“All assassins are left-handed,” declared Sherlock Holmes in a detective story written not by Sir Arthur Conan Doyle but by American humorist Mark Twain. Twain, of course, was skewering two traditions: the lore of the all-knowing British detective and the stereotyping of left-handers as deviant personalities. And yet, a faint echo of this latter tradition was heard recently, when a Canadian research group asserted that left-handedness is more common among homosexual men and women than among heterosexuals.
As I’m an out-of-the-closet lefty and gay man, the idea of an association between these two traits passes my personal “smell test.” But is it good science, and if so what does it mean?
The study in question was done by Martin Lalumière, Ray Blanchard and Kenneth Zucker of the University of Toronto, and it was published in the latest issue of Psychological Bulletin. The study was a meta-analysis. That is, the authors did no original research on the topic but instead did a statistical analysis of data from a number of previously published reports. Combining all these studies, the authors have data on 14,808 straight men, 6,182 gay men, 1,615 straight women and 805 lesbians. Result: gay men have a thirty-four percent greater chance of being left-handed or ambidextrous than do straight men, and lesbians have a ninety-one percent greater chance of being left-handed or ambidextrous than straight women.
Thirty-four percent and ninety-one percent sound like major differences between the gay and straight groups, but they’re really not. Indeed, this way of presenting the data obscures the fact that most homosexuals are right-handed, and most left-handed people are heterosexual. Knowing someone’s handedness tells you very little about their sexual orientation or vice versa.
The authors acknowledge that the correlation they found between handedness and sexual orientation is a weak one. They note, however, that handedness is established around or even before the time of birth, so they take their findings to mean that sexual orientation is influenced by prenatal processes of brain development.
Here again, I have a gut sympathy with their point of view, since my own research on the brains of gay and straight men has led me to a similar conclusion. But I have a harder time with their attempt to pin down the mechanism involved. Specifically, they suggest that a fetus’s general vulnerability to stresses can cause it to deviate simultaneously from two specific pathways of development the one that leads to right-handedness and the one that leads to heterosexuality. It is this general vulnerability, they argue, that causes the link between homosexuality and left-handedness.
The core of their argument is the notion that right-handedness and heterosexuality are specifically organized by developmental mechanisms, whereas left-handedness and homosexuality are the consequences of a loss of specificity. In other words, it’s not that there are positive factors that cause a person to be left-handed or homosexual but that the usual factors that cause right-handedness and heterosexuality are impaired.
Left-handedness really does have this flavor of being a “loss-of-specificity” trait, because people we call “left-handed” are in fact nearly always ambidextrous if one tests a variety of behaviors. I play tennis with my right hand, for example. Reflecting this fact, researchers prefer to use the phrase “not consistently right-handed” rather than “left-handed.”
Sexual orientation, on the other hand, does not have this flavor, at least in men. Although studies differ in the details, the consensus is that there are more out-and-out homosexual men than there are bisexual men. To my mind, this suggests that there is a specific mechanism leading to male homosexuality, rather than a mere loosening of the mechanisms that cause heterosexuality.
Among women, on the other hand, several studies suggest that there are more bisexuals than outright lesbians a finding that would be more consistent with the Canadian researchers’ hypothesis. And, indeed, the reported association between handedness and sexual orientation is significantly stronger in women than in men. Still, I suspect that there is at least a subset of women who are lesbian in consequence of a very specific program of prenatal brain development. Of course, there are also potential cultural explanations for the sex difference: a degree of bisexuality may be less threatening to a woman’s sexual identity than it is for a man.
There are other reasons why Lalumière, Blanchard and Zucker’s argument seems unpersuasive. First, both sexual orientation and handedness are poorly understood traits that are probably influenced by a number of different factors. To establish an ostensible correlation between two such traits does not really say much about how either of them is established. Furthermore, the notion of a general “vulnerability” in fetal development is a rather vague concept it could have any number of genetic or environmental causes so it may not be very useful in guiding future research.
Luckily, the Canadians’ report comes at a time when left-handedness no longer seems as “sinister” (the Latin for “left-handed”) as it did in the age of Sherlock Holmes. Little effort is made these days to make left-handed kids into right-handers, at least in our culture. Let’s hope that today’s gay kids will likewise be allowed to express their sexual orientation without fear of “therapy.”