Archive for January 9th, 2013

Whenever someone claims they can tell if a piece of writing was written by a man or a woman, I have to suppress a knowing smile. They have only a fifty percent chance of being right, and a near certainty of embarrassing themselves with rationalizations if they are wrong. Writing, apparently, is a skill that has very little to do with gender.

I first became aware of this basic fact through the reactions to James Tiptree, Jr. As a young teen, I remember critics praising Tiptree for a supposedly masculine prose style. When rumors emerged that Tiptree might be a woman, many explained at great length why that could not possibly be so. Then it was revealed that Tiptree was actually a woman named Alice Sheldon – and in a perfect demonstration of double-think, many of the same critics began explaining how they knew that all along, and pointing out aspects of her prose as evidence for what was suddenly an obvious fact.

Something of the reversal happened a few years later with F. M. Busby, a writer of intelligent space opera. Because Rissa Kerguelen, one of Busby’s greatest successes, featured a female protagonist, dozens of people assumed that Busby was a woman. A man, they argued, couldn’t possibly write such a sympathetic female character. But Busby was a man – although one fond of saying that “An intelligent man who isn’t a feminist isn’t.” The reasons that he went by his initials were that he disliked his given names of Francis Marion, and that his publisher considered his nickname “Buz” too informal for a book cover.

Having these two counter-examples, I have always been skeptical about efforts to identify gender through writing samples. Like too much alleged social science, such efforts always assume that certain subject matter and stylistic choices are somehow innately masculine or feminine (gay, lesbian, transgendered, or queer are always left out). A male writer, for instance, might be supposed to use “I” and to write short, unqualified statements. By contrast, a woman might be said to be more tentative in offering an opinion, and write about emotions or domestic subjects. Needless to say, such divisions say more about the devisers of such studies than any actual differences.

In fact, I’ve always found such studies rather dismissive of writer’s abilities. Most writers I know would have no trouble imitating the so-called masculine or feminine prose styles of such studies. Once they knew the required mannerisms, all that would be needed is a few hundred words of practice.

Moreover, whenever I have tried any online versions of such studies, the results have been random. For example, this morning I ran samples of my writing through Gender Genie, an online adaptation of one such study. My journalistic articles registered consistently as male, and my personal blog entries as female. My fiction registered as both male or female, although neither very strongly. Similarly, two women writers of my acquaintance registered as male, and a male friend as female. I would have tried more samples, but at this point, it was obvious that the results had such a large margin of error as to be unreliable in any given case.

And apparently, my personal observations were correct. Recently, fantasy writer Teresa Frohock invited readers of her blog to identify the gender of the writers of ten different writing samples. Of 1,045 guesses, only 535 were correct – a number slightly above random chance, but well within statistical variation. As Frohock noted, despite all the elaborate rationalizations and the stereotyped ideas that men were more likely to write epic stories and women emotional-driven ones, people were unable to tell men from women based on how and what they wrote.

In other words, exactly what my experience would predict.  Excuse me while I cackle, “Told you so!”

But this subject goes far beyond a mildly diverting observation. The obvious conclusion is that, if writing samples don’t reveal who is male or female, then why are most people so quick to assume that supposed differences in male and female brains are significant? If the products of those brains are indistinguishable from one another, then the brain differences can’t matter much, either. As often happens when gender is discussed, too many people tell themselves comforting stories, then look for reason to believe the stories instead of examining the evidence.

Read Full Post »