Todd Akin’s empirical question

So, as you may have heard, Mr Akin, the Republican Party Senate candidate in Missouri (hi, Missouri!) has some weird views on rape, conception, and abortion. Basically, this is what the Todd Akin t-shirts are going to be saying this November:

It seems to me, from what I understand from doctors, that [conception following rape] is really rare. If it’s a legitimate rape, the female body has ways to try to shut that whole thing down.

Ha! “That whole thing!” Lovely. Now, much of the controversy surrounding these views stemmed from his use of the term “legitimate rape“, but really that was (it seems) just a temporary bout of word salad. He misspoke. You know, like when (Warning: old joke ahead) you want to say “Please pass the salt” but end up instead shouting “YOU RUINED MY LIFE!

Anyway, Akin himself says he misspoke — i.e., he didn’t mean to say “legitimate” — and so we can’t really ever know otherwise (I guess we are to assume that he is not misspeaking when he is telling us that he misspoke).

But the more substantive controversy remains. Because Akin’s main point was not that rape was/is/can be “legitimate”,  but that — get this — a woman’s body is able to prevent pregnancy when she is being raped You see, it’s just nature’s way of preserving justice in an otherwise randomly violent world. Or, more to the point, it’s God’s way (and that’s the Christian god, you guys. Other gods don’t offer this service).

So think about what this means: If women who are raped can’t get pregnant, then women who get pregnant can’t have been raped; therefore, we shouldn’t legislate to allow abortions for pregnant rape victims, because women who claim to have become pregnant through rape are probably liars. You see how elegant this implied-but-unstated logic is?

Note to Todd Akin: All this stuff is covered pretty well in Wikipedia

Just to get it mentioned straight away, the theory is bunkum. Despite Akin’s claim that “doctors” told him (go doctors!), there is no science to show that there is any credibility to the idea.

However, statistically speaking this is actually a tricky question. In fact it’s a lot more difficult to explain than many media reports have suggested. This is for a few reasons. First of all, as you can imagine, the data here are hard to gather. While many cases of rape are reported to the appropriate authorities, many others are not. And data on non-rape intercourse is only gathered as part of research studies, which means that it is incomplete and possibly non-representative.

But a more conceptual awkwardness is that the benchmarking needed to address the claim is quite nuanced. What we need is a baseline rate of pregnancy-per-instances-of-intercourse. This sounds all very well in theory, but in practice is complicated by the way fertility fluctuates across the menstrual cycle. For example, if a woman has sex more than six days before ovulation or two days after, her chances of becoming pregnant will be less than 1%. If she has sex five days before ovulation or one day after, this increases to 5%. Even if she has sex the day before ovulating (which will maximize the probability of pregnancy), the likelihood of conception is no more than 30%.

So, essentially, for about five days per cycle, a woman’s chance of becoming pregnant after intercourse ranges between 10% and 30%. For a day either side of this, it’s around 5%. But for the remaining 21 days or so — i.e., for two thirds of the time — the probability of post-intercourse pregnancy is less than 1%.

So this is all pretty complicated on one level. But not so bad on another. Because we can use this info to arithmetically calculate the expected pregnancy rate across the menstrual cycle on average. And this turns out to be around 4 to 5% (I worked this out all by myself and all; see here). In other words, if a group of 1000 women — representing all points on the average menstrual cycle — have sex, then around 46 of them — 4.6% in other words — will become pregnant.

And what is the rate of pregnancy among rape victims? Yes, you’ve guessed it. Pretty much 5% on the nose. Which, as you see, actually turns out to be slightly higher than the 4.6% rate seen generally.

Now, of course, it goes without saying that caveat, disclaimer, limitations, more-research-is-needed, generalizability, sample selection, and all that. But what is certainly the case is that even with rough and ready back-of-envelope calculations, we can see that there is no trend for pregnancy rates in rape to be any different from pregnancy rates in general. There is no holy-spirit-as-condom effect.

Here’s more of it, in a book this time

So that’s that idea eliminated from all human discourse forever then. Well, perhaps not. Because the “doctors” who told Todd Akin that women’s ova were somehow equipped to distinguish and repel the sperm of rapists were actually not coming up with anything new. Conservative commentators, many of them in positions of significant authority, have been peddling this theory for years. Here are some lovely sound-bites as collated by Randy LoBasso over at PhillyNow:

“…Conceptions from rape occur with approximately the same frequency as snowfall in Miami!” — Federal Judge James Leon Holmes (1979)

Women’s bodies “secret[e] a certain secretion which has a tendency to kill the sperm”, rendering the odds of a rape pregnancy as “one in millions and millions and millions” — Rep. Stephen Freind of Delaware County (1988)

“…people who are raped — who are truly raped — the juices don’t flow, the body functions don’t work and they don’t get pregnant…To get pregnant, it takes a little cooperation…” — Rep. Henry Aldridge of North Carolina  (1995)

Hmmm. So it looks as though the no-pregnancy-when-raped idea is one that defies empirical argument. It all involves juices, secretions, and stuff like that, not data. Who’dathunkit? Even on Tuesday, these guys were still at it, employing anecdotal evidence to trump all those numbery statistics:

[On raped 12-year-olds becoming pregnant?] “Well I just haven’t heard of that…” — Rep. Steve King of Iowa

So, basically, if he hasn’t heard of it — personally — then he can assume it never happens.


People sometimes bemoan the promotion of a scientific worldview as a kind of arbitrary fandom running riot at the expense of nice, humany stuff like the arts. The Todd Akin episode shows us why this isn’t true. Science is a way of winning arguments, a tool for resisting the oppressive, misogynistic, medieval, anti-victim primitivism of privileged hegemonic reactionaries like Akin, King, and the others.

Clearly, these arguments are not easy to win. With grown, well educated men in positions of legal authority willing to blurt out dogmatically motivated hearsay as if it were actual fact, it is no surprise that millions of people come to believe that that is what knowledge looks like.

Akin may be getting slaughtered right now for using the term “legitimate rape“. But it’s the underlying pseudoscientific theory of human fertility that his comments represent — and reinforce — which deserves our longer-term attention.

One Comment

  1. Brian Hughes

    Reblogged this on The Science Bit and commented:

    I don’t usually re-blog my own posts, but this week, like many other folks, I have been taken aback by Donal Trump’s ongoing war on women — in particular, his claim that women who have abortions should definitely *not* be not punished, or punished, or not.

    It has been said that Trump has equally enraged both right-wing conservatives and left-wing liberals with his comments. This is because left-wing liberals are supposed to be pro-choice, and right-wing conservatives are not supposed to say they think women should be punished (it doesn’t poll well with, for example, women).

    Instead, they are supposed to say things like what Todd Akin blurted out in 2012; namely, that nature (i.e. God) has ways of ensuring that abortion is never truly necessary.

    Hence, the following re-blog. The view Trump was trying to express has a very long history, and is unlikely to go away any time soon. Especially if he becomes President…

Leave a Reply