How Many Rape Reports Are False?
How many women falsely accuse men of rape?
A lot of statistics are floating around the Internet: Two percent, say many feminists, the same as other crimes. Twenty-five percent, say other groups who quarrel with the feminists on many issues, or maybe 40 percent. Here's the real answer: We don't know. Anyone who insists that we do know should be corrected or ignored.
The number of false accusations is what statisticians call a "dark number" -- that is, there is a true number, but it is unknown, and perhaps unknowable. For a deep dive into the reasons it's so hard to know, I commend you to Cathy Young's new piece at Slate, in which she details all the problems that confound investigations into false rape accusations.
Here's what we do know: The 2 percent number is very bad and should never be cited. It apparently traces its lineage back to Susan Brownmiller's legendary "Against Our Will," and her citation for this figure is a single speech by an appellate judge before a small group of lawyers. His source for this statistic was a single area of New York that started having policewomen conduct all rape interviews. This is not data. It is an anecdote about an anecdote.
The 41 percent number beloved of men's-rights activists is better; it involves a peer-reviewed study by Eugene Kanin of a police department in some unknown small city. False reports could only be declared if the victim herself withdrew the charge. However. We're talking about one city, in which 109 rapes were examined over a period of nine years. As feminists point out, victims might have withdrawn the charges simply because they found it too traumatic to engage with the police department, not because the accusation was false . And the study itself is now pretty elderly. A lot has changed in 20 years, including, possibly, the number of false rape accusations in this city and the rest of the nation. This number should be used only with grave caution.
But so should any other numbers, such as the 8 percent figure that is commonly attributed to the FBI. When you dig into the research itself, you find it is often heavily inflected with the authors' prior beliefs about what constitutes the "real problem": unreported cases of rape or false reports? So Kanin is frequently chided for accepting the results of a police department investigation that included offering the victims a polygraph, because this is intimidating for true victims as well as women making false reports, and it could raise the incidence of false negatives. On the other hand, if the rate of false rape reports is quite high -- much higher than that of other crimes -- then this might be a reasonable precaution. It's possible that by encouraging police departments not to polygraph rape victims, we have fixed a cruel system in which innocent victims are bullied into recanting. It's also possible that we've increased the number of false accusations that proceed to investigation and conviction.
Shorter: You cannot treat "percentage of reports that were found to be false by investigators" as "percentage of reports that were actually false." Some women may simply have recanted to disengage from the system. Some police officers may decide a case was false when it wasn't. On the other hand, we also know that false accusations can make their way through the system pretty far -- witness the Duke lacrosse players and Brian Banks.
What we know is that we don't know. We should not presume that every rape victim is telling the truth because it would make it easier for victims to come forward. Nor should we presume that every rape accusation has a 50 percent chance of being false. We should look at the facts in each case and judge them with the knowledge that some women do lie about rape -- for revenge, to cover up some problem in their own lives, to get attention and sympathy from others. And also with the knowledge that men lie, too, violating their victims a second time in order to cover up their crimes. And that while men have gone to jail for rapes they did not commit, many other men have avoided the jail time they deserved for terrible crimes against women.
That's not a very satisfying answer, because rape is inherently a hard crime to prosecute. If someone comes into a police station with their face bashed in, you can be pretty much certain that unless they're a professional boxer, a crime has occurred. If a rape kit shows evidence of sexual intercourse, however, all that tells you is that ... something happened. Because this is something that a lot of people do to each other voluntarily, you cannot proceed immediately to the arrest. Usually there are only two witnesses, telling different stories. Often drugs or alcohol were involved, and intoxicated people make lousy witnesses.
We don't want that to be true. Rape is an especially heinous crime, and heinously unfair -- it is mostly something that stronger men do to weaker women. How can we pile on an extra dose of unfairness -- by failing to prosecute so many of the crimes?
Feminists would like to rectify that unfairness by treating rape accusations as presumptively true, making it easier for victims to come forward. That's understandable. But there's a risk that this makes it easier for false accusations to get through the system, resulting in destroyed lives for men such as Brian Banks. Men's-rights activists would like to make it harder for innocent men to get caught in a web of lies, so they want rape accusations to be interrogated with deep suspicion. But treating rape victims as possible or likely liars may make it harder for them to go forward, leaving rapists free to stalk their next victim.
No one wants to openly advocate for a hard choice that will end in injustice for someone. So instead we get the war of bad statistics, with each side claiming certainty when all we really have are dark crimes and dark numbers.
Though to be fair, Kalin, the author of the study, goes into some detail about the nature of the recantations, and they're both pretty specific and pretty plausible -- i.e., "my husband is overseas and I'm afraid I'm pregnant." They mostly occurred a couple of days after the report, not after extended contact with "the system."
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Megan McArdle at firstname.lastname@example.org
To contact the editor on this story:
James Gibney at email@example.com