Gender is not a social construct

[SheWhoMustNotBeHeard]

A very insightful interview with Debra Soh, the Canadian scientist on Triggernometry. She talks about the research in differences in male and female brains, trans people's brains being more female if they are tw and more male if tm but that results are confounded by their sexuality. She talks about transitioning children and why the affirmative approach is harmful.

She also disagrees that gender is a social construct. I didn't really understand what she was saying tbh. She didn't say what is was either. Can someone clarify it for me?

@DianasLasso

Will have to watch later, but as I understand it there are two schools of thought on differences in cognitive behaviour between men and women.

1 - what you might call the "nurture theory". Emphasises that differences in cognitive performance have tiny d values - i.e. the difference between the mean performance for women and that for men is tiny compared with the variability within each population. Points out that the brain is immensely plastic and responds to external stimuli, and that we live in such a sexist society that right from infancy girls and boys are treated differently ie exposed to different stimuli. So there's no way you could construct an experiment which would tell you whether cognitive differences (tiny and overlapping) can be mapped onto neurological differences (again, tiny and overlapping), and whether such differences were "hard-wired" according to sex or developed contingently as a result of brains being exposed to different stimuli in a sexist society. This is Cordelia Fine and Lise Elliott's position.

2 - the nature argument: the small but measurable differences in cognitive performance can be mapped onto small but measurable differences in underlying neurological structures, which in turn are at least in part driven by chromosomal sex differences, not just brain plasticity responding to differing stimuli. I think this is Deborah Soh's position (and Simon Baron Cohen and Steven Pinker).

I'll be interested to watch to see how Soh deals with the claim that, given what we know about brain plasticity, and also how early in life parents and carers start differentiating how they treat girls and boys, how one could create a clean experiment to separate out nature and nurture.

I think position 2 is actually the more common one among people whose area of study it is, and is considered moderate, as it really includes both nature and nurture. Fine's position as I understand it is considered rather on the edge, and tends to argue that difficulty showing a clear conclusion implies a particular reality.

I hate to come on such an intelligent thread and lower the tone but I wanted to say Steve Pinker is a duck.

I mean, something like that.

The Blank Slate had a great premise, but it eventually became clear the man has virtually no idea what he's talking about and was happy to use any number of specious and illogical arguments to try to prove his ideologically driven theory.

So I'm obviously new to the debate and still trying to figure things out.

The male versus female brain. If brains are different, could it be that the majority of males have typically male brains but plenty of males don't and the same with females? Have there been many studies to show the differences? But at the same time, how do we know that nurture hasn't influenced those brains? So, how do girls who go to single sex very academically driven schools brains differ to those who go to a normal comp? Is there a difference?

It will useful to see her references and bibliography.

I thought her view on transitioning children was spot on and it echoed what we all have been saying, especially about the future law suits.

I also found what she said about research could happen in the future quite chilling and probably accurate.

The only researcher I knew in the area (male prof in a Russell league university) was in camp 1, but this was a while back. I do suspect there's a lot of "looking for the lost keys under the streetlamp" going on; it's undoubtedly the case that quite a few researchers are enormously invested in finding evidence that there are sex differences, so they put an awful lot of energy into coming up with studies which will find them.

The key issue for me, Goosefoot, is how you construct a clean experiment which could show the "proportion of the variance" attributable to nature, and the proportion attributable to nurture. For a lot of nature/nurture questions in psychology and physiology, twin studies are the gold standard, but obviously that doesn't work for sex differences, because girl/boy twins are always fraternal rather than identical.

I think sensible discussions of it are also hampered by the tendency of the press to present what are in actuality two massively overlapping distributions with a tiny d-values as being more like a couple of separated delta functions!

The thing is that if everything was completely plastic and responsive to our thoughts and feelings there would be no need for medical intervention. A person would just morph into what they want...

She seems to contradict herself about male and female brains. She says that there are traits which are more common in one sex or the other, but that there is a lot of overlap, meaning that there is no such thing as a male or female brain. She then says that people who identify as transgender have brains which are more typical of the opposite sex (in some way which she doesn't elaborate on). But so do many people who aren't transgender, since there is a large overlapping distribution.

She also says that 99% of people identify with the 'gender' which goes with their biological sex, which is clearly nonsense, particularly for women.

it's undoubtedly the case that quite a few researchers are enormously invested in finding evidence that there are sex differences, so they put an awful lot of energy into coming up with studies which will find them.

I do worry about the motivation of people who are intent on finding sex differences in the brain.

OP a picture might help. This is a diagram I've borrowed from wikipedia (attribution here: commons.wikimedia.org/wiki/File:Cohens_d_4panel.svg) of the distributions for d-values of various sizes (could be height, could be age at which a toddler reaches a vocabulary of 50 words, could be percentage attained in a standardised maths test at age 8, could be distance a child can throw a tennis ball...)

One thing to notice is that in no case have we got two completely separated "spikes". For small d values, we have distributions in which the variation between two randomly selected men, or two randomly selected women, is likely to be far bigger than the difference between the means of the two populations.

Height has an overlap of round about a d value of about 2 (that's from memory, don't quote me). Cognitive differences typically less than 0.5 - i.e. the distributions look very, very similar.

If someone gave you a tissue sample from a brain, or an MRI scan of a brain, you might be able to locate it on the probability distribution and read off the probabilities - say 65% that it came from a man, or 63% that it came from a woman. But you couldn't establish the sex of the person simply from the brain scan. That's not how it works.

There aren't "sexed brains", there are brains - with very variable characteristics - within sexed bodies.

Re:twin studies / Adoption studies / heritability - this way of measuring heritability doesn’t work for male/female or differences between groups with any Readily identifiable difference (eg different racial groups). The reason being, that by virtue of being female vs male, people treat you differently. Your environment is More different when comparing to a sibling of the opposite sex vs a sibling of the same sex. So you can never control for environment. Unless we start doing manipulation studies but these are unethical obviously.

Certainly observation studies (Eg comparing women’s achievements once barriers removed to before, women’s performance in all female vs mixed environments, etc) suggest a strong role for nurture. We can’t rule out nature but we can’t rule it in either.

She also says that 99% of people identify with the 'gender' which goes with their biological sex, which is clearly nonsense, particularly for women.

Sohs not a feminist - she’s using the TRA definition of gender which is gender identity, not the feminist one which is sex role stereotypes. She’s saying that 99% of (Eg) women are willing to accept being referred to as female / with female pronouns which is true.

@veryvery

The thing is that if everything was completely plastic and responsive to our thoughts and feelings there would be no need for medical intervention. A person would just morph into what they want...

I think you've misunderstood neuroplasticity. No-one's saying it's easy.

Take London cabbies as an example. Learning the "knowledge" is phenomenally difficult. It requires years and years of study. Very few people turn out to be able to do it. But those who do end up with measurably bigger hippocampuses than the rest of the population.

www.bbc.co.uk/news/health-16086233

No-one's "born" to be a London cabbie. Cabbies haven't been around long enough in human history for this to have been evolutionarily selected for. It's a response to repeated stimuli, showing that our brains aren't blank slates or hard wired, but that stimuli can actually change the make-up of the brain.

Have there been many studies to show the differences? But at the same time, how do we know that nurture hasn't influenced those brains? So, how do girls who go to single sex very academically driven schools brains differ to those who go to a normal comp? Is there a difference?

A great question OP. I do not know about any such studies. Neuroscience is just beginning to accept widespread Neuroplasticity. Fine and Rippon do cite some studies. I recall studies demonstrating that sexed brain differences increase over development, which is what one would predict if nurture is causal to brain differences. Of course the “nature camp” could turn around and argue that the brain differences are caused by sexual changes during development (Eg puberty), and the Behavioral/social differences result. It’s very hard to tease apart.

Sohs not a feminist - she’s using the TRA definition of gender which is gender identity, not the feminist one which is sex role stereotypes. She’s saying that 99% of (Eg) women are willing to accept being referred to as female / with female pronouns which is true.

Thanks for that explanation. I found her comments about gender to be quite garbled, so when she said that gender wasn't a social construct she didn't seem to be talking about gender at all, because she went straight into gender identity and gender expression. But that makes more sense if that's what she means by gender.

I was thinking how can socially constructed gender roles (or sex stereotypes) not be socially constructed?

The "more similar on average to the brain of people of the opposite sex" claim that Soh makes (which I believe the research literature hasn't actually established one way or another) still doesn't settle the ontological question.

Suppose for the sake of argument (Student T test, Welch's test, whatever) you found that the distribution of test results for TW looked like it had been drawn from the female population rather than the male population for a certain characteristic, significance at the 95% level.

This still wouldn't establish that TWAW. That's a philosophical question about ontology, not a scientific one. You could interpret that result as lending credence to the claim that TWAW only if you already thought that internal feelings/thought processes were more important to being a woman than reproductive biology. Conversely, if you thought reproductive biology was the feature underlying "being a woman", you'd interpret one and the same scientific result as explaining why a group of male people strongly believed that they ought to have been born women, even though they weren't.

(Bit like the so-called "god" region of the brain which was in the news a few years ago - a region of the brain which lit up in MRI scans when people reported having religious experiences, and which, when electrically stimulated, produced religious experiences. For a religious believer, its existence explained how we were able to experience the divine, a bit like the visual cortex explains how we can process visual information. For the atheist, it explained why some people had the illusory sensation of the existence of the divine, even though there was no external reality it corresponded to. It didn't actually settle the question of the existence of god one way or the other.)

Not misunderstood. I just wasn't been entirely serious.

Thanks DianasLasso. I think I understand what you are saying; the picture helped. I will have to look at the thread again when I feel more with it.

Can someone explain to me in very very simple terms, what she meant by gender isn't a social construct? What is it then? She seemed to imply that gender roles was a social construct though.

Can someone explain to me in very very simple terms, what she meant by gender isn't a social construct? What is it then? She seemed to imply that gender roles was a social construct though.

One problem that bedevils this whole debate is we have one word - gender - doing double or even treble duty.

As it was traditionally used in social sciences (anthropology, sociology, feminist theory) it meant "the set of roles, dress codes, acceptable behaviours, occupations, legal status, etc. a specific culture deems appropriate for members of one sex or the other." This sense is clearly a cultural construct - Medieval Iceland: weaving is women's work; Medieval England: weaving is a manly occupation and therefore paid twice as much as spinning.

Then there's gender as a synonym for sex because sex is well, y'know, bit naughty and all that. This crept into even the academic literature.

Then there's "gender identity" - a sense of one's own internal manliness or womanliness or in-between-i-ness or neither-ness, which some people report having, and some people don't experience.

My guess is Soh is trying to claim the last of these is rooted in neurology and is thus biological rather than sociological. (There are various counters to this claim: one is that rates of people transitioning seem to vary across cultures, but do seem to be tied to some extent to how rigidly enforced the sex stereotypes understanding of gender is within a society; another is that accounts of what this internal womaliness or manliness consist of often seem to involve a list of sex-stereotypes; a third is phenomena like social contagion in girls schools, for instance, where a number of girls in one class suddenly decide they're trans-boys at a statistical frequency far in excess of the population at large).

Suppose for the sake of argument (Student T test, Welch's test, whatever) you found that the distribution of test results for TW looked like it had been drawn from the female population rather than the male population for a certain characteristic, significance at the 95% level.

If it was still within the normal range for men (even if it was at the extreme 'female' end of that range), there would be no reason to interpret it as anything other than a male result. If it was outside the usual range for men, it would just extend the male range a bit further in that direction, because the sample is of biologically male people.

If you find a trait in a male brain which is you previously believed was only present in females, that doesn't mean the male brain is female, it means that that particular trait can also be present in males.

So basically from those pictures, to a round number, half of women have male brains and half of men have female brains?

@midgebabe

So basically from those pictures, to a round number, half of women have male brains and half of men have female brains?

Pretty much, yes . Which is why talking about "male" and "female" brains is such a stupid idea.

OldCrone - I'm talking about populations in a technical statistical sense. One of the things statistics regularly does is this: you have a small bag of widgets, with a certain percentage of them faulty. They could have been produced by machine A or machine B, and you know how what percentage machine A produces that are faulty, and what percentage machine B produces that are faulty, over very long runs (1000s) of widgets. How likely is your bag to come from machine A or machine B?

This is what a lot of the brain studies of trans-identifying people claim to do at a population level rather than an individual level. They claim not that individual brains are male or female, but that the subsample of transwomen's brains has the same statistical properties as the distribution of brain features seen in a control group of natal women, and is statistically distinguishable from a control group of natal men.

What I'm arguing is even if this is the case (and the studies I've looked at have had severe problems with small sample sizes and quality control of data - NB I am a physical scientist, not a neurologist, but I do know a reasonable amount about statistical methods) - even if it is the case, that wouldn't solve the philosophical question of whether that match in statistical properties was because TWAW or because the similarity in brain structures explained why a group of natal men had an overwhelming psychological feeling that they ought to have been born as women.

Yes, I know what you were getting at, Diana. But I think you can extend my argument to a distribution which looks different from what you might expect.

Using your example, in this case we know which 'machine' the sample comes from, and even though it looks as though it's more likely to have come from machine B, we actually know it comes from machine A.

So the sample is just an atypical sample of men, not a sample of female brains in male bodies.

that wouldn't solve the philosophical question of whether that match in statistical properties was because TWAW or because the similarity in brain structures explained why a group of natal men had an overwhelming psychological feeling that they ought to have been born as women.

Isn't the question whether they are literally female brains in male bodies, or just atypical male brains in male bodies which causes them to feel a certain way? One of these requires a form of belief, the other is just science.

Oh, I agree - the crucial thing is we know the widgets came out of machine B, which to my mind makes it much more plausible that the evidence (such as it is, and it's actually pretty weak) supports the hypothesis that this might explain why some biological males feel the way they do.

But also you have to establish that it's unlikely (95% level or more on a two-tailed distribution) that the same statistical distribution of brain features would arise from simply randomly sampling the male population. And as I said, most studies I've seen have very small sample sizes. Recall that we're generally looking at two populations with a very small d value between their distributions - so to unequivocally decide (or more correctly, decide at a certain significance level) that a subsample of people statistically matches one population rather than the other requires a big sample size.

I also suspect we quite often see "p-hunting" going on in these studies - where the researchers slice and dice their data until they hit on a feature, however ad hoc, which finally delivers that elusive 95% significance level. ('Cos 95% signficance isn't telling you "can't happen by chance", it's just telling you "only happens by chance one time in twenty".)

By 'not a social construct' she means girls are naturally girly, and boys are naturally boyish.

That's why it's the conservative thesis. She's saying there's lots of truth to sex stereotypes ie women are 'naturally' more submissive, nurturing etc and men are 'naturally' more aggressive, power and status seeking etc.

As opposed to feminism, which suggests these stereotypes are not part of the brain but part of the culture a child is raised in.

She's not suggesting that trans people have the brains of the other sex, but that their brains have gone 'wrong' for their own sex. Transwomen's brains are more girly than is natural. Transman's brains are more butch than is natural.

Everything revolves around the ideological standpoint that differences between men and women are global, natural and highly significant, and lead to observed differences in things like pay, and employment etc.

I think it's likely that both nature and nurture have a role to play in average observed differences between the sexes, especially in areas related to mating and procreating. In other words, I'm sympathetic to a limited role for evolved differences, whilst observing that the brain responds and changes according to culture from birth onwards. It's anti what we know about brains to essentialize differences as solely natural.

She's not suggesting that trans people have the brains of the other sex, but that their brains have gone 'wrong' for their own sex. Transwomen's brains are more girly than is natural. Transman's brains are more butch than is natural.

What ‘went wrong’ with the brains of the individuals in that study was puberty blockers. There was nothing ‘wrong’ before the participants started taking these.

No argument from me. Opposite sex hormones, same sex orientation plus the effect of environment/culture on the brain may create observable differences which point to nothing other than the brain being impacted by one or more of those factors.