best of my knowledge.

                But if the vervet experiment were repeated across a range of primate species, with some preferences consistent among most species and consistent with human preferences, then that would indeed be strong evidence for specific genetic determination of specific human gender-aligned toy preferences.
                Do you understand now why this isn’t so?

                * And of course that these objects aren’t part of our culturally- and historically-varied evolutionary heritage until very recently.

              • Posted February 3, 2011 at 2:28 pm | Permalink

                SC, you repeatedly say that I have not rebutted you, simply ignoring my original rebuttal rather than offering a counter-rebuttal. Yet you complain that I and others have not taken the time to answer you point for point.

                I’m saying you haven’t rebutted my points.

                Re: conflation of gender-ID and behavior. You ignore the fact the self-identification is, in and of itself, a behavior.

                No, it isn’t. Stating/claiming it is a behavior, in the same sense that all expressions are behaviors.

                What about obvious behaviors/preferences such as gender-typical sexual attraction for the opposite sex?

                Sexual orientation is not a behavior.

                Do you propose that this is not in large part genetically driven? And if not, exactly where is the line between gender-typed behaviors and preferences which are/are not genetically driven?

                You think this is helping your case, but it’s hurting it. There is a close mapping, with few exceptions, historically and cross-culturally, of sex identity (the sense that you’re a male or female) to genetic sex. It also appears to be the case that the proportion of gay people has remained constant, though this isn’t my area of expertise and there’s a lot of complexity in how sexual orientation is defined. This suggests a strong genetic (and perhaps perinatal) component. In contrast, there’s enormous historical and cross-cultural variation in the interests/attitudes/behaviors/aptitudes associated with and expected of men and women and hetero/homosexuals, suggesting a weak, if any, genetic basis for these. Moreover, this latter variation doesn’t appear to affect the former – having certain interests doesn’t appear to make people change their sexual identity or orientation. This suggests that people sense that they are female, and then learn what that means in their culture.

                And even if we all are sub-consciously wired to know “I am male” or “I am female”… how are we to know who is “like me” other than to observe behaviors?

                Seriously? You don’t know what a male is? Did you have a father?

                And what about other species? Is it learned cultural norms that cause male bison to leave the herd as young adults to return at full maturity to battle for mating privileges? [repeat ad nauseum for countless species and behaviors] What magical assumptions must be made to explain that somehow human evolution resulted in a gender-agnostic genome when it comes to psychological traits (or a politically correct subset of psychological traits)? Occam’s razor is not a friend to your position.

                No, it’s not a friend to yours. Other species (for our purposes here) don’t have culture. They haven’t evolved in cultures, and sex differences in behavior don’t vary across “cultural” groups. Humans, in contrast, are cultural animals – we’ve evolved in cultures and we exhibit enormous cross-cultural variation. You’re suggesting that it’s best to suppose that things having to do with sex and gender are innate, but the better explanation is that things that vary widely historically and culturally are cultural.

                On to the other point I will address: male preference for football vs. figure skating. You bring up 18th century bias that figure skating was a male sport as if it is somehow damning to my thesis. But I was referring to the here-and-now. Within the current cultural context, there is a strong male preference for football over figure skating.

                But the problem with this is that you can’t just think in terms of a single culture in the present if you want to make arguments about genetic characteristics, which need to be transcultural.

                Proposing a significant genetic component to this preference, as opposed to merely conforming to cultural norms post-gender-identification, does not rule out such learned cultural norms also being a factor.

                Anyone can propose anything. That’s all people like you seem to do is propose arguments that support your culturally-ingrained notions. When you find research you think supports your presuppositions, you latch onto it.

                Furthermore, even if two centuries ago figure skating was “just for men” (as with pretty much *all* sports at the time),

                And what does that tell you? What do you think it told men at the time? Do you think they had their own version of eco psych to support their exclusion of women from the sport?

                it would be interesting to look at what men’s *preferences* were at that time.

                Of course anyone can. These were not outside culture.

                I rather suspect that most men of the era *preferred* the more more aggressive sports which were precursors to games like modern football.

                They may well have (or not – I don’t know; today, they might also prefer golf to figure skating). You can’t say anything about women’s preferences since they were excluded. It would be one moment in a single culture, in which preferences are shaped by encouragement, opportunities, constraints, and rewards. You could not use it to make a genetic/evolutionary argument. (You could, of course, make up stupid stories for why it’s “so,” and this is what people do. If men figure skate, it’s because they like competition; if women do, because it’s pretty. If men want to teach, it’s because they want to lead and be authorities; if women do, it’s because they’re nurturing. If men study biology, it’s because they’re curious and analytical about nature; if women do, it’s because they’re more attuned to nature and thinking about relationships. The proportions change and even flip, and the stories continue to be invented.)

                My point here is that you bring in an irrelevancy, the time dependency of cultural norms for gender-typed preferences, when I have not disputed that cultural norms are a significant factor.

                What don’t you understand? This cultural variation is a huge problem for biological claims. It means you can’t pluck one moment out of a single culture’s history and claim its evolutionary-biological roots. (Nor can you look only at one sex, implying that the other would not have the same broad preferences in changed circumstances.)

                If you could somehow magically eliminate the effects of culture, I would expect, on average, the male preference for football vs. figure skating to significantly diminish. But I would also expect a significant preference to remain which would reflect underlying genetic predispositions (most likely having to do with physically aggressive, directly adversarial competition).

                It doesn’t matter what you would expect or on what unfounded basis. You can’t make claims based on nothing. You can assert your expectations and suspicions all you want. They’ll always be worthless.

                If we look at, say baseball vs. softball, in the magical zero-cultural-effect universe, I would expect little discernible genetic effect, as there is so little difference between the games… except for the following. The zero-cultural-affect universe is not absent of culture, merely absent of impact on gender-typed preferences. The cultural perceptions of the nature of the two games would remain intact.

                What?

                So, softball may retain a general perception of being a less competitive more “for fun” sport than baseball (event he names, “soft-” vs. “hard-” have this implication),

                They’re different balls.

                and to the extent that this were true, I would expect a slight genetically driven preference for baseball among males.

                I do not care what you would expect. Got it? You’ve either missed my point about this study or ignored it completely. There are many sports these kids could have participated and competed in. Softball for kids is a competitive game, and they also play soccer, racquet sports, field hockey, basketball,…, – many of these are aggressive. These sports give competitive, athletic girls the opportunity for playing time and to compete. Baseball, football, and hockey don’t nearly as much. That these kids being raised as girls asserted a preference for sports in which they would have fewer competitive opportunities suggests that it was something other than an interest in (aggressive or other) competing (not that they didn’t have this, but plenty of girls do), but more in participating in “boy things.” (And note again my comment about this looking atypical for boys.) As I pointed out, baseball, which is the frequantly listed sport, is not aggressive or more competitive compared to many sports girls play. Even if your notion about perceptions re softball were true, it wouldn’t be evidence that evolution led them to choose baseball as an aggressive or competitive sport in which they could compete, because relative to other options it isn’t and they largely can’t. My explanation makes more sense.

                Even in situations where culture shapes perceptions of somethings in a completely gender-neutral way, there could be gender-typed genetic predispositions which shape our response to those gender-neutral perceptions. And even though cultural norms play a huge role, there can still be gender-typed predispositions some of which reinforce those norms and some of which may go against those norms.

                You just keep making assertions. Anything could be. It’s doubtful given the variation. Different notions of these genetic predispositions rise and fall with changing cultures, but all they are is speculation based on limited knowledge and a failure to recognize limitations.

  18. JdL
    Posted February 1, 2011 at 7:49 am | Permalink

    As one of the ‘lurkers’ who requested a good example of EP, I’d like to point out that I am not an evolutionary psychologist and EP is not my discipline. And like Rik (the other lurker responding above), I did not defend any of the papers you criticized. I just expressed my reservations about your approach to evaluating EP.

    That said, your response to the lurkers’ request is a bit odd.

    Nothing in the above post suggests that this is actually an evolutionary psychological paper. (A quick scan of the original article suggests the same). The paper provides evidence for a genetic basis for trait (complex) development, which is a precondition for evolutionary hypotheses. But that is all. Numerous traits have a genetic basis, but that in itself tells us nothing about why they evolved (byproduct, adaptation, …) – nothing about the ultimate questions that evolutionary pstychologists are interested in. The authors’ focus appears to be exclusively on the proximate level; no evolutionary explanations of gender differences are suggested.

    (Note also that development of a sex-differentiated trait through socialization is not, in principle, at odds with it being an adaptation since sex-differentiated socialization could have a genetic basis (Matt Ridley’s phrase ‘nature via nurture’ springs to mind). If genetically influenced socialization leads to faithful reproduction of adaptively relevant sex-differentiated traits, then NS could favor it.)

    • Sven DiMilo
      Posted February 1, 2011 at 9:46 am | Permalink

      You can test a hypothesis directly, or you can test it indirectly by examining its assumptions. One of the assumptions of any evolutionary hypothesis is a heritable genetic contribution to the phenotype of interest. Since this is perennially controversial for human behavior, it’s a valid approach.
      And it’s one study. Not the final word.

      • JdL
        Posted February 1, 2011 at 10:37 am | Permalink

        I agree of course with the points you make about assumption testing and this being only one study, but I think that qualifying this as an evolutionary psychological paper is a bit of a stretch nonetheless.

        No *specific* evolutionary hypotheses are being tested here, so a defining feature of papers in EP proper is missing. Indeed, the paper does not mention evolution at all, the authors are not evolutionary psychologists, and they do not relate their findings to the EP literature.

        I we accept your line of reasoning, then it would follow that ALL papers in behavioral genetics, for example, would count as evolutionary psychological. But that is simply not how these terms are used.

        ‘Provides evidence for genetic basis of trait X’ is simply not sufficient grounds for qualifying research as evolutionary.

      • gillt
        Posted February 1, 2011 at 1:00 pm | Permalink

        One of the assumptions of any evolutionary hypothesis is a heritable genetic contribution to the phenotype of interest. Since this is perennially controversial for human behavior, it’s a valid approach.

        I think you’re bringing in a media bias here. A genetic component for complex human behavior is not generally controversial. Sans defined haplotypes how can these studies be anything more than speculative?

    • whrrr
      Posted February 1, 2011 at 11:13 am | Permalink

      I think Jerry’s point in posting this was to give an example of a paper that simply attempts to find biological (evolutionary) roots for behavior. Most of the previous evo-psych examples have been hand-waving and/or extrapolations far beyond the data.

      • whrrr
        Posted February 1, 2011 at 11:14 am | Permalink

        This was a reply to JdL

        • JdL
          Posted February 1, 2011 at 11:39 am | Permalink

          That could have been his intention, but he still, in my opinion, misapplied the term EP (see my reply to Sven DiMilo).

          Since the whole point of this exercise was to cite some EP that Coyne does like, it would’ve been nice if he had cited a paper that would actually recognized as EP proper by evolutionary psychologists.

  19. Azkyroth
    Posted February 1, 2011 at 2:06 pm | Permalink

    Another weakness that comes immediately to mind is reliance on parental self-reporting for how the children were treated with regards to gender roles.

  20. Morgan
    Posted February 2, 2011 at 5:46 pm | Permalink

    The problem I have with studies like these is principally that the conclusion seems often to be stated in a way that implies that everyone’s ‘biological gender’ (assuming ‘biological gender’ means ‘chromosome combination’) has an equally strong influence on their behaviour and gender identity.

    Because while there clearly is a strong correlation between whether the child’s gender identity and whether they’re XX or XY, it’s not a simple relationship. Not all of the children in the study felt masculine to the same extent, and in the general population…well, explain the existence of trans people.

    And people raised to a gender in line with their biological sex who are more or less uncomfortable with their expected roles.
    As far as I know I’m genetically female, and I always got on better with boys and felt more or less uncomfortable around other girls, preferred trousers to skirts, and while I did play with dolls, I also thought my doll prams were perfect for chariot races which was much more fun than demurely pushing the dolls around, I then spent most of my school career in fuming resentment at the boys and PE teachers who wouldn’t let me play rugby and football, and boys who refused to fight me because they ‘didn’t hit girls’. When I was nine I decided I was going to be a boy, got my mum to cut my hair short and started demanding that people call me by a boy’s name. That didn’t last long, but purely because no-one took it seriously – which they might have done if I’d been in these kids’ situation.
    It sounds like those are the kinds of feelings the researchers are hearing from these kids. Obviously the fact that they’re getting it from ALL of them is significant, but only studying a group like this does often seem to lead to ignoring the degree of variation among the population at large. From me, and from a lot of cis girls, this kind of behaviour has only ever been seen as tomboyishness; from a genetic male it’s seen as their inherent maleness showing through.

    There’s also the conflation of gender identity and behaviours. Because the gender-appropriate behaviours always seem to me to be very restrictive, and not to take into account the number of happily cisgendered people who would spill over the boundaries of ‘gender-appropriate’ behaviours all the time. For a start, given how often athleticism is raised as a typically male attribute…well, so no woman on a sports team is conforming properly. And a lot of the stereotypical behaviours support each other – if you happen to be particularly active you’re probably also more likely to prefer masculine clothing because it’s usually more practical than skirts and dresses.

    I’m not denying that genetics have an influence on gender identity and gendered behaviours – but I don’t think sex chromosomes are the be all and end all of gender expression, which often tends to be how the conclusions from studies like this are framed.

    (To be honest, it makes me feel kind of like I’m not allowed to exist, with my non-binary gender identification and my masculine personality and my vagina and XX chromosomes. So sorry if this comment is rambly and not especially helpful, I’m basically just getting FEELINGS everywhere, ugh.)

    • S.K.Graham
      Posted February 2, 2011 at 9:53 pm | Permalink

      Morgan, I do not see where you take from this paper that the authors (or anyone else) are claiming that the presence or absence of a Y chromosome has an “equally strong” effect on behavior across all individuals. The paper provides evidence in favor of a strong genetic effect, period, not a uniform effect. It may be that science journalists and lay readers may allow a result like this to reinforce their biases regarding gender stereotypes. That is unfortunate.

      Anyone with a lick of comprehension of genetics *should* know that the genome is composed of thousands upon thousands of genes which interact with each other and the environment over the course of development in complex ways which we have barely scratched the surface on. While the presence/absence of a Y chromosome may act somewhat like a “master switch”, it clearly is not the whole story and there are countless other genetic switches which may be off/on in various combinations which vary from one individual to the next, resulting in plenty of atypical variation within genders. There is nothing about this paper to suggest that “tomboys” should not exist.

  21. S.K.Graham
    Posted February 3, 2011 at 7:12 pm | Permalink

    [in reply to Salty Current -- I'm avoiding the super-narrow columns]

    “[Vervet paper] …all they could say is “We gave these vervets some random objects and seemed to find some sex differences in the objects they touched. We have no clue why.”

    Uh. No. First, they can say more than that. Second they (I just went and re-read to be sure), they conclude much less than you seem to think they do. You are battling straw-women here.

    “1) They are vervets. If you are designing a study about sex differences in object preferences in vervets, it has to be based on what vervets do.”

    Which is what they did. They put the objects in the room with the vervets and watched what the vervets actually did. This is an utterly ridiculous thing to say.

    “…To call cooking pots female-oriented ‘toys’ in a study of vervets and expect any sex differences in preference for cooking pots is ludicrous.”

    This just shows your utter lack of comprehension of plain english, apparently. The “female-oriented” label in the Vervet paper is clearly a label of convenience based on human preferences in modern culture. Nowhere is it suggested otherwise. At no point did the authors state an expectation, and even if they had, it does not matter how ludicrous it is if the test is performed objectively.

    ” All cooking pots are to vervets is objects with a certain shape, color, and texture….”

    And the authors say as much. No one pretends that the vervets understand what a “pot” is or that the human use of “pots” for cooking is even relevant. A non-cultural preference for the pot, whether in vervet or human, may have to do with color, or curved shape, or most anything.

    ” (It is possible to study sex differences in preferences for objects of different shapes, colors, and textures in different species, but 1) the expectations and study design should be based on the lives of that species, including sex differences in activities if they exist,”

    This is a pathetic bias on your part. Why “should” such a study be based on the “lives of the species”? Are you so lacking in imagination that you cannot comprehend why a study using objects foreign to the species might produce interesting results?

    “and 2) if you’re testing for color, shape, texture, size, whatever, other factors have to be controlled for or your results are uninterpretable. Of course, they were not doing this.)”

    More bullshit. The results are interpretable for what they are. They were not testing for color, shape, textures, size, and they explicitly note that they cannot tell what specific properties of the objects account for the vervet preferences. They were primarily testing for the *existence* of biases, and whether the biases happened to coincide with biases of modern humans. The vervet paper authors find this *suggestive* of gentically modulated biases that are consistent across primate species.

    That’s it. That is all the paper concludes. The biases exist. The biases (for these 6 objects, in this experiment) coincide with human biases. And this is suggestive of primate-wide bias. Which is a suggestion for doing much more research before anything can be said conclusively.

    “b) Vervets do not hunt.”

    So what? Nothing in the vervet paper depends on them hunting, or not hunting. It is never asserted that they do. You seem to confuse some preliminary discussion of *possible* reasons for *human* gender-typed toy preferences as somehow being statements about the vervets. Again, a reading comprehension issue, perhaps?

    “…To call certain objects male-oriented “toys” based on researchers’ notions of toys related to hunting, expecting male vervets to show a preference for them, is dumb.”

    Again, the “male-oriented” label is based on extant modern human preferences, and is simply convenient. The authors expressed no expectations (whether or not they had such). The relationship to hunting was discussed *speculatively* as a reason for some human preferences, not vervet preferences.

    “If you’re suggesting that a common ancestor of humans and vervets had cooking and sex-segregated hunting, show it.”

    This is just silly.

    “The only conceivable ones of these objects that could be expected to show gender preferences for females are those “representing” vervet infants…”

    Really? That is the only conceivable possibility? A little biased in your preconceptions, perhaps?

    “…, and establishing this is very difficult.”

    The vervet paper makes no pretense of even attempting to establish this.

    ” It’s extremely questionable that a human doll would represent a vervet infant to a vervet, or that it could be distinguished from a stuffed dog, which they included as a neutral “toy.””

    Again, so what? The male/female/neutral labels were based on well-established human preferences, not vervet preferences. They are convenient labels, particularly as the data show vervet preferences happened to coincide with the human. Which *is* interesting. Hardly conclusive with such a small study, but it is interesting. As the authors said, it is *suggestive* of a pattern which might exist across primate species. I would be very interested to see the results of many more such tests involving many more different toys and primate species.

    “d) To call this “play” or these “toys” for these animals is not really defensible. There really isn’t evidence that they “played” with these objects to any meaningful extent. The researchers recorded “touches,” …”

    This is a particularly inane ‘critique’. They reported “approaches” and “contact” as something that can be measured objectively. The paper is not about the “play” of vervets per se, so one would not expect much discussion on the actual activity and whether it resembled “play”. One would expect such in a paper in which vervet behavior, per se, was the subject — perhaps such appear in their references. This research is intended to investigate genetic/evolutionary underpinnings of gender-typed primate behaviors and preferences. Measuring “touches” is a simple objective measure. A gender bias for “touches” is still a gender bias in behavior as relates to certain objects. Measuring preferences of human children is done by similar objective observations like “contact time” and how long an infant looks at an object. The point is that the behavior biases exist. Establishing the nature of what exactly consitutes “play” across species is not the point of the paper.

    “…which doesn’t make much sense if you’re arguing that they were drawn to different objects for nurturing or throwing or rolling or whatever.”

    Again. Reading comprehension. The authors never argue these things. These are speculations within discussion portions of the paper only, often in reference to other work, not conclusions. The authors are very clear that the experiment is not designed to examine *why* the preferences exist or what properties of the objects trigger the preferences.

    Me: “But if the vervet experiment were repeated across a range of primate species, with some preferences consistent among most species and consistent with human preferences, then that would indeed be strong evidence for specific genetic determination of specific human gender-aligned toy preferences.”

    You: “Do you understand now why this isn’t so?”

    Say what? So if in similar studies, chimps, gorillas, baboons, and a dozen different species of monkey all show a marked gender-aligned toy preference consistent with the gender-aligned preferences of modern human babies and toddlers, across dozens of different “toys”, you would find some twisted logic to dismiss the obvious conclusion? You really are, as Sven said earlier, blinded by your ideology. The fact that the “toys” are not natural to the primates would not detract — in fact it would support the conclusion. The whole point is that in all these closely related non-human species, there can be no cultural bias regarding gender orientation of modern human toys, so the there must be some kind of geneticly programmed cues to which they are respond. We would not yet know *why* the preferences exist (in the sense of evolutionary adaptions, etc.), but we would know that the preferences do exist and that they must have a genetic and evolutionary basis. We would know that, of course, the preferences would not be due to any modern human understanding of the purpose, function, symbolic significance of the toys. With enough toys and expiriments, *if* patterns emerged, we might start to identify features of toys that atracted the different genders — hard vs. soft? sharp edges vs. not? straight lines vs. curves? colors? It would of course take much more research to understand the adaptive significance of such gender biased preferences, but we do not have to know the adaptive value (if any exists or existed) to establish the existence (or lack thereof) of particular gender based preferences.

  22. Rob King
    Posted December 5, 2012 at 4:40 am | Permalink

    Its useful to know where all the evo psych folk have bneen going wrong now Jerry has supplied the details of what he considers good science. Small samples. No controls. And most importantly–no “adaptive story to rationalise findings”. Or, as we might put it, “no appreciation of adaptation to guide hypothesis formulation”. Still, it looks like he is gradually falling into the “you are doing it wrong” rather than his previous “you should not be doing it at all” position. Now, does anyone care enough about Jerry’s support to produce a range of experiments that suit his tastes?

  23. yboris
    Posted December 27, 2012 at 3:24 pm | Permalink

    Reblogged this on YBoris.

  24. random guy
    Posted December 30, 2012 at 8:54 pm | Permalink

    EP aside, I wonder if such situations are really strong evidence of anything at all.

    I think it has been shown that parents, unconsciously, treat their children differently according to gender, from the moment they’re born. Even if they say they’re all liberals and uber-pro gender-neutralism and feminism. There’s also a study where people evaluated the behaviors and expressions of a single individual baby as stereotypically manly or girly, depending on what was they were told the baby’s sex was. I think that the parents knowing the real sex could then create some sort of snowballing Pigmalion effect.

    I’m not saying that there’s nothing innate besides genitalia, that’s 100% construction. But I think it’s a “greyer”/mixed determination. Some tiny differences, that are not even always present (I think of two bell curves nearly total overlap), or present in the “correct” way, are reinforced and/or countered by socialization. So, basically, perhaps there’s some determination but enough room for social influence, so that the current distribution of men and women in various professions and activities isn’t simply a purely “biological reality” that couldn’t be significantly changed with different education/culture/environment. I don’t particularly have desires for some utopic genderless society that does not resemble anything those with traditional roles though. I just think that there’s some wasted potential for men and women, if they’re trained from an early age to reinforce their differences too much.

    I think that ideally such studies would have as subjects “normal” people and study how they vary in degrees according to biological determinants that were unlikely to have been affected by behavior/socialization. However that may be. Perhaps taking prenatal hormones in normal boys and girls and then looking how they fall in a spectrum, according to careful, preferably “blinded” evaluations. If the effect is strong, perhaps environment could be ruled out easily, if it’s not that clear perhaps one could study dichorionic preferably MZ twins that had different levels of prenatal hormones in a similar fashion.

    I think that’s even likely that such studies may have been done already, I’m curious about it.


7 Trackbacks/Pingbacks

  1. [...] This post was mentioned on Twitter by TheBritishAtheist, Jerry Coyne. Jerry Coyne said: Evolutionary psychology: gender "construction": http://wp.me/ppUXF-7bH [...]

  2. [...] weitere Studie, die für eine biologische Begründung der Geschlechterrollen spricht (via). Diesmal geht es um Kinder mit einer „Cloacal exstrophy„. Weil bei dieser Krankheit [...]

  3. [...] more here: Evolutionary psychology: gender “construction” « Why Evolution Is True Posted in Psychology Tags: crappy, crappy-papers, criticized-their, lurkers-come, [...]

  4. [...] original post here: Evolutionary psychology: gender “construction” « Why Evolution Is True Tagged with: all-eight • been-told • children • four-had • initial [...]

  5. [...] Evolutionary psychology: gender “construction” « W&#1211&#1091 Evolution I&#1109 Tr&#965&#1077 [...]

  6. [...] http://whyevolutionistrue.wordpress.com/2011/01/31/evolutionary-psychology-gender-construction/ [...]

  7. […] and utterly wrong.”  Jerry Coyne, who also spilled some ink over Douthat’s latest? I don’t think so!  The latest I’ve seen emanating from that realm, Joshua Greene’s Moral Tribes, […]

Post a Comment

Required fields are marked *
*
*

Follow

Get every new post delivered to your Inbox.

Join 25,540 other followers

%d bloggers like this: