A follow-up on my post about racism in law enforcement

Some people have reacted to the post I published on racism in law enforcement earlier this week and the conversations I had with them have convinced me that I should write a follow-up to clarify some points.

First, some took issue with me because I said that blacks are more likely to offend than whites, which they claim neglects the fact that, for instance, a white man who lives in a trailer park is more likely to offend than a college-educated black woman. Of course, this is no doubt true, but nothing in my post suggested otherwise. When I say that black people are more likely to offend than white people, I just mean that, on average, black people commit more crimes than white people. (If you don’t think it’s true, I will have more to say on this soon, so watch this space.) This fact was relevant because the argument I was criticizing purports to establish that it’s not the case. Obviously, there are plenty of factors that have nothing to do with race per se which explain why blacks offend at a higher rate than whites, but I made no claim whatsoever in my post about the causes of this fact. (As I said in my post, it’s almost impossible to have a honest conversation about race in the US, especially with the people who claim to want one the most. It would be nice if, first of all, one could make true claims without being called out because people immediately assume they have some nefarious agenda.) However, in many cases, the characteristics that are more directly causally related to criminal behavior are not easily observable. Thus, as long as race is correlated with criminal behavior, it can be rational for cops to use it as a proxy, even if correlation doesn’t correspond to causation. As I noted in my post, it doesn’t mean that they should, but that’s another question.

However, as I also said in my post, the question of whether racial profiling is morally permissible is not as obvious as many people seem to think. In particular, if you are a consequentialist, it’s going to depend on a lot of empirical facts that are not easily determined. One must know how efficient different forms of racial profiling are, how much they affect social trust, etc. Even if one can do that, one also must determine how to balance those considerations, which is hardly obvious. In a comment to my post, newt0311 brought up what he calls “neighborhood profiling”, i. e. the practice of allocating more police officers to neighborhoods where crime rates are higher. As he pointed out, given the degree of segregation in American cities and the crime rates in black urban enclaves, the effects of that kind of policy are in practice very similar to more straightforward forms of racial profiling and, therefore, it should arguably be regarded as a form of racial profiling. But I think even people who think it’s obvious that more straightforward forms of racial profiling are not morally permissible would agree that it’s not as obvious in the case of this indirect form of racial profiling and, if they don’t, it’s because they haven’t spent enough time thinking about it.

A friend also claimed that, regardless of whether this kind of practice is morally permissible, there is no question that it’s illegal. Since I’m not a lawyer,  I have no idea whether this is true, which is why I was careful not to talk about the legal aspect of this debate. Other people have claimed that, while the fact that the hit rate for whites is higher than for blacks doesn’t imply that the offending rate for blacks is lower than for whites, together with the fact that cops stop more blacks it implies that they are being inefficient. However, as I explained in my reply to kevin’s comment, even this weaker claim is false. Depending on the details, the fact that the hit rates for whites is higher than for blacks even though cops stop more blacks may give you a good reason to think that cops are not efficient, which in some cases may justify the accusation of racism, but you need more information before you can make that accusation. The same friend who brought up the legal question also made the even weaker point that, if the hit rate for blacks isn’t higher than for whites, it suggests that cops could probably afford to stop black people less often without drastically reducing their efficiency. I think he is probably right that, in many cases, whatever adverse effect this might have on efficiency, it might be more than compensated by the gains in social trust resulting from such a change. But this isn’t the argument I was attacking and, if that’s what liberals want to say, then they should say that instead of making false claims resting on fallacious arguments.

I also said in my post that, although liberals often insist on the dangers of having a honest conversation about race (of course, that’s not how they put it, but that’s what they mean), they tend to ignore the danger of not having such a conversation. (I wish I didn’t have to use the word “conversation”, which has been deformed beyond recognition by social justice activists, but it’s the most economical way of saying what I want to say.) I have been asked what I meant by that, so I wanted to conclude with a few words on that issue. Anybody who follows the scientific debates that have to do with race knows that it’s a radioactive issue. As a result, people are reluctant to say things which they believe to be true and, in many instances, won’t even go near a topic which could lead them to say non-PC things. (For instance, see this story about the problems 3 economists ran into, after they published a paper in which they concluded that the racial gap in the US could be explained in part by the fact that minorities spend a greater part of their workday not working. To be clear, I’m not endorsing the study in question, since I haven’t read it. I’m just using it to illustrate the fact that, in the current atmosphere, even social scientists have very good reasons not to investigate some issues.)

It should be obvious why this is dangerous, because unless we can understand the causes of the problems faced by many blacks in the US, it will be difficult to solve them. In particular, since this is what I was talking about in my post, it may prove difficult to fix the racial disparity in incarceration if you overestimate the importance of discrimination in the criminal justice system and I think liberals definitely overestimate it. For instance, it could lead you to waste money and squander political capital to equip cops with body cameras, when these resources could perhaps have been better used to deal with e. g. housing segregation. (I use that example because I personally think that it’s the single most important factor in the problems faced by the black underclass.) It may also be that, when body cameras don’t have any significant effect on racial disparity in incarceration (which they won’t), you will find yourself with a bunch of nice white liberals (the same who give me shit for saying things which shock their delicate sensibilities) who will not be willing to address the real problems (such as, again, segregation), because they have been led to believe that discrimination was the main issue and they think that, having paid to equip cops with body cameras, they have done enough. (I know many people think discrimination plays a major role in housing segregation, but I disagree and plan to write a post to explain why at some point.) So liberals should stop behaving like Torquemada with anyone who says something controversial about race.