UPDATE: There is now a timeline of the major events in these accusations, and the responses to them, on Jason Thibeault’s Lousy Canuck blog. It includes several additional reports of harassment and sexual assault, and several additional pieces of corroboration of these reports. It is being updated as new information comes in and as new events unfold.
So I got this comment on my blog from Hannah Barnhardt:
I have a question about how to handle allegations of rape and sexual harassment. In the local atheist group that I am only now tenuously connected to (because so many members display open disdain for women and feminists), Karen’s allegations have been discussed only briefly, and with criticism and disbelief. Basically, they’re saying: “Well we ARE skeptics after all, and skeptic means we need PROOF! DUURRRR”
But with something like rape, or the kind of sexual harassment Karen experienced (and I do understand Karen has lots of proof, but I’m talking about a case where perhaps, like many cases, there’s not much proof beyond the victim’s testimony), what is the best way to handle cases where there’s not much physical proof? Because I understand how little rape/harassment is actually prosecuted and how difficult it is to accuse someone, I favor giving the accuser the benefit of the doubt.
I guess I’m asking: what’s the best way to respond to these people, who say that there must be ample physical evidence in order to actually DO something about harassment or rape? In the real world, it would be awesome if every person who experienced this kind of abuse had ample physical evidence, but it just doesn’t happen that way. I don’t for one second believe that that means we shouldn’t believe the victim. What do you think?
A good question, and one that has been much on my mind in the last few days.
Here’s what I think, what I want to say to people who are saying this sort of thing: I think you should be really careful about not letting your skepticism turn into denialism.
Here’s what I think:
1: Extraordinary claims require extraordinary evidence. But claims of sexual harassment, abuse, assault, and rape are not extraordinary. They are depressingly ordinary. So the level of evidence we should need to believe a claim about sexual harassment, abuse, assault, or rape is substantially lower than the level of evidence we should need to believe a claim about, say, Bigfoot.
2: Sexual harassers, abusers, assailants, and rapists are typically very good at covering their tracks. They don’t generally commit their acts in front of witnesses or video cameras, or leave a paper trail. Depending on the kind of harassment or assault we’re talking about, they often don’t even leave physical evidence (and when they do, it often doesn’t get collected, since collecting it typically requires the victim to report the assault almost immediately, and subject themselves to further emotional and physical trauma). And perpetrators often cover their tracks in other ways — such as getting the victim drunk, which our culture regrettably tends to see as evidence of consent.
So the kinds of evidence we’re likely to find supporting an accusation of sexual harassment or assault are not straightforward, obvious physical evidence. The kinds of evidence we are likely to find are:
* Multiple similar claims made against the same person from different people. Especially when these claims show a similar pattern of behavior.
* Other people saying that the victim told them about the harassment/ assault shortly after it happened — with stories that are consistent both with the accusation and with one another.
* Other people corroborating behavior that falls short of harassment/ assault, but is consistent with it. Example: If an accused assailant is accused of getting victims drunk first, and someone says they’ve seen this person deliberately getting people drunk while hitting on them, or have experienced this themselves — that would support the accusation.
* Paper trails, email trails, or other kinds of evidence that either directly support the claim — or that show behavior that, again, falls short of being direct evidence of harassment/ assault, but is consistent with it.
(Note that this doesn’t refer to the types of evidence we’d accept in a court of law. See #4 below. And note that “support” doesn’t mean “absolutely prove.” See… oh, the rest of this entire post.)
To make an analogy that skeptics should understand: Think about how creationists say, “Where’s your evidence for evolution? I’ve never seen life spontaneously generate from a peanut butter jar! I’ve never seen fish evolve into mammals in one generation!” Or think about how global warming denialists say, “Where’s your evidence for global warming? Why isn’t the Antarctic turning into Florida? Why was it so cold in Minnesota last winter?” No, of course not. That’s not the kind of evidence you’d expect to see to support evolution or global warming — because that’s not how evolution and global warming work. The kind of evidence you’d expect to see to support evolution is exactly the kind of evidence we do find: evidence from genetics, geology, anatomy, fossil records, etc., all consistent with one another. The kind of evidence you’d expect to see to support global warming is exactly the kind of evidence we do find: evidence from long-term studies of weather patterns over years, decades, centuries, and millennia.
So be a good skeptic. Think about how sexual harassment, abuse, assault, and rape usually work. Think about what kind of evidence you’d expect to see for them. And then think about whether that kind of evidence is present in this case.
3: False allegations of sexual harassment and rape are actually very low. The consequences of making allegations of sexual harassment or rape are very high indeed: public shaming, having one’s personal history — especially one’s sexual history — being subjected to extreme public scrutiny and censure, being traumatized by callous law enforcement officials if the crime is reported, harassment, threats, and more. And the consequences are especially high when the person you’re accusing is powerful: if they’re famous, if they’re rich, if they’re influential, if they have political power.
4: In the conversations we’re having about these incidents, we’re not talking about what kind of evidence would support publication in a peer-reviewed journal, or a judgment in a court of law. We’re talking about what kind of evidence would support judgment in the court of public opinion. We’re talking about what kind of evidence would support staying away from people if we’re at an event with them. Exercising caution if we have to deal with them. Warning other people to exercise caution around them. Not inviting them to speak at conferences. Not attending conferences, or speaking at conferences, where they’re speaking. Not buying their books. Not continuing to cite them as shining examples of skepticism at its best. In the most serious case, we’re talking about what kind of evidence would support firing someone. (And yes, for the record, I would want more evidence to support firing someone than I would to support not inviting them to conferences.)
This is a generally well-understood principle. The severity of the consequences affects how much evidence we need to believe an accusation. If several of my friends tell me, “Hey, your friend is a creep, they kept cornering me at your party,” and one person tells me, “Hey, your friend is a serious creep, they cornered me at your party and groped me”… that’s not going to be enough evidence for me to call the police, but it sure is enough evidence for me to stop inviting that person back to any more parties. Even our legal system has different standards of evidence for different situations: there’s a higher standard of evidence for criminal charges, for instance, than there is for a civil case. And the court of public opinion, and of of personal opinion, have different standards as well. Which they should. The standards shouldn’t be trivial, or non-existent — and for accusations of sexual harassment, abuse, assault, and rape, they should be pretty darned high. But there is a wide, wide world between “These accusations could lead to a conviction in a court of law,” and, “These accusations are entirely without merit.” It is a huge mistake to treat these as the only options.
So. Think about the accusations that are being made. Think about the fact that sexual harassment, abuse, assault, and rape are, unfortunately, very ordinary. Think about the rarity of false accusations. Think about what kinds of consequences are being considered here. And perhaps most importantly, think about what kind of evidence you’re actually likely to see with sexual harassment, abuse, assault, and rape… and whether you’re seeing it here.
As of this writing, August 12, 5:21 p.m. Pacific time.
In the Ben Radford situation: There is an email trail. There is independent corroboration from more than one person, who witnessed the behavior or who Stollznow told about it. There is the acknowledgement from CFI, after an investigation from an investigative firm that they hired, that Radford behaved inappropriately at conferences, and harassed Karen Stollznow with unwanted correspondence.
In the Michael Shermer situation: There are multiple reports from different people. There are other people saying that the victim told them about the harassment/ assault shortly after it happened. There are other people corroborating behavior that falls short of harassment/ assault, but is consistent with it (in this case, Shermer getting the person very drunk while flirting with them).
In the Lawrence Krauss situation: I can’t say anything about that right now, because the blog posts reporting on the accusations against him have been taken down, apparently under threat of lawsuits. If you’ve been following the story, you can probably remember what was reported before it was removed, and you can look at these questions — are there multiple claims from different people, are there other people saying that the victim told them about the harassment/ assault shortly after it happened, are there other people corroborating behavior that falls short of harassment/ assault but is consistent with it, is there any sort of paper trail or email trail — and decide how you would answer them.
UPDATE REMINDER: There is now a timeline of the major events in these accusations, and the responses to them, on Jason Thibeault’s Lousy Canuck blog. It includes several additional reports of harassment and sexual assault, and several additional pieces of corroboration of these reports. It is being updated as new information comes in and as new events unfold.
I’m not asking what verdict you’d come to if you were on a jury. I’m not asking what you’d decide to publish if you were the editor of a journal. I’m asking you to pay attention to the difference between skepticism and denialism. And I’m asking you to not be a denialist.
Being a good skeptic doesn’t only mean knowing when to reject claims. It means knowing when to provisionally accept them. It means not demanding more evidence for sexual harassment, abuse, assault, and rape than you would for Bigfoot. It means not continually moving the goalposts of what kind of evidence you’ll accept to believe these reports. It means not telling victims who don’t name names that their vague accusations can’t be taken seriously… and then telling victims who do name names that they’re just trying to ruin reputations, and shouldn’t make public accusations outside of a courtroom. It means not saying to religious believers, “No, I can’t prove with 100% certainty that there is no god, there’s almost nothing we can prove with 100% certainty — but based on the available evidence, I can conclude with a reasonable degree of certainty that there is no god”… and then saying to victims of sexual harassment or rape, “Can you absolutely prove that it happened?”
Skepticism is not denialism. Don’t be a denialist. This shit is too important to be in denial about.