I’m going to try a different approach to Sam Harris’s accusations. Since one of the problems with grappling with the objectionable ideas Harris has thrown out is that they’re fuzzily presented and laced with caveats to hide behind, I’ll just state my position as clearly as I can on a couple of the contentious issues, and why I think that way. Maybe contrasting them with Harris’s arguments will at least clarify the differences.
I found this the least of Sam Harris’s objectionable views, so I’ll put it first. He doesn’t like blogs. I do. I think they’re a fine way to engage public discussion, pro and con.
It is difficult to overlook the role that blog comments play in all this. Having a blog and building a large community of readers can destroy a person’s intellectual integrity—as appears to have happened in the case of PZ Myers. Many people who read his blog come away convinced that I am a racist who advocates the widespread use of torture and a nuclear first strike against the entire Muslim world. The most despicable claims about me appear in the comment thread, of course, but Myers is responsible for publishing them. And so I hold him responsible for circulating and amplifying some of the worst distortions of my views found on the Internet.
What, exactly, is so destructive of of a person’s integrity about having a large community of readers? I know that Sam Harris has a large community of readers, too; quite likely larger than mine. I also know from the comments that they’ve left here and on twitter that they can be quite sycophantic. What is the secret to the apparently obvious incorruptibility of Sam Harris?
As for what I’ve said, I’ve never addressed the subject of using nuclear weapons against the Muslim world (I’m against it); I haven’t said that Harris advocates the widespread use of torture; I do think he’s racist in his thinking, but then, we all are. That means that if my commenters have expressed despicable distortions of his views in my comments, they must have gotten them from somewhere else. Most of them, apparently, have gotten these ideas from reading Sam Harris’s books. Therefore, the person who should be held responsible is…Sam Harris.
Now also, if he’d bothered to read these comments with an unbiased eye, he’d notice that there are people commenting here who detest Sam Harris and everything about him; some who like some of his ideas, and like others; and others, especially since he recently linked to me, who are vigorously defending him. This isn’t a propaganda organ. People are arguing over the issues in those comment threads. A Sam Harris opponent could also claim that I was circulating and amplifying defenses of the odious views of Sam Harris.
This is the reality of open discussion: you don’t get one view. Not everyone agrees with you. Not everyone agrees with me, even on my own blog.
And yes, I’ve been getting email telling me both that I’ve been too hard and too soft on Harris. I’m more inclined to agree with the latter view right now.
I categorically reject it: torture should never be used, under any circumstances, because a) it corrupts the institutions that allow it, b) violates the rule of law, and c) freakin’ doesn’t work.
Sam Harris gave us an independent, straightforward explanation of his position:
Predictably, this article refers to the fact that I have discussed the ethics of torture in the past—and it does so in order to brand me as a moral lunatic. From reading this piece, and hundreds like it, one would never imagine that my position on torture is more or less identical to the one prescribed in that handbook of evil, the Stanford Encyclopedia of Philosophy. (Read the entry on torture there, especially the section entitled “The Beating,” and then tell me that being categorically “against torture” is a morally uncomplicated stance to adopt.)
I thought the Stanford Encyclopedia of Philosophy article was terrible. It’s largely anti-torture, but goes out of its way to justify “one-off emergencies in which the use of torture is morally justifiable”…which is what I think Harris’s position is. My problem is that every time these mysterious “one-off emergencies” are trotted out, they’re either wildly unlikely hypotheticals or awful after the fact rationalizations of brutality.
Take “The Beating”. It’s a horrible little story: car thief takes a car with a child in it, abandons it, and is shortly arrested at a railway station. According to the account, it was 20 minutes between the time the car was stolen and the time they had the thief in a truck on the way to the police station. There’s urgency in getting the perpetrator to say where the car was abandoned, because a child left in a car on a hot day is at risk. That urgency is used to justify beating the guy into semi-consciousness to get the information.
Part-way through his umpteenth, “It wasn’t me”, a questioner clipped him across the ear as if he were a child, an insult calculated to bring the Islander to his feet to fight, there a body-punch elicited a roar of pain, but he fought back until he lapsed into semi-consciousness under a rain of blows. He quite enjoyed handing out a bit of biffo, but now, kneeling on hands and knees in his own urine, in pain he had never known, he finally realised the beating would go on until he told the police where he had abandoned the child and the car.
My first thought, though, was “wait — they caught the guy on foot shortly after the car was stolen. It isn’t going to be in the next county — they know it’s going to be near the station. Why don’t they look for it?” I read this story as one of the police enjoying an excuse to beat someone to a pulp, rather than just driving around nearby streets to find an abandoned car with a kid in it. Instead of doing police work, they got to play the role of thug. The story would have also had a very different ending if, instead of semi-consciousness, the thief were rendered completely unconscious, or dead. This is not an example to laud or use to guide your ethics, except as a bad example.
It’s also interesting that at the end of the article, it’s said that if torture is allowed in emergencies, the torturers “should resign or be dismissed from their position”. Were they? This ethical consequence is left out of the story.
But this isn’t my main complaint with the article. It’s a fundamental flaw in all the pro-torture arguments: they all assume that torture works, that it is an effective way to extract the truth from someone. I couldn’t believe that the Stanford article didn’t even consider this basic assumption, that you can get useful information by way of torture, and I read it a couple of times to see if I’d missed it — but it really does assume throughout that torture is just a distasteful way to get an answer. All you have to do is beat the car thief long enough, or start pulling the fingernails off the terrorist who planted a nuclear bomb, and eventually they’ll tell you the truth.
It seems to me, though, that what torture is good for is getting the tortured to tell the torturer what he wants to hear. That’s what it has always been used for; truth doesn’t come into play. If the car thief were determined to cause the child harm, he could have sent them off to a random address to get the beatings to stop; if the terrorist were a fanatic determined to wreak destruction at any cost (could that be?), he could send the interrogators off on all kinds of wild goose chases while the timer ticks down.
So this is a difference between us: we both oppose torture, but I’m incapable of finding a single convincing counter-example, and he can.
I think that racial profiling is inconscionable and useless. It punishes the innocent for the crimes of a few, it creates new loopholes that a determined terrorist can exploit, and it generates a false sense of security by artificially exonerating whole classes of people for reasons that have nothing to do with their likelihood of committing a crime. It’s also a case of chasing the past without looking to the future; I’m hoping the people responsible for security aren’t sitting there thinking terrorists are idiots and that they’ll just keep doing the same thing, with no creativity in their planning at all.
I can see where behavioral profiling would be useful in some cases; if someone is looking nervous, or is engaging in unusual behavior, sure, check them out a little more carefully, figure out what’s wrong. But that’s not what Sam Harris is talking about. The giveaway is his frequent excuse that it’s not racism because his version of profiling would also target him. Why? It’s not his behavior. I’ve seen Sam at the airport; he’s calm and confident, like always, and doesn’t have any unusual tics; he doesn’t carry a suitcase gingerly, like it contains a bomb that might go off. He’s probably the last person I’d imagine to start praying or haranguing the crowds about their damnably heathenish ways, nor is he going to unfurl a prayer rug and start ululating. There aren’t any behaviors that he exhibits that aren’t the same as the thousands of ordinary businessmen in nice suits milling about in the terminal.
So let’s dismiss these demurrals that he’s not talking about racial profiling. Of course he is. He thinks he’d be likely to be singled out for scrutiny because he looks like he’s of middle-eastern descent. The fact that he’s willing to bear extra examination is nice and socially responsible, but it doesn’t matter: it doesn’t improve our security to have Sam Harris and many other people given preferential rigor.
Now I’ve adressed the profiling issue multiple times: my initial post, some further comments, a response to his rebuttal (where I again point out the absurdity of thinking it’s not racist if it targets yourself), and my appreciation of Bruce Schneier’s arguments. I think I’ve been clear: I do not approve of racial profiling at all, and neither does a real security expert. It doesn’t work!
I won’t say more. But it seems only fair that I leave the last word to one of my “growing army of trolls”, Marcus Ranum, who has a security background and knows a little bit about this subject.
Schneier points out:
(TSA screeners can’t sort based on religion; they have to sort based on something they can detect. And since there’s no such thing as “looking Muslim” — it’s a belief system, not an ethnic group — they’re going to sort on something like “looking Arab,” whatever that ends up meaning.) Then, you’re going to have to analyze the resulting security system. How does it work, and how does it fail? What’s the false-positive and false-negative rate? (You’ll have to do some theoretical analysis, at the very least refuting current research.)
And that’s it, right there. The rest of the debate is just noise. And you gotta hand it to Bruce, he included a link to “Strong Profiling is Not Mathematically Optimal for Discovering Rare Malfeasors” in PNAS ( http://tinyurl.com/cjcbc96 ) I believe he included that reference in “Liars and Outliers” – his latest book. It’s quite good; I recommend it. Though it’ll maybe make the philosophers and social scientists scream.
Harris’ response to Bruce’s point is pathetic. First he says You have delivered a litany of concerns about profiling that are (in my view) easily answered. and then proceeds not to answer them. Instead he goes off on a tangent about how islamic terrorists have clearly stated their intent and are not shy about talking about their plans in public. Harris ignores the fact that generally terrorists don’t discuss their plans while they are waiting in the security line. Harris continues to ignore (I can’t believe he’s stupid, so I assume he’s arguing in bad faith) the point that you can’t identify a muslim visually unless they are carrying a sign. He then side-tracks about the base-rate fallacy. Ouch, this is really bad.
Bruce isn’t a debater, BTW (though he kicked my ass at the RSA crypto commons this spring when I debated him about software liability) he’s too honest and he’s mostly concerned with educating people, not winning. I am not impressed with Harris’ honesty in this debate.
Harris digs in deeper by arguing about the Israeli behavioral profiling process which is a false equivalence to “religious profiling” or “racial profiling” because, yes, you actually can tell if someone is sweating or clutching a detonator or holding their bag extremely gingerly or the peroxide in their coke bottle is eating a hole in their hand… Most security experts, BTW, are pretty impressed with El Al’s security screening process but will say in the next breath, “… but it doesn’t scale.” I have said that I don’t know how many times, myself.
Schneier tries to get Harris to stick to the point:
That’s behavioral profiling, completely different from what we’re discussing here. I want to stick with your ethnic profiling system.
And Harris dodges it:
Well, I disagree. And the Israelis, who are generally credited with being the masters of behavioral profiling, appear to disagree as well. A person’s behavior can only be interpreted in context. What does a man’s sweating profusely and looking agitated mean? It means one thing if he is a morbidly obese senior from Alabama traveling with his wife and their church group, who is struggling to get all the trinkets he purchased in Jerusalem into a bursting suitcase; it means another if he is a 23-year-old man traveling on a Pakistani passport who is doing his best to not make eye contact with anyone. The distinction between behavioral profiling and everything else that can be noticed about a person is a myth. However, we can table this issue for the time being.
All the things above – passports, ticket purchases, luggage, etc, etc – are legitimate profiling techniques because they actually are something you can decide on.
OK, I’m going to stop here. I had a few bits of respect for Harris going into reviewing the debate closely and now I see that not only is he wrong, he knows it and is dodging the topic and playing debaters’ games rather than arguing in good faith. That’s pathetic. And Bruce is too nice to slam him for it.
As Bruce tries to point out in the debate, the whole profile process revolves around criteria that can be decided – because if they can’t be decided, they can’t be used. Then, once you’ve decided, you can look for correlations. If Harris was being honest he’d say “people with hooked noses” (or whatever the stereotype muslim he has in mind) and then security people could determine whether or not hooked noses are a decent metric. I suspect we know the answer to that. Based on 9/11, there are certain criteria that are searched for: one-way ticket, recently purchased, passport from a certain country, no frequent flier miles, etc.