So whatever happened to the “Believe Women” mantra, brought to us care of #MeToo? Yesterday’s unqualified axiom seems to have been washed away by today’s intra-progressive controversy. The reasoning here seems to be: Elizabeth Warren accused Bernie Sanders of sexism. But Bernie is more progressive than Liz. So the accusation can’t possibly be true, because if it were true, its truth would ruin the most progressive mainstream candidate’s shot at the presidency. Hence the accusation must be false, and Elizabeth Warren is a bit of a bitch for making it. From which it follows that the “Believe Women” axiom must also be false, though we’re not to say so out loud.

Gee, that was easy. Who knew that moralized axioms could so lightly be adopted, and so lightly be cast aside? 

What makes this dispute particularly ludicrous is the widespread assumption that it’s “sexist” to predict that a woman can’t be elected president. It certainly seems an unwise prediction, given Hillary Clinton’s having won the popular vote in 2016, and Elizabeth Warren’s respectable polling numbers now. But how is it sexist? The prediction (if it was made) doesn’t say anything about the talents or abilities of women. It speaks to the presumed sexism of the American electorate. Maybe it’s an uncharitable assessment of the American electorate (if there is such a thing), but it can’t be sexist to judge the American electorate sexist, can it? Because if they’re not sexist, but sexism is a pervasive problem, where exactly does it reside? In non-citizens and minors who can’t vote? But set this whole issue aside. The controversy is too far gone for anyone to be raising questions at this point about its basic coherence.* 

Speaking of coherence, and getting back to my real point, I so far have not encountered a single justification for “Believe Women” that makes any sense, at least in the original unvarnished sense associated with the slogan. Here’s a charitable reconstruction of the hypothetical argument for it:

1. Women have often been on the receiving end of sexism, up to and including sexual assault and rape. Their accusations have often not been believed. Because they haven’t been believed, but have often been true, many miscarriages of justice have taken place, with men escaping justice for their malfeasances.
2. False accusations of men by women are rare.

I don’t dispute (1). I would dispute (2), but let it go. False accusations are not the kind of thing that get accurately reported (any more than sexual assault itself), but I don’t expect people who’ve never been falsely accused of anything to grasp this. So I’ll just throw my own arbitrary pseudo-statistic out there: people who have never been falsely accused of anything aren’t rare.

Suppose we accept both (1) and (2). What is the combination supposed to entail? How is it supposed to entail that women have achieved the epistemic status of cops and prosecutors, whose accusations, by virtue of being accusations made by people of a certain privileged description, regularly trump the presumption of innocence in a court of law?** Because if trumping the presumption of innocence is not what “Believe Women” means, it’s unclear what it means. Not that attempts haven’t been made to explain it. But unless we just believe the attempts because they’ve been made, I’m not inclined to think that they succeed.***

I suppose we could add one ad hocery to another and say that “Believe Women” was only supposed to apply to cases of sexual assault, not sexism. Since Warren wasn’t alleging assault, we needn’t believe her. But who believes this? And why would they?

The problem presented by (1) is unquestionably a problem. But it isn’t unique to the situation of sexual assault against women. It’s true of any situation where one person is alleged to have violated the rights of another; where the putative victim makes an accusation but lacks the proof to convict; and where a miscarriage of justice takes place as a result. That can happen, and often does happen. But Believe Accusers is not a legitimate response to it, and certainly not one when fashioned for the purposes of one specific sort of accusation to the exclusion of others.

Ironically, one situation where it happens–where one person makes a difficult-to-prove allegation of another–is precisely when one party enjoys an automatic presumption of veracity by virtue of making an accusation. Think of the doctrine of qualified immunity in the case of the law enforcement. It’s precisely because the police enjoy this immunity that they lie so much, and it’s because they lie so much but enjoy a presumption of veracity that they’re responsible for so many rights violations. To hand this power over to women is not an improvement but a form of moral regression.

Feminism is supposed to consist in “the radical notion that women are human beings” (see “Marie Shear”). But if that’s so, maybe we should draw the equally radical corollary that they act like human beings: given a chance to abuse unaccountable power, some of them will often abuse it–in just the way that men perennially have. You might have thought that the likes of Kellyanne Conway, Sarah Sanders, Stephanie Grisham, and Katrina Pierson had resolved this dispute, but alas, no. (Or Hillary Clinton, for that matter.)

The deep moral question posed by Believe Women is: can we permissibly sacrifice the innocent to the desire to rectify the miscarried justice of the past? The deep epistemic question is: if we assume that the aggregate incidence of false accusation is low, can we prima facie infer in a given case that a given accusation is likely to be true?

My answer to both questions is “no.” But even if the answer is “yes,” it’s not yes only in its application to women (or even more narrowly, to women accusing men of sexist malfeasance). It’s yes because we should believe accusations for being accusatory. I hope you find that as implausible as I do. What makes Believe Women such a disingenuous slogan is its willingness to poison the well by invoking sexism against the sheer existence of doubt: if you reject the slogan, the implicit accusation goes, you endorse sexism, or even worse, sexual assault and rape. What if you’d just like to be careful about accusations? I hate to sound melodramatic, but a world in which this question is treated as prima facie suspect is a world in which justice has no chance at all.

To end on an irenic note: in fairness to the slogan, its more conscientious adherents are now modifying it in a more plausible direction, suggesting that “Believe Women” should not be used as an excuse for rumor mongering or ignoring inconvenient facts. And the spirit of the slogan is laudable and true–a correction of centuries of sexist oppression and lies whose victims have mostly been women (and whose perpetrators have mostly been men). But that’s the problem with slogans: first they oversimplify, then they require modification in the direction of reality, and then, after the damage is done, they reluctantly get it. We can now expect a PR campaign suggesting that Believe Women never had the original, utterly implausible meaning it obviously did have: believe women to the point of sacrificium intellectus. A change in meaning will certainly be a change for the better. But it will also mean leaving some falsehood behind. And admitting it.

*By the way, two years ago, it wasn’t sexist to make the prediction, at least not at Politico. It was being made by feminist academics. I won’t bother to dig them all up, but with the rise of Warren, Harris, and Gabbard in the polls awhile back, there was a spate of articles in the press wondering whether a woman was in fact electable to the presidency, some optimistic, some pessimistic. The debate was not over womens’ fitness for the presidency, but their electability. Until the recent debate, no one thought that the pessimists were sexists. On the contrary: the pessimists were regarded as more realistic about the prevalence of sexism in American society. Elizabeth Warren has single-handedly turned this plausible assumption upside down: apparently, it’s now sexist to suggest that America is sexist.

**Because why would a cop lie? Don’t laugh. This is the rhetorical question I’ve repeatedly heard from judges–both male and female–in municipal court proceedings. Judges will look directly at defendants who accuse cops of lying, and tell them with a straight face that they can’t imagine that Officer So-and-So would perjure himself. Then they infer that because they can’t imagine it, he must not be lying. I’m not aware of any study that tabulates how often this happens. Should we then infer that it almost never happens?

***It’s hard enough to come by a statement of the claim itself that makes any sense. (Here’s a notable exception.) From the Wikipedia entry:

“Believe women” is an American political slogan arising out of the #MeToo movement.^[1] It refers to the perceived necessity of accepting women’s allegations of sexual harassment or sexual assault at face value. Sady Doyle, writing for Elle, argues that the phrase means “don’t assume women as a gender are especially deceptive or vindictive, and recognize that false allegations are less common than real ones.”^[1]

“S‘s face value acceptance of testimonial evidence that p” could refer to two completely different things: (1) S‘s acceptance that p could very well be true as stated, despite strenuous denials of p; or (2) S‘s acceptance that p is true simply because it was stated despite any evidence in its favor beyond that fact. “Acceptance” also has to be distinguished from “taking seriously.” You can take a claim seriously but regard it as false.

If “Believe Women” refers to (1), I don’t dispute it. But if it refers to (2), I reject it. Sady Doyle’s version of the thesis (in the Elle article mentioned above) is a conjunction. The first conjunct is entirely true (both the prescription and the descriptive claim embedded in it). The second conjunct may or may not be true. Whether true or not, there’s nothing culpable about doubting it, and it has no clear implications for the inferences one draws in a given case.

The logic behind “In case of conflict, believe female testimony over male testimony” is no different from that of racial profiling: “When in doubt, assume that a young black male is doing something sufficiently suspicious to justify a Terry stop.” Both involve inferences from (what are alleged to be) higher aggregate propensities for a given behavior to action based on suspicion in a given case.

If I learn that black students generally have lower SAT scores than white students, and am told that SAT scores predict academic success, should I then infer that a given black student is unlikely to succeed in my class, or as an advisee of mine? Would it really help things if I were to be blackmailed into the conclusion (so to speak) by being told that by setting the aggregate statistics aside in the individual case, I was confessing my hostility to academic excellence?

It’s notable that Doyle seems not to have any problem with profiling: false accusers, we’re told, either fit a profile, or evaporate into insignificance. What if the data sets, derived from “academic studies, journalistic accounts, and cases recorded in the US National Registry of Exonerations” are incomplete–just as is (correctly) assumed of data sets regarding sexual assault?