You’ve probably heard about the newly reported research in which a genetic link was found to homosexuality in a study of gay brothers. Kelly Servick has a good writeup on it here. The study looked at 409 pairs of gay brothers, and found a region on the X chromosome that was similar across the sample. This sort of shotgun approach, comparing a trait (in this case, gayness) with a bunch of DNA (I oversimplify) is very likely to get results that look real but are the result of random association. But, it is also possible to find real links. I am agnostic as to whether or not this study found something interesting. But I do have a few remarks to make about how you get to be gay.

Consider the following list of things:

<li>Sexual attraction (to whom you are attracted)</li>

<li>Erotic response (what is erotic, including physically, to you)</li>

<li>Attachment (with whom to you seek attachment, and of what kind)</li>

<li>Sex drive (do you have it and where is it driving too?)</li>

<li>Society norms (especially for your subset of society)</li>

<li>The details of social norms, i.e., what categories of sexual orientation exist around you.</li>

<li>Your relationship to social norms (your comfort level ... do you seek "normalcy" or prefer something else?)</li>

<li>Whom you know or encounter and where they are with all of the above things.</li>

<li>And many more things that ultimately may relate to sexual orientation.</li>

This list can be written in many different ways, and every item on this list really represents a number of other sub items. These things are not mutually exclusive and the list is not exhaustive of that which relates to sexual orientation. Feel free to provide your own lists in the comments, if you like.

Many, most, maybe all of these things have individual ontogenies for any individual. The ontogenies may start before birth. We are bathed (or not) in various maternal hormones in utero. We are bathed in our own hormones in utero. The effects the hormones have depend on the relationship between the amount of hormone and the abundance and distribution of receptor sites, and on the timing. The abundance and distribution of receptor sites itself is probably influenced by the process. It is very complicated. Differences between one individual and another may related to external or non-genetic factors. In fact that may be very common.

Hormonal effects and interactions continue after birth. Again, timing, relationships between kinds and relative amounts of hormones, and receptor sites, still apply. Causes may be numerous.

The above only applies to that related to hormonal changes, which may affect a number of somatic (body related) features including brain features.

Then there are the non-hormonal factors, including cultural and social ones. Again there are complexities to the ontogeny of an individual with respect to these factors. And, these complexities are dynamic; culture and society can change right underneath you. And the non hormonal and hormonal factors may interact.

Much of this can be thought of as a process of negotiation. One negotiates internally, one negotiates with one’s social groups, one negotiates with society, culture, even the law.

Here is a simplified model linking the DNA identified in this study to homosexuality. Various switches are turned on or off, buttons pressed or not, during a person’s development. They do everything in some individuals to “make a person be gay.” But there is one element missing. If you have the DNA profile associated with the sample of 409 brothers, you get to be gay. If not, you probably won’t be. But, the “yes-no” value (reminder: oversimplifying here) found in this DNA actually has another purpose. It has to do with how many hairs you have on the back of your hand. The variation across men in hand hair is accounted for by variation in these genes. But in some individuals (but not all) it also happens to be the final ontogenetic link in the chain to a particular sexual orientation that in the sociocultural context that the 409 pairs of men live in results in gayosity. In another society, another culture, at another time, it results in being more likely to be a blacksmith than a farmer.

Note: That was a made up example. But in the absence of a biologically, developmentally, sensible link between some DNA and a trait, we can certainly carry out amusing and instructive thought experiments.

This complexity of links between causes and effects is probably true for the vast majority of variation found in human behavioral traits. Not this exactly, but something like this. The steps involved can be characterized in a certain way with respect to a trait under study, but all or most of those steps actually relate as well to other things. Also, some of those steps might have multiple causes. A particular manifestation of sexual or erotic attachment may arise in one person for one reason, in a different person for a different reason. In other words, the list I provide above can take many forms, not just because I’m being vague about what is in the list. The list can simply be different for different people who end up with the same “trait” as we happen to define the trait for the moment.

There is a reason for this vague connection, or in many cases, lack of connection, between inherited genes and behavior. A strong link between genetics and behavior has been shown to be very highly adaptive in some organisms. Here’s an old example. In a particular species of fruit fly, the larvae have a gene with two alleles. One allele causes the larvae to forage tightly in space, making a lot of turns in its search for food. The other allele causes the larvae to forage widely, to make few turns, and cover a larger area. Each allele is adaptive in a particular context and the fruit fly species has diversity at this locus. So, the fruit fly female mates with multiple males, produces a diverse batch of offspring, and the ones with a particular pattern of alleles at that locus have higher fitness. For now. In a different environment, maybe a few generations later (as the orange juice they are feeding on changes its characteristics as it rots in that glass you left on your desk) the genetic arrangement with the higher fitness changes.

But, humans are different. Humans are like the fruit fly, needing different traits at different times, but instead of those traits being programmed by genes, they are learned. Added on to the individual by enculturation.

This applies to some extent to all mammals because mammals have brains that matter to behavior. It applies very much so to primates, especially apes, and even more to humans. We have diversity in behavior, but we get it from our cultures. We learn to be a functioning adult; it is not pre-programmed. There probably are some pre-programmed behavioral features, but those are the features that would generally apply. But even those may be largely divorced from genetic inheritance on the grounds that behavior generally does not emerge from genes coding for neural structures. Genes in humans can’t code for neural structures at the level of the cerebrum, because of the way cerebrum develops, and that is where most of the relevant behaviors exist.

We can be pretty sure this is the case because of the huge cost we pay for it. Childhood. Childhood may be the most important human adaptation, and it may be the most costly. Human females can die in childbirth. That is nearly unheard of among mammals, outside of humans and our domestic stock. The babies can die in childbirth as well. That is because of our oversized brainy heads. Human babies are born helpless and spend several years nearly killing themselves at an alarmingly high frequency, and only survive childhood because of the adult humans taking care of them (or in some cases, wolves or ocelots, I suppose). This is costly to the adults. It limits reproductive output in the adults. Childhood also limits the reproductive output of the child, because it extend the time before reproduction, and decreases the chance of survival until reproduction.

Childhood, a brain that learns, the heavy reliance on the things the brain learns, and the long time it takes to make all this work demands a brain that is not overly programmed genetically, and results in a species with an extraordinary characteristic found in no other species: we are a multitude.

If you look at numerous species in most mammal families, you will find a wide range of behavioral and ecological repertoire. Measure body size, sexual dimorphism, typical system of mating, food getting, diet, defense, inter and intra species competition, etc. across all of the geomyids or voles, across all the species of dogs or all the species of cats, across the antelopes, across the African forest monkeys, etc. and you’ll find many features such as those mentioned that vary very little within species, but vary greatly across them within that taxonomic group.

Then look at humans. They look more like a taxonomic family than a species. Human cultures vary in these and other features as greatly as larger mammalian taxonomic groups.

But, when you capture an infant at birth from one human group and have it raised by another group, the infant grows up with behaviors typical of the adoptive group, not its natal group. That pretty much falsifies the idea that variation in our behavior is linked to variation in our genes.

By the way, if you move new born antelope, rodents, primates, etc. between species you may get some of the same effect. Cross species adoption does result in a bit of a behavioral chimera sometimes. But, it is only possible between some species and tends to work when the interactive parts of the system happen to be aligned. A parent bird will feed mouth-gaping carp for a while if they’ve lost their mouth-gaping baby birds. Within mammals, we’d expect a fair amount of post adoptive learning across species, because, as I noted above, learning how to be typical member of your species applies to some degree to mammals in general, more so to primates, more so to apes, and vastly more so to humans. Vastly.

Imma let you get back to finding links between genes and behavior. But first, remember, culture rules.

Final note. Part of the reaction to this new research, and this has happened with all prior research on homosexuality, is in reference to the sociopolitical outcome. If you are born gay, Conservatives can’t legislate against you, but if it is a choice, you might be a criminal. That sort of thing. This is balderdash. The Nazi’s killed all those people because of their genes. Many value free choice. Some will see being born gay as being born broken. People who are born a certain way, in many sociopolitical contexts, are vilified for it. You can’t win the sociopolitical game by claiming a certain human behavior or trait is built in or choice. You win that game on its own terms. And, lately, we mostly are winning.

Prior to birth there are a number of factors than can influence things like gender or sexuality in a human. You have probably heard of the finger-index (not the index-finger … actually called the 2D:4D ratio). The ratio of length of two of your fingers seems to be associated with certain trends; Men with a certain ratio tend to be more athletic or more gay, for instance. The mechanism for the finger ratio variation is probably straight forward but has not been nailed down yet. Likely, a surge of steroid hormones enhances growth rate of whatever bones are forming at that time (I simplify somewhat) and if such a surge occurs at a certain time, a slight shift in bone length ratio affecting fingers occurs and because of the timing, a slight change in something else also occurs, something having to do with what will eventually be adult behavior.¹ I am not arguing here for the strength of this association or its meaning, but available evidence shows that there is something going on. To the extent that this particular relationship is true, we see an adult outcome (related to gender, sexuality, or other behavior) being the result of something that is biological and prenatal, but not likely genetic. While the overall pattern of the hormonal environment of a fetus may be broadly determined by genes, variations in the details are just as likely determined by other things. In many contexts, one steroid hormone looks a lot like another, or can convert into another as they float around in the blood supply, so any large surge of steroids could act like sex hormones or growth hormones even if they are merely stress hormones, and there is an exchange of hormones between the mother’s blood supply and that of the fetus. Since the mother’s hormonal environment is heavily influenced by her environment (especially stress hormones), the ultimate cause of steroid hormone-mediated developmental variations in a human is very likely to be strongly environmental, if not entirely environmental, even though it all happens before birth.

Then there is the stuff that happens after birth. One study carried out in Australia seems to show that adults in a specific culture (Australian middle class) treated infants very differently depending on their knowledge of the infant’s sex.² For instance, a boy would be moved around more, tossed about a bit, handed boy-specific toys, and so on, while a girl would be held more calmly, not tossed about, hugged more, and handed girl-specific toys. In that study, the “sex” of the infant (boy vs. girl) was “known” to the adult on the basis of obvious clothing choices and pronoun use, and in fact, the infant was always a boy. After months of treatment as one sex or the other, depending on what that treatment consisted of, one could potentially get a gendered difference. Movement, touch, voice, etc. all form part of the environment in which the infant’s neural system, including the infant’s brain, develops. This would make a difference.

And so it goes throughout development; At numerous stages along the way, a human is affected by hormones, bathed in gendered behavior, and eventually, starts to observe her or his own environment and act accordingly. One study seemed to show that at about Kindergarten age, boys were more conscious of how they would fit into a group than girls, paying special attention to what other boys were doing before making certain choices. If this was a general pattern in a particular group of people, one might see girls engage in a wider range of available stereotypes while boys restricted themselves to a narrower range.

While it is possible that there is some hidden Jungian subconscious difference between nominal boys and girls resulting in different thematic tropes in their behavior (i.e., girls like circles and boys like lines or some such thing), the degree to which kids past a certain age … say six or so … gravitate towards gender specific toys or other objects, or engage in gender specific behaviors, is way too finely tuned to be the product of anything other than high cognitive function. While we know that across cultures, different colors are associated with different genders, within a culture most boys and girls know what the boy vs. girl colors are and to varying degrees express this knowledge as strong preferences, perhaps with boys expressing a narrower range of preferences than girls. Most likely, culturally specific gender preferences for things like toys and clothing are learned early, become deeply ingrained, are unlikely to be genetically determined at any level of detail (if at all) but may be attended to by boys more than girls.

There are many factors that would determine a person’s gender over a lifetime. The above mentioned inter-uterine hormonal conditioning is probably fairly complex, with multiple moments in time when one or another thing might happen, and where one version of the developmental scenario would lead towards one gender orientation than another. After birth, more of the same but less hormonal and more cultural, and later on, with puberty, the hormones kicking in again, but with a twist: Early conditioning may determine the nature of later hormonal activity by setting up differences in receptor sites or sensitivity, or other aspects of hormone feedback systems.

In speaking of humans it is easy to assume that other animals, who lack the complex and often costly (and therefore presumably ‘important’ in some way) trappings of prolonged development and culture have simpler systems for determining gender. For the most part, I would argue that rodents do in fact have simpler systems of gender than do humans, with the caveat that I’ve just compared an entire order of mammals (and a rather speciose and diverse one at that) with a single species. But what would you make of a gender-shaping system in rodents that was actually very complex, in which ‘culture’ was the main determinant of adult male-ness?

In rats, males get to be males in large part because they have testes that secrete testosterone, which in turn causes other changes. But according to at least one study, the degree to which testes will secrete testosterone is determined by anogentital licking behavior of the mother. This behavior is, in turn, brought on by some sort of cue produced by the newborn male. Without this licking, the testes do not produce much testosterone and andorgenization of the rat does not take place.³

OK, so I was exaggerating slightly when I said that rat “culture” determines adult gender, but prior to hearing this you probably assumed that there was a gene or set of genes that simply coded for which sex the rat would be when it grew up. And yes, you can get some interesting results when the mother rat is replaced with a lab tech and various different variations of the licking thing are tried out. (Using tiny wet paintbrushes.)

And I could go on. But I want to make two points about development and behavior, especially gender. One is that whatever genetic component is working, most aspects of adult behavior and orientation are shaped by non-genetic factors and those genetic factors that may exist come in the form of basic species-specific (but almost certainly gender-differentiated) “drives.” I’ve discussed the importance of drives here, and if you want to read a whole book about the link between drives and everything you do in your life check out Mean Genes: From Sex to Money to Food: Taming Our Primal Instincts (See all Self-Help Books)

The second point is that as something complex (and both personal and social) as gender orientation emerges in a person it must be true that it comes to whatever point it comes to after a series of many turning points. If every single factor is thought of as a simple binary choice (and I use the word “choice” with no reference to human decision making) between two cannalized options, then the number of possible outcomes could be thought of as 2ⁿ where ‘n’ is the number of times a binary choice is encountered. So, if there are, say, three hormonal moments in utero, and one more after birth (puberty) and, say, three life stages that have major influences on gender (and I oversimplify) then the number of possible routes a person may take from conception to adulthood would be 2⁷. That is 128. If these different paths lead to mostly different outcomes, wouldn’t there be over 100 “genders” among humans?

The interesting thing about this is that a cursory examination of potential human gender diversity from a purely biological point of view suggests that there are at least dozens of “genders” but the vast majority of cultures define (or even allow) only a few. Perhaps culture, in this case, is more restrictive than biology. Which, to a behavioral biologist, is not much of a shock, though it might be if considered from a broader social science perspective.

So, the next time you are in charge of making a form to collect personal information from people, when you are designing the “gender” question, you might consider something other than a couple of checkboxes. Perhaps a drop-down list. Or, best of all, just have people write a short essay. Make ’em think, that will.

¹Be careful with this idea: While I’m sure there are several aspects of 2D:4D research that are valid and interesting, it is often somewhat over-reported. Also, the numbers are tricky. The measurement is often done on fleshed and living fingers, but should really be done on the bones directly (using X-ray technology, not sacrificing the subject and defleshing them!). And the meaning of this trait is somewhat open to interpretation. I’d be comfortable sorting out males from females in a skeletal population with good preservation of hands but no pelvic remains, but more reluctant to use this for sorting out ethnic groups, gender orientations, or assertiveness levels. For a recent review see Bailey and Hurd, 2005. Finger length ratio (2D:4D) correlates with physical aggression in men but not in women. Biological Psychology. Volume 68, Issue 3, March 2005, Pages 215-222.)

²The specific research to which I refer was shown on a documentary about sex differences; For an exemplar published study on this work see Frisch 1977. Sex Stereotypes and Adult-Infant Play. Society for Research in Child Development. Vol. 48, No. 4 (Dec., 1977), pp. 1671-1675

³See this study and references therein: Moore and Morelli, 1979. Mother rats interact differently with male amd female offspring. Journal of Comparative and Physiological Psychology, Vol 93(4), Aug 1979, 677-684. doi: 10.1037/h0077599.

Moore, C., & Morelli, G. (1979). Mother rats interact differently with male amd female offspring. Journal of Comparative and Physiological Psychology, 93 (4), 677-684 DOI: 10.1037/h0077599

The answers are Yes, argued by Stephen Cici and Wendy Williams of the Dept of Human Development at Cornell, and No, argued by Steven Rose, a neuroscientist at Open University.

I would like to weigh in.

The real answer, as is so often the case, is “You dumbass, what kind of question is that? Think about it further and rephrase the question!”

But I don’t think they are going to do that.

I find it very interesting that even though the question does not mention IQ across gender, the details of the ‘debate’ (disguised as ‘rules’) actually specify that the commentators will tackle both race and gender links. Kinda proving that Nature is indeed playing the race card.

I like the idea of addressing both the questions of gender and race in relation to any differences (IQ or whatever). The course that I have taught in many forms in the past, and will likely teach again next Spring, does this. I like to do this because of the very important difference of differences. Gender is, biologically, much much more “real” than race. Gender is demonstrably real (in many aspects) and race is demonstrably not real (in almost all aspects). Also, almost all race differences we see bandied about are linked to nefarious racism one way or another. Gender differences, however, run the full spectrum from really destructive to very positive, with a lot of difficult ambiguity in the in between parts. So, looking at the myriad of purported gender differences first, then race second, turns out to be very very interesting. (One could do it the other way round as well, but for various reasons this works better in the context of my class.)

Let me say a few things about each of these papers first (citations below), then I would like to make a few broader remarks about gender, race, and “IQ.”

Steven Rose does a very good job of explaining all the reasons why the answer to this particular question should be “No” … although I hope he would also agree with me that this is not exactly the question that should be asked. He rightly discusses motivation, noting that we are busy comparing certain “races” by IQ while utterly ignoring equally oft constructed multichotomies of difference.

The categories judged relevant to the study of group differences are clearly unstable, dependent on social, cultural and political context. No one, to my knowledge, is arguing for research on group differences in intelligence between north and south Welsh (although there are well-established average genetic differences between people living in the two regions). This calls into question the motivation behind looking for such specific group differences in intelligence, sheds doubt on whether such research is well-founded, and begs whether answers could possibly be put to good use.

He does not spend enough time on, but does address, the fundamental flaw of the question: If race is not a valid categorization of people, then how do we justify funding scientific research of it? He also notes that while people may bellyache about adjusting IQ scores across ‘racial’ groups, no one seems to complain about nor notice the adjustment of IQ scores between gender, whereby boy’s scores are raised to make them seem equal to girls. Who are smarter, obviously.

The other side of the coin argued by Cici and Williams is the usual drek that should not pass for scientific discourse. Race should be studied because … it is truth. Race should be studied because Stalin tried to stop this kind of thing. Race should be studied because … Larry Summers and James Watson and others have been victimized by the Liberal Left.

Whatever whatever.

I would like to note that the “yes” side is being argued by geneticists. That is pretty typical. Geneticists don’t study intelligence, they study genes and they overrate the value of knowledge of genetics and always have. The “no” side is argued by a neurbiologist. Neurobiologists understand things like culling and plasticity. Do you know what culling is? If not you don’t have a valid opinion about race and IQ. That would be like not knowing what an “Internal Combustion Engine” and a “transmission” are and thinking you have a valid idea of how to fix your car’s drive train. You’d be wrong.

About Gender vs. Race and IQ (or any other trait): Gender is both very real and highly constructed. It is probably often more constructed by context and upbringing than ever race is, but there are real aspects of gender. The vast majority of individuals who are constructed as women cannot inseminate a person with viable sperm in the absence of special technology. The vast majority of individuals who are constructed as men cannot carry and birth a baby at this time. Except in that one movie. This is for a number of biological reasons. The evidence suggests that a certain number of measurable gender differences in behavior between various genders are linked to biological differences and probably have something to do with hormonal conditioning which, in turn, may be mediated in some cases by behavior and cultural or social environment (so even hormonal differences are not entirely independent of constructed context). But there is all sorts of biological stuff going on there. And everything in the above paragraph applies to rats as well as humans.

Of course, you don’t inherit your gender, exactly. Well, OK, there is an ongoing argument that gay-osity is heritable. Maybe or maybe not. The argument seems to gain strength then get shot down again and again, like one of those tings many people need to believe is true but isn’t. If it is true, it is pretty wishy washy and depends a lot on stuff that is in turn hard to pin down. But your basic maleness vs. femaleness with respect to reproductive parts and so on is basically not inherited but is provided genetically, as we all know.

“Race” on the other hand is inherited, but in a very complex way. Since race is a social construct, two elements are needed to produce a certain race. First, there must be a construct extant that responds so some signal (like skin color or language dialect), then there must be a signal produced by a particular genetic variant (like skin color) or, in some cases, just a construct (like language dialect).

Imagine a racist act. Many racist acts occur in a broader social context and can be understood by all the people in that cultural milieu as such. Racists acts often have names or commonly understood index terms associated with them. Most people know at least roughly what the racist act is, how it is done, to whom (which race) it is done and by whom (which race) it is done, etc. That is the socially constructed racist act, and linked to it is a socially constructed race.

Then there are the people. Among the people there will be allelic variation … everybody has the same genes, but the genes themselves have variants … alleles … that result in different phenotypes. So among the people there will be individuals of one socially constructed race and individuals of another socially constructed race, and the defined differences and identities will be an interaction between the alleles and the social constructs.

So if you have a handful of alleles that make you seem to be a Native American, for instance, some professor of higher education may look at you and think “Oh, another one of these guys. Last Native American I had to deal with …. well that didn’t go so well. Let’s get rid of this guy.”

That was the expression of a genetic trait possessed by the victim of a racist act. The genotype was the set of alleles that code for Native Americanosity, and the trait, in its fully expressed glory, was a racist act that emerged from the social context.

The same sorts of things happen with respect to both gender and race. In all cases it is hard to draw lines or make clear links between genotype and phenotypes. It is not so hard to understand the power relationships that usually drive the acts themselves. Even if most people engaged in these gendered and race-driven act are not cognizant of the power relationships, they are usually there.

Research in gene-behavior interaction is important. Research in genetic variation is important. Research based on either a race model (of any kind) or a simple two-step gender model is neither important or valid because such research is based on assumptions that not only cart-before-horse but are also sufficiently discredited to be abandoned. And, I suspect that not too much of this research is actually being funded anyway. A fair amount is published, but I’d love to see the actual link between funding source, proposal, research, and publication. I’d wager there is some disconnect there.

Steven Rose (2009). Darwin 200: Should scientists study race and IQ? NO: Science and society do not benefit Nature, 457 (7231), 786-788 DOI: 10.1038/457786a

Stephen Ceci, Wendy M. Williams (2009). Darwin 200: Should scientists study race and IQ? YES: The scientific truth must be pursued Nature, 457 (7231), 788-789 DOI: 10.1038/457788a