Posted tagged ‘stereotypes’

The male blogger (well, just about)

January 10, 2009

Yesterday someone drew my attention to an interesting website, Genderanalyzer. This allows you to enter a URL of a blog site, and it will assess the text and tell you whether it was written by a man or a woman. So of course I let it have a go at this blog, and this is what it came up with: ‘We guess is written by a man (54%), however it’s quite gender neutral.’ Far from feeling this violates my masculinity, I am rather pleased with that verdict.

But then I wondered what else I might let it assess. How about Barack Obama’s famous Chicago victory speech, ‘yes we can’? Well, no doubt about that at all – though perhaps Obama’s fame is such that the software already knew him; at any rate, it concluded that ‘we have strong indicators that [the speech] is written by a man (98%).’ Hillary Clinton also is true to her gender, if not quite as definitively: her speech to the Democratic Convention last year was very likely ‘written by a woman (87%)’. Our own Taoiseach Brian Cowen, facing all his challenges, is much more borderline; when he introduced the paper on Building Ireland’s Smart Economy, his maleness was only 62%. But then again, when his predecessor Bertie Ahern addressed the National Forum on Europe a year ago, he was only 54% male, just like your current blogger.

Stereotypes come through fine: Sky Sports’ Andy Gray writing about Sam Allardyce’s appointment as manager of Blackburn Rovers is 99% male, bless him (though you’d have to wonder about the missing 1%). But then again, what’s this – the wonderful Nigella Lawson is, ahem, 74% male! Really?? Exactly the same result as for Margaret Thatcher’s resignation speech in 1990, but could you compare them?

What does any of this tell us? Interestingly, my 50 or so attempts to get Genderanalyzer to give a verdict on the gender of various writers and speakers was overwhelmingly accurate; the latter two are the only two it got ‘wrong’, though to be fair in Margaret Thatcher’s case I was expecting that. Some of the more borderline cases may actually be explained by the fact that I was assessing texts written for the speaker by someone else (whose gender we don’t know), or even by a team of people. But overall, it seems that gender differences as detected in speech are real enough, and on the whole that seems fine. Our task remains to ensure that this is not reflected in the way in which society creates advantages or imposes disadvantages or allows prejudices.