Someone on The Internet is Wrong

Someone on The Internet is Wrong

Here's an article. Here's my first thoughts. And here's where the real discussion begins.

Needless to say, it's an emotive subject. Bringing up Bayesian inference isn't really as effective as you might expect when you're talking about something people feel so strongly about. It's just a shame that people can be so secure in their conclusions, convictions and rhetoric that they won't examine their arguments for the same.

In this case, I think the author is broadly correct. But I don't think the author's arguments are. And I think to ignore that would be doing a disservice to the correct conclusions reached.



I guess it's just fun to you - must be - to think of women scrambling around trying to give significance to the fact that a serial rapist wasn't really, all the time, through and through, a rapist, since, you know, he never raped and murdered his female co-workers or the cashier at the grocery store. And that's just as indicative of his likelihood of raping someone as the fact that he actually raped and murdered a slew of prostitutes. For math's sake, of course.

But I'm of the mind that sometimes women's safety is more important than mathematical thoerems, and that, where women's safety is concerned we might be better served to ignore certain numbers. Even numbers Bayes would like.
And I'm of the mind that, especially where anyone's safety is concerned, we're best served by using the full extent of our knowledge, all our rational capacity, and everything we can learn and have learnt. I can't stand wilful ignorance, and it's especially repugnant when people's lives are at stake.

Without letting rebuttal get in the way of you self-destructing (we should "ignore" facts, when it's convenient? female rapists don't exist? unicorns? what?!), I've never said an incident where a person didn't rape someone else is "just as indicative of his likelihood of raping someone" as an incident where that person did rape someone. In fact, I repeatedly said the opposite.

You conflate my argument (rape incidents give us conclusive evidence of someone's proclivity to rape, non-rape incidents give us inconclusive, weak evidence of someone's proclivity to not-rape - evidence nonetheless) with an entirely different one. One that's so far removed from what either of us think that you can easily caricature the argument as taking the matter lightly, or being enveloped in mathematics.

Still, should it be convenient, I'm sure you'll abide ignoring what someone is saying. Your beliefs are "better served" that way.