Apparently the imbalance in male and female numbers of top science types is such that everyone should be reading one specific book, Inferior: How Science Got Women Wrong by Angela Saini. So much so that people are raising money so that free copies can be sent to schools or summat. A pity that it’s this book as it manages to get the core of the subject wrong. The modern understanding is not that women are worse at anything at all, let alone science. It’s that there are certain skills, attributes and desires spread across the population, some of them more common in males, some in females.
And that’s it. To fail to understand this is to fail to understand the very heart of the science itself. This being something that this book does not understand, thus this campaign isn’t going to be useful:
Two scientists have launched a campaign to get a copy of a book that debunks accepted scientific “facts” about women into every state school in the UK.
The physicist Jess Wade, best known as “chief troublemaker at Imperial College London”, and Claire Murray, a chemist and beamline scientist at a UKsynchrotron, are raising funds to buy copies of Inferior: How Science Got Women Wrong and the New Science That’s Rewriting the Story by Angela Saini. The actor Daniel Radcliffe has described it as one of his favourite books.
The pair are hoping to raise £15,000 via a crowdfunding campaign in an effort to encourage more girls and young women to educate themselves about the structural barriers they face and how to overcome them.
As I say it’s a pity that it’s this book for it does indeed get the central contentions wrong. It’s attacking a straw man, never a good philosophical nor logical stance. More than one strawman in fact.
But if reading Inferior isn’t always a comfortable experience, neither, it would seem, was writing it. While Saini admirably covers the research of experts from opposing camps, deftly fielding points and counter-points from each, she comes across as little short of hostile in her take-down of developmental psychopathologist Simon Baron-Cohen and his research into sex differences in the behaviour of babies.
Well, apparently this is her debunking of that research:
But in charting research into sex differences from cradle to old age, Saini discovers that many of society’s traditional beliefs about women are built on shaky ground. Gender identity is very different between boys and girls, and there are also slight differences in toy preferences – with some evidence that biology might play a small role. But for everything from fine motor skills to vocabulary, colour preferences to aggression, the overlap between boys’ and girls’ behaviour is huge. Differences, if they exist at all, are tiny.
But then that’s what the research actually says. There are varied skills, attributes, desires and interests spread across human beings. The variation between individuals is vast. That determined, even noted or extant in men and women varies not that much – not as much as it does over individuals. But there is a different distribution that we can note across the population which is at the very least correlated with sex.
To take the bile out of it. Heights of individuals very wildly. Hmm, no, between them, it’s not that height varies for an individual. There are women very much taller than the average man, men very much shorter than the average woman. Yet it’s still true that across the population a man is likelier to be taller than a woman. That is, we can say nothing certain about an individual, we can only remark upon probabilities. All that Baron Cohen and other research is saying is that this is true of other attributes.
She’s trying to disprove him by agreeing with him. Sigh.
The opening chapter makes other but similar mistakes.
Larry Summers is said to have talked of “issues of intrinsic aptitude” between men and women, that a biological difference exists between men an women. Well, an obvious retort is that if biological differences don’t exist then we’re going to have problems producing the next generation, those differences being rather what they’re for. But that’s also not what Summers said. Rather, when asked about the lack of female professors in certain hard sciences at the most prestigious universities he pointed to the difference in the very tails of the distribution. Those very tails being what prestigious universities regard as the talent pool they go fishing in. And there’s nothing controversial at all about saying that those extreme tails trend male. It’s true of both ends too, there are more males among geniuses and cretins. The reason saying this is uncontroversial is that it’s true.
Even her insistence on pointing out actual discrimination doesn’t work. She tells us of how children do interfere with female careers more than male. OK, it’s a pretty obvious thing to note. As she says, childless single women don’t face a gender pay gap. We agree that far. But then she says that randomisation of male and female names on job applications shows that, the women being less likely to be hired, is discrimination. But you’re not allowed to ask about children these days, neither existence nor plans or them. And we’ve already agreed that children do affect such matters. So, is not hiring the woman for the management position taste discrimination or rational such?
The biggest problem though is that she’s horribly misrepresenting what the current science is trying to say. It isn’t that “men and women are different” in brain power. Rather, that individual variation is vastly greater than group but we do still find variations across groups. This being entirely unimportant when we consider an individual but rather so when we try to measure group performance. Obviously. The test of whether to hire a female scientist is that individual. But looking across all science to see the number of women is going to be influenced by that population distribution. We can thus – as we think we are – be entirely fair to each and every individual and still end up with a lopsided system result.
We’ll not be reading the rest of it. And we do think that if every school is to be fed a copy of a book on the subject it should be one that gets all of this right.