Friday, November 18, 2016

Evidence by Curated Anecdote

Suppose I have a bag with a hundred balls in it. The balls are made either of wood or iron; I don't know how many there are of each, but I know that there's at least one iron ball and at least one wooden ball. I start out with no idea whether there are more iron balls or more wooden balls.

Then I examine a ball, and see that it's iron. How should this impact my estimate of the number of iron balls? It's tempting to think it obvious that my estimate should go up—I've seen an iron ball, so that's at least some evidence in favour of there being a lot of iron balls in there.

But not so fast. It all depends on how the ball was chosen. It depends a lot on how the ball was chosen. If the ball was pulled out at random and proved to be iron, that'd be evidence for more iron in the bag. But what if it was pulled out with a magnet? Then the selection method guaranteed it was going to produce an iron ball. And I already knew there was at least one iron ball in there. So the fact that I got an iron ball gives me no new information whatsoever. My estimate of how much iron there is in the bag should change not at all, if that's how the iron ball got selected.

I think something like this situation, and its attendant tempting fallacy, applies in much more real-world cases with morally significant implications, too. In particular, I'm thinking today about estimates of the prevalence of bad behaviour among certain demographic groups. For example, there are many people who, when thinking about anti-Trump protesters, think of unruly mobs of violent people. There are also many people (mostly different ones) who, when thinking of America this month, think of people committing hate crimes against Muslims and black people.

If challenged about why one should think anti-Trump protesters tend to be violent, or that there are a lot of hate crimes in America, people tend to point to examples: news features about anti-Trump protesters attacking Trump supporters, or about threats of lynchings against black teenagers. I think that most of the time, such news stories are terrible evidence for what they're being used as evidence for. Even setting aside questions about whether the stories are true—let's assume they are—they do not actually provide any evidence at all for their general conclusion. The reason for this is that the cases are structurally analogous to the magnetic drawing method described above.

The world is a big place, full of all kinds of people. So even before we do any serious investigation at all, we know that there are some people who commit hate crimes against black people, and that there are some anti-Trump protesters who are violent. We should all agree that this is obvious. There are some people like that. We disagree about how prevalent these things are, but we all agree they exist.

We also know some things about the media. Namely, that, since lots of people are interested in reading about violent anti-Trump protesters and hate crimes, various media sources will be motivated to find and report on at least some such cases. Furthermore, the media are good enough at finding these things that it's nearly certain they'll do so.

In other words, the antecedent probability of there being some cases in question is practically 1, and the conditional probability of the media reporting on them, supposing they exist, is also practically 1. So when you read about a Trump protester being violent, that should increase your estimate of how violent Trump protesters on the whole are by practically zero. It's like the magnet pulling out the iron ball—it was going to find one, no matter how many there were, so the fact that it found one does not make it likelier that there are lots.

Things would be different if the media worked very differently—if, for example, it picked Trump protesters at random, and then reported on what they were like, no matter what they found. If that is what happened, then reading reports of violent protesters would be significant evidence. But that is very unlike the way our actual media works.

Our own anecdotal experiences actually work a bit (a bit!) better in this respect. If the anti-Trump protester you happen to be standing next to starts beating somebody up, that is evidence in favour of violence in that group. (It wouldn't be if in advance you'd somehow implemented a strategy of standing next to the person you thought likeliest to be violent.) For what it's worth, I spent several hours at anti-Trump rallies in Philadelphia last week. I observed no violence.

I haven't seen comparative numbers about the frequency of hate crimes in America over the past week. As far as data goes, that would be the gold standard. But I am pretty confident they've gone up, probably by a lot. This isn't based on the many news stories I've read about examples—the USA is a big enough place that it's not implausible to me that there are dozens of such cases every week, and that they're being reported and disseminated more now. But I have more specific, more personal experiences that are a bit more similar to random sampling. I haven't been victimized myself, but I do personally know someone who was physically attacked and racially insulted. And, at the university where I happened to be visiting last week, black students had that very day been threatened with lynchings.

This isn't super definitive data, but I think it's a lot more telling than lists of stories turned out by media outlets motivated to turn out lists of stories. The main point is, if you want to know how much something supports a given hypothesis, it makes a huge difference how you found it.

Monday, November 07, 2016

The Mikado in the 21st Century

Last week I participated as a guest of the New York Gilbert & Sullivan Players in a forum discussion on racism in The Mikado. After cancelling their 2015 production over concerns about racism, they've hired some new staff and brought in an advisory panel to put together a new version of the show. Because of my background as a performer and consumer of Gilbert & Sullivan productions, my own Japanese-American identity, and some things I've written (though not publicly) on the subject before, I was invited to give a keynote address as part of that forum.

I thought it was a productive conversation—I had the sense that all participants (which included some NYGASP officials with significant responsibility about the new production) were well-motivated and sincere, and had some understanding of the issues. Unfortunately, we didn't have time to take audience questions, and lots of things didn't really get addressed. But I thought the forum was an encouraging step. I have found that these conversations are extremely difficult to have. But for at least a few hours last week, we were having a real one.

For anyone interested, here are some of the things I said, along with some other things I hoped to say but didn't have time for.

Thursday, October 20, 2016

Donald Trump: trial balloon or vaccine?

Let's assume that Donald Trump is about to lose the Presidential election. This seems overwhelmingly likely today, but there have been times when a President Trump looked like a real possibility.

538's polls-plus winning probability over time.

Donald Trump has run an incredibly racist, xenophobic, and misogynistic campaign; he's elevated to prominence a dark underbelly of US culture. This has been clear from the very start of his campaign. Unfortunately, this by itself didn't stop him from being a halfway viable candidate with a real chance to win. The reason this race got away from Trump is that he is a totally inept politician. He has no personal discipline or ability to stay on message. Even recordings of his bragging about sexual assault, and over a dozen women who have accused him of such actions, do not seem definitively to have shut him down as a viable candidate; among his base, there seems to be a widespread willingness to forgive and forget. A skilled politician who performed real remorse might conceivably have gotten similar favourable treatment from a decent number of middle-of-the-road Americans. People are surprisingly and depressingly fuzzy about a lot of these questions.
Numerous serious allegations of sexual assault would be damaging—perhaps fatal—to any candidate in 2016. But Trump certainly did himself no favours.

Fortunately for civilisation, Donal Trump is a disaster of a politician who will lose an election unlikely to be even close. But the Trump candidacy does raise a sobering question: what if he hadn't been such a blatantly repulsive specimen of humanity? What if he had the discipline and common sense to stay on message and employ more moderately-palatable spin against his unpopular opponent? What if someone with Trump's politics had a gentle face and a kindly voice?

I had a chance to visit the Johannesburg Aparteid Museum this past summer. One of the striking and disturbing things I saw there was footage of South African Prime Minister Hendrik Verwoerd smiling and cooing the most horribly racist sentiments behind a friendly and grandfatherly tone.

As I remarked to my friend at the museum, if Trump had looked and sounded like that, I think Clinton would have had very little chance in this election.

The thought of a modern-day Trump-like demagogue who knows how to dress up his racist, authoritarian demagoguery in pretty propaganda is a terrifying one. In the counterfactual world where we had that instead of Trump in 2016, running against the same unpopular Hillary Clinton, I am convinced that this hypothetical candidate would have won. It might have been the biggest disaster in American history.

What I'm unsure about is whether in the actual world, the prospects for such a future candidate are greater or lesser in the wake of Trump. One theory is that Trump has been a trial balloon. This was something of a test run, to determine what works and what doesn't work in a fascist, white supremacist candidate for US President. The alt-right learns its lessons from Trump, and comes back on another occasion with a candidate with similar politics, and similarly able to stoke its horrifying base, but who has never bragged on camera about sexual assault and who doesn't look and act like an impetuous 12-year-old when insulted. It's certainly a worrisome thought.

But here's the more optimistic possibility. The alt-right has learned a lot about the electorate this year—but so too has the electorate learned a lot about the alt-right. Given the resounding defeat Trump seems well on his way to receiving, it may be that his own disastrousness may tarnish the racist ideology itself. The future, more polished, version of Donald Trump may find himself much less viable than he'd otherwise be because he reminds people of Donald Trump. Rather than Trump as trial balloon, we have Trump as vaccine—after defeating the less dangerous version of this candidate, Americans are better-equipped to avoid succumbing to the sneakier one.

The latter, happier interpretation seems particularly plausible if Clinton runs up a big lead on Trump. Let there be no ambiguity about the disastrousness of this campaign. Let politicians be terrified for a generation or two of reminding people of Donald Trump. Here's hoping.

Saturday, October 01, 2016

Why Don't More Faculty Speak Out About Sexual Assault?

This is a (slightly modified for public consumption—links added, some personal bits and references to other conference speakers’ remarks removed) draft of remarks that I’ve prepared for a panel entitled “The Impact of Speaking Out: Graduate Students and the Role of Faculty”, as part of a conference at UBC on conversations around sexual violence.

Sunday, September 25, 2016

Platforms, Free Speech, and the Boundaries of the Pale: A Methodological Remark

Sometimes people say things in public, and other people think those are terrible things to say, and so they complain, or protest, or distance themselves, or decline to invite that speaker for something else, or apologize on their behalves. And then, as we all know, sometimes yet other people get real offended by those first other people for a perceived threat to the free expression of ideas.

This is a super common pattern. (A pretty substantial amount of public hand-wringing about Millenials falls into this category.) Philosophers will recognize a recent high-profile controversy of this form.

The argument against step two above—the argument that says you shouldn't protest or apologize or whatever—goes like this: free speech is important, even when (especially when!) it is uncomfortable and challenges your preconceptions. It would be both unjust and intellectually foolhardy to censor the expression of some ideas just because some people find them distasteful or threatening.

The argument on the other side goes like this: words matter, and the expression of some ideas is actively harmful; we shouldn't encourage or enable people to harm others, especially when those harmed are members of already-marginalized groups. No one is entitled to these particular platforms, and by the way protest speech is free speech too.

In most actual instances of the schema I come across, my sympathies are largely with the protesters. But there's something a bit funny about the level of generality at which the argument tends to take place. The details of the objectionable presented content make a pretty big difference.

For example, the principles often articulated against protesting, if taken literally, would generalize to defences of views that I think most of us would recognize as clearly beyond the pale. Like, if the keynote address at the American Society for Conservative Philosophers (a name I just made up but maybe it's real, it doesn't matter, it's a fictional case) were given by a white supremacist calling explicitly for, say, stripping the voting rights of Americans of colour, everybody would be making a tremendous fuss about it and they'd obviously be right. Maybe a few individuals would crawl out of the woodwork and say "we must be willing to encounter ideas that challenge our preconceptions" or "these ideas about racial purity have been around a long time don't you think there might be something to them?" or "are conservatives not allowed to talk about social and political questions about race?", but I think most of the people who say the parallel things about real-life cases wouldn't be much impressed. This example shows that the principles cited are not at all plausible in full generality.

I think something similar is also true of the other side. Suppose a local farmer is invited to give a commencement speech at a high school in their community, and students protest and demand that they be un-invited, on the grounds that such farming amounts to a morally indefensible program of animal suffering. I would have a certain amount of sympathy with these students. (I think most animal farming does amount to morally indefensible animal suffering.) But I think I'd ultimately be persuaded by the line that this isn't very good grounds for protesting the speech, even if the students make the case that the speech itself would cause harm.

One might try to locate the difference between these cases in the moral seriousness of the harm, or the amount of offence caused, but I'm inclined to look to more contingent social matters. Really blatant white supremacy is, in North American academic circles, rare enough and marginalized enough, not merely to offend, but to shock. It's not one of society's live options. It is beyond the pale. This isn't an intrinsic feature of the content of the view—it's a relational status to society more broadly. By contrast, the idea that it's fine to cause massive suffering in non-human animals for small human benefits is widely accepted. If one merely levelled the horrified stare (the moral cousin of the less incendiary incredulous stare) at everyone who thought it was OK to raise animals for food, one would only serve to alienate oneself from society at large. People who think farming is fine need to be convinced; people who think America is only for white people should be shunned.

The lines move over time, as some questions become settled. At the margins, there will be controversy, even among people who agree that a stance is morally objectionable, about whether it is beyond the pale. I think that many questions about the moral status of homosexuality are at present right around the margins. The idea that homosexuality constitutes a defect of some kind is rightly and rapidly becoming much more of a fringe position, but it is still widely enough held, by powerful enough people, such that many people will still be inclined to treat it as a view worthy of respectful disagreement, rather than the horrified stare.

The debates that ensue tend to frame the issue as one about free speech versus harmful attitudes towards marginalized groups. That's partly right, but I think that quite a lot of what's going on is best conceptualized as a fight about where to set the boundaries. It's a border skirmish in the culture war. If the progressives keep winning, this will all look very different in retrospect. Hopefully in twenty years we will look back with some shock at the idea that people were arguing in 2016 about whether it's OK to descibe homosexuality as a kind of disability. Maybe by 2036 the argument will be about whether we need to take seriously the idea that animal farming is morally permissible.

Sunday, May 22, 2016

Presumption of Innocence

The Universal Declaration of Human Rights has it that "everyone charged with a penal offence has the right to be presumed innocent until proved guilty according to law in a public trial at which he has had all the guarantees necessary for his defence." This is a kind of legal protection against criminal conviction—you cannot be declared by the state to be guilty of a crime, or punished for it, unless your guilt has been proven.

The presumption of innocence is rightly considered a pillar of civilised society. But people have a tendency to over-apply it in irrelevant cases. The presumption of your innocence means that the state can't punish you for a crime unless it proves that you committed it. That's it. It has nothing to do with how one individual should treat or think about another, or whether an organisation should develop or continue a relationship with an accused individual. The presumption of innocence doesn't protect you from being unfriended on facebook, or shunned at conferences, or widely thought by other people to be a criminal. It just protects from being criminally convicted.

The weird world of academic philosophy this week is digesting an article that includes serious allegations of sexual harassment by Yale professor Thomas Pogge against philosophy students. While most of the people whose reactions I'm reading are expressing disgust at Pogge's behaviour and gratitude for Fernanda Lopez Aguilar, the woman who has come forward with the story, there are some (including Pogge himself, as well as some of the People who Leave Comments on Blogs) who caution the broader world from jumping to conclusions, on grounds related to the presumption of innocence. Pogge's statement discusses "trial by internet"—but since "the internet" is not contemplating the use of state power against his liberty, whatever metaphorical sense in which he is undergoing a "trial" is not one where he enjoys the presumption of innocence.

The admonition not to pass judgement about the allegations is simply the admonition to ignore them. "Don't believe anything unless it's been proven in a court of law." But this is just a ludicrous epistemic standard. Do you care whether powerful men in academic philosophy are using their stature to coerce students into compromising sexual situations? Then you should be interested in credible testimony to the effect that this one has been. Don't be tempted by the fallacious inference from it hasn't been proven in court to you have no way to tell whether it's true. (This, incidentally, is also why it makes sense for universities to have sexual assault policies.)

Moreover, it's not in general "jumping to conclusions" to accept someone's word. We learn from other people about what happened to them all the time. Talk of courtroom standards of evidence can make us forget the fact, but in general, we can get knowledge from other people. There's a temptation to think "it's just a he-said–she-said situation, so there's no way to know what really happened". While there are of course some cases that are like that, it's not in general true that any time testimony is disputed it can't be known to be true. Sometimes someone tells me something, and I thereby come to know it, even though someone else denies it. This is especially likely when the thing told is generally plausible, and fits well with other things I know, when there's no plausible explanation about why the testifier would say it if it weren't true, and the denial is both weak and self-serving. All these features seem to me to be present in the current case (and in many similar cases).

I believe Aguilar's allegations. They are serious and credible. They cohere with other stories many of us have heard about Pogge. The asinine hypothesis that she'd make them up out of spite, or for some kind of personal benefit, is the risible product of a preposterously misogynistic imagination. I think it'd be an epistemic error not to believe them—and a moral error too. It would be derelict for us as a community to ignore that which it's reasonable for us to believe—indeed, what which I think we know.

Wednesday, May 11, 2016

What is proof beyond a reasonable doubt?

As I finish up a few long-standing projects on knowledge (check out my draft monograph!) I'm hoping to start exploring a few new-for-me areas. In recent years my philosophical discussions on this blog have been relatively esoteric, trying out a new objection to so-and-so's third paper on such-and-such; when I was in grad school I used it as an early thinking-out-loud forum about more basic stuff. If I can swallow my pride enough, I may return to that form over the next several months. One area I'm quite ignorant and naive—but also curious—about is law. In particular, I'm curious about how legal concepts like evidence and testimony interact with epistemological concepts like evidence and testimony. I'm a complete novice in this field, so if I'm overlooking things that seem totally obvious to any readers, I'd be grateful to have it pointed out, especially if it comes along with reading advice.

One of my entry-points into the literature here is my UBC colleague John Woods's book Is Legal Reasoning Rational: An Introduction to the Epistemology of Law. (I'm maybe 40% of the way through a first read of it now.) Here's one thing I've found interesting so far. Woods writes, on p. 103:
Spike cannot be convicted unless a jury unanimously finds that his guilt has been proved beyond a reasonable doubt. Certainty beyond a reasonable doubt is sometimes said to be moral certainty. Under either name, as Gifis writes, it is a conviction based on persuasive reasons and excluding doubts that a contrary conclusion can exist. This is much too strong a formulation. There is plenty of case law that allows for conviction even in the face of a perfectly reasonable case to the contrary, recognized as such by the jury. On an alternative formulation of the standard. A juror is said to be morally certain of a fact when he or she would act in reliance upon its truth in matters of greatest importance to himself or herself. Although better, this appears to omit the essential requirement of criminal proof that this readiness to act, this moral certainty, be grounded in the juror's belief that the burden of criminal proof has been met. In other words, it seems not to be sufficient for criminal conviction that a juror be in a state of moral certainty of the accused's guilt. It also matters how he came to be in that state. 
So we have these notions:

  • Proof beyond a reasonable doubt
  • Certainty beyond a reasonable doubt
  • A juror's being in a state of moral certainty
Woods slides between the first two without comment. Maybe the thought is that there is this tight relationship: proofs establish certainty. (In logic or mathematics, this seems plausible—if I have a proof of something, then it is certain for me.) But as Woods points out, it's another matter whether my psychology reflects this certainty. I might have a proof but doubt anyway; or I might be irrationally certain, even absent a proof.

I wish Woods had explained further what he had in mind when he discusses appropriate conviction in the face of 'a perfectly reasonable case to the contrary'. (I am not sure who Gifis is—I don't see a citation.) Can there be 'perfectly reasonable' cases that have been proven not to be the case? If 'provability beyond a reasonable doubt' is closed under deduction, then it seems like there must be. (If it's been proven beyond a reasonable doubt that Jones committed the crime, then it's provable beyond a reasonable doubt that Jones wasn't framed.)

The alternative formulation mentioned seems to target the psychological state of being in a state of moral certainty. This feels at least analogous, maybe more, to the epistemological notion of belief or commitment. Some epistemologists, like Jennifer Nagel and Brian Weatherson, have emphasized that whether one is in this state depends (causally or constitutively, respectively) on the practical situation—when the stakes are high, one is less likely to believe. Maybe moral certainty is something like, being such that one would count as believing even in very high stakes cases. And although this state isn't, as Woods points out, sufficient for appropriate conviction, a normative connection seems plausible: one should only be in this state if one has a proof of corresponding strength. Don't be sure unless it's proven.

It's hard for me not to think in terms of knowledge. One contextualist way of mapping these thoughts would have it that, given a high standard for knowledge, all and only that which is proven beyond a reasonable doubt is 'known', and jurors should vote to convict if and only if they know the defendant guilty. I think this could make sense of these issues—although it'll probably be too simple in other respects, such as the restriction of admissible evidence. (Some things jurors know, they're not supposed to consider. (Or maybe a weird standard could count them as unknown? Something to think about.))

Anyway, I know I'm being really naive here, but I'd be interested in digging in a little bit, if anyone has ideas or reading suggestions.