Bill Bennett, Abortion, and Race

| | Comments (10) | TrackBacks (1)

I do not agree with the president on everything, though I've often defended some of his less popular statements and policies, against liberals who think his views are too conservative and against conservatives who think his policies are simply not conservative. It should never be said that I defend him no matter what he does. I happen to agree with him more than any liberal or moderate would, and I also happen to agree with him more than most conservatives would. That's not because I agree with whatever he would say but because I simply think he has the right views on many things that I don't think either major party gets quite as right. That being said, I wanted to disagree with something he did that I suspect most people I know would agree with him on. Last Friday, he made a statement criticizing Bill Bennett's infamous comments Wednesday morning about abortion and its social effects. I don't think the president was right in declaring Bennett's comments "not appropriate". I've seen two complaints against what Bennett said, one completely stupid and the other based on a genuine but misguided worry, indeed a logical fallacy. This second sort of issue comes in two forms, the more prominent and more misguided one and the less prominent and a little more reasonable one that I think ultimately is still not a good reason to criticize Bennett.

What Bennett said was pretty clear and straightforward. He was responding to a caller who thought you could argue against abortion on social consequences grounds. In particular, the caller wanted to say that abortion is wrong because if we didn't have it we'd have more taxpayers, and the GNP would be higher. Society would be more productive. Regardless of whether such a claim is true, if abortion is wrong it's for much deeper reasons than that. Bennett wanted to distinguish between the mere consequences of an action and what makes it wrong, thus siding squarely on the non-consequentialist side of ethical theory (which is the right view). He wanted to make this point by showing that an action can have good consequences without being a good action. So he picked an example that most everyone would consider horrendous with consequences that, independent of other factors, would be good. Since the original example was just like that, his point would be made, and it would be a very effective way of making the point. So he chose the example of aborting all black babies, which no one in their right mind would consider a good thing, not only for the horrific element of killing a whole generation of kids (presumably without consent of either parent). What's even worse is that it singles out just black babies to kill, which is verging in the direction of genocide. Yet surely such an action would reduce crime, Bennett points out. This is thus a very good example of the sort of argument Bennett wanted to make his point. He could have made it raceless and just had every fetus being killed regardless of race, but that wouldn't have been as bad a case, and he wanted it to be very bad.

The stupid complaint against Bennett's argument is that he had described a terrible action. But that complaint is simply to misunderstand the argument. The role his example has in the argument is that it's a counterexample to a particular claim. To serve as such a counterexample, it needs to be a very bad action with a good consequence. The worse the action, the better the example. The more sophisticated complaint against Bennett understands the nature of the argument but has trouble with his consequence. They don't think it's very nice, or polite, or something to use a particular empirical fact in his argument. That empirical fact is that more crime takes place in predominantly black neighborhoods, more crime takes place at the hands of black people, and more crime is against black people than is the case in white neighborhoods, by white people, and against white people. This is a fact that's been observed by empirical study and is not disputed by anyone. There is one caveat, however, and it's an important one. Steven Levitt says:

It is true that, on average, crime involvement in the U.S. is higher among blacks than whites. Importantly, however, once you control for income, the likelihood of growing up in a female-headed household, having a teenage mother, and how urban the environment is, the importance of race disappears for all crimes except homicide. [hat tip: Juan Non-Volokh]

As far as I know, Levitt's point is undisputed. It may not be, but it doesn't matter. Bill Bennett may or may not be aware of Levitt's statement, but that doesn't matter either. His point still stands, even if Levitt is right (and I wouldn't be surprised if he is). Levitt's point is that race itself as a factor isolated from other factors does not lead to high crime. I'm not sure why anyone but a racist would think it would. I would have assumed without even looking at any other factor that it's the sociological patterns within black America that lead to the higher crime rate. Levitt seems to have isolated those. Growing up in a female-headed household, having a teenage mother, and being in a more urban environment apparently all affect the crime rate. It's not race itself. Therefore, we shouldn't conclude that race itself causes a higher crime rate. Does that mean Bennett is assuming some racist premise? I've heard Democratic commentators repeating such a mantra not a few times in the last few days. The problem is that it's a fallacious argument. What Bennett said is still true.

Bennett said that aborting all black babies would decrease the crime rate. The fact that what leads to the higher crime rate among blacks is not race does not change that fact. There is a higher crime rate among black people. That's because black people are more likely to be in more urban areas, black people are more likely to be born to a teenage mother, and black people are more likely to grow up in a female-headed household. Those factors do increase the crime rate among black people, and if you aborted a whole generation of a group who will have a higher likelihood of crime due to such factors, then it would indeed be likely that the overall crime rate would go down, other factors being equal. This argument requires no assumption that black people are by nature more likely to be criminals. The claim that Bennett must be assuming something only racists would believe is simply false. What he said makes perfect sense without such an assumption.

Now there's one slightly more reasonable complaint someone might make, and I've only seen it suggested just once. Someone might argue that everything Bennett said was true and not deriving from some racist view of blacks as criminals, but it could have a bad effect on whoever might hear it. It might promote the view that black people are by nature criminals. I don't think this is a good complaint, however, even though it's a much better argument than the other two. I don't much like any argument of this form. It basically amounts to saying that we should give in to simplistic thinking and not bother to make careful distinctions because people might misunderstand you. It's tantamount to saying that we shouldn't prmote careful thinking because people don't know how to think carefully. Isn't the proper solution to woolly thinking to model careful thought and let people observe how to think issues through in a way that demonstrates the very sort of thing people need to see to improve in this sort of thing? I just don't think giving in to simple-mindedness is a good reason to refrain from using a counterexample that really does illustrate a good consequence of a bad action, even if some people will mistakenly conclude something that doesn't follow. So I think President Bush was wrong to call Bennett's remarks inapprpriate. That seems to me to be giving in to forces in our society that he simply shouldn't be giving in to.

1 TrackBacks

Listed below are links to blogs that reference this entry: Bill Bennett, Abortion, and Race.

TrackBack URL for this entry:

The Parable of Bill Bennett from Philosophy, Computers, and Bad Writing on October 6, 2005 12:08 AM

The Parableman takes a look at Bennett and his recent comments: Parableman: Bill Bennett, Abortion, and Race The role his example has in the argument is that it's a counterexample to a particular claim. To serve as such a counterexample, it needs to... Read More


Good post. I'm interested in the following claim though:

"He could have made it raceless and just had every fetus being killed regardless of race, but that wouldn't have been as bad a case"

That seems false to me. I've written about it some more, here.

Ah, good point. I think what I want to say can still be said. It's not that this is in every sense worse, but I think it's worse in the sort of way that he wanted. I've left what follows on your blog, but I figure it would be good to put it here too.

It's clearly a worse consequence in the head-counting way of measuring consequences. I think what Bennett might say, though, is that you're assuming not so much consequentialism as a theory of the good that he wouldn't agree with. An action can be worse than another even if the second has worse consequences in terms of, say, hedonism. Indeed, genocide of a generation by arbitrarily singling out a whole race seems worse in some ways than simply killing a generation of children in general.

Hedonism counts up the amount of pain caused and pleasure robbed, and it clearly comes out as worse to kill them all than it would be to kill some of them. If you expand the theory of the good to include desire fulfillment, as you would of course do if I remember your view correctly, I think the same would happen. If Bennett thinks there are intrinsic goods beyond that, he might well think an outcome is objectively worse even if it's better according to a hedonist or desire-fulfillment view of the good. My suspicion is that Bennett would hold to quite a number of irreducible intrinsic goods, and some key ones might be so much lower in the genocide case than in the blanket case that he would consider it worse.

One suggestion as to one such thing that's worse is something about the motivation of the person doing it. It seems to me to be much more profoundly evil to single out a particular group for extermination than it is to kill willy-nilly, which usually signals not so much evil as lack of moral sense whatsoever. The fact that the motivation is evil makes not just the action in one sense more evil but even makes the consequence worse in some sense, even if it doesn't make it worse in every sense.

It's the sense of profound moral evil that Bennett really would care about for this case, I think, not so much the sense of bad consequences. He wasn't saying that the good consequence of more GNP would be outweighed by the bad consequence of genocide. He was saying that the fact that it's genocide, a true evil, makes it wrong even to look at the consequences.

It occurred to me also that Bennett couldn't have used the example of aborting all of a certain generation irrespective of race. That wouldn't give you the same result. Killing a generation of a population with a higher crime rate with respect to other populations will likely lower the crime rate. Killing a generation of everyone isn't going to have the same sort of effect. He needed to single out a population with a higher crime rate, or the example wouldn't have served as a genuine counterexample to the claim of his caller.

I question Bennett's use of the term "crime rate" , maybe because I'm not understanding the context of its use. Crime rate is the number of crimes (reported or committed) as related to the population. I looked on the FBI website which shows the number of arrests, there are no statistics on the number of reported crimes, and it shows of a total (rounded) 9.5 million arrests that 7 million were white and 2.5 million were black. Relating that to the population gives a total crime rate of 4776 per 100,000 people, 3375 for whites and 1284 for blacks. Looking at these figures shows a higher crime rate for whites. Knowing that it would be an illogical thing to connect the abortion of blacks with reduced crime rates unless there were, as you point out, an intention to demonstrate a more heinous and absurd proposition. He did not go on to say anything about his intention other than to make a logical statement. It's that logic that I don't think exists considering the premise, inference and conclusions he used. What am I missing here? Maybe it's the definition of "crime rate" and whether this was to be understood as the black crime rate compared to the white crime rate as related to the racial composition in society. any comments? (I will bookmark this page and see if there are any) thanks

Bennett didn't mean crime in general. This was a radio conversation, and he was giving an argument off the cuff. What he was referring to was the established fact that violent crimes, particularly murder, have a much higher incidence among black communities. Black people are six times more likely than whites to be murdered, and black people are seven times more likely than whites to commit homicide. Almost all of these cases are blacks killing blacks. The black rates of violent crimes involving drugs are even higher. (These numbers are from 2002 and are listed at the DOJ website.) I think the numbers you're pointing to include all crimes, and that's why they hide the much higher rate of violent crime among blacks.

I've also read that the FBI statistics on this take most Hispanics as white. I haven't found anything to substantiate that, but most Hispanics in the U.S. are not considered white by most of themselves or by most Americans, and if they happen also to have a higher crime rate the numbers will be inflated. As I said, I have nothing to substantiate that, but I've read it. The more important point is the distinction between violent crimes and other crimes, because it was violent crime that Bennett meant.

I'm only looking at the logic of the statement and whether it was logical or not logical statement to make given the context. I'm not judging whether it was right or wrong or a good thing or bad thing to say. I need more information to make a determination of the logic of his statements. and much of that would be determined by the context of the discussion.
you state that what bennett MEANT was violent crime although he didn't state that nor did he state that it was black-on-black crime.
was bennett also meaning to infer a hypothetical one-time abortion, or a 14-year period of abortions or a longer-term period of abortions because it would make a difference to whether in fact there would be reduction in crime. if it was a one-time thing there may be a small blip in the population of blacks (or any race subjected to that event) but that would hardly be enough to effect a substantial change in the future crime rate.
I have listened repeatedly to the tape of the call
and it seems that bennett switched from the subject that the caller had brought up to another subject, which takes away from the logic of saying what he did. He even switched from the subject of freakanomics which was crime rate and abortion and he injected race into it which was neither the emphasis of freakanomics nor the subject of the call in.
looking purely at the logic of what he said it seems to be not a logical thing to say in the context of the call

Of course he switched subjects. That's what you do when you give an analogy. It wouldn't be an analogy if you didn't give an analogous example. It would be strict repetition, which wouldn't show anything.

The form is as follows. The caller justified his view of abortion because of one effect that it might have. Rather than question whether it would have that effect, Bennett wanted to give an example of something that was uncontroversially wrong even if it had a good effect. So he said what he said, which is uncontroversially wrong even if it might have that good effect. It doesn't even matter so much if it would have that good effect. What matters is that, even if it did have that effect, it wouldn't make it ok to do the action. You're missing the point if you focus on whether it would have that effect, because he would just respond by giving a different example.

I think he did mean a long-term policy, though. And I don't know if he meant violent crime. He may not have been thinking about the particulars but may just have been referring to the statistics he knew, which turn out to be about violent crime. So he was referring to the violent crime rate, whether or not he was thinking of it as the violent crime rate.

i see what you're saying and I know it was an analogy, but the segway was just not there; non-existent, no "sort of like..." or "think of it as the argument of..." or other lead-in, it was just POW! out there. there had been no mention of race in the caller's question and no mention of race in the freakanomics, that's what threw me.. I found myself saying, "where did that come from?" what was the logic of introducing race? I read bennett's explanation of his comments but intuitively looking at it for a cause when it was said, I can't see a cause for bringing up race other than that it may have been something that he had been thinking about separately from the call-in subject

Well, I could see what he was saying immediately. It's a standard style of argument in philosophy, and Bill Bennett has a philosophy degree. I believe he's even taught philosophy classes. This is just an ordinary argument from analogy, and all philosophers I've discussed this with seem to have picked up on it as easily as I did. This is a standard counterexample, which philosophers use all the time. Counterexamples often have to be far enough from the original example that if you don't understand the argument's structure you'll think it's a change in subject, but that's how counterexamples work. It really is a change in subject, but it's to show that the principle being used in the original case is simply false. That's exactly what Bennett was up to here, and it's one of the most basic arguments in the philosopher's arsenal.

I understand your point but it would have been better, from this listener's standpoint, for Bennett to have added some clarification or explanation as to why race just popped up into the conversation. I believe that both "abortion" and "crime rates" are such complex social parameters that they cannot just be thrown together in an analogy and claimed to have a logical connection. Factual premises are required to be able to draw a logical connection and the connection drawn in his argument was merely an untested hypothesis, not a proven fact yet it was stated as a logical conclusion even though it may have only been an analogy.

Leave a comment


    The Parablemen are: , , and .



Books I'm Reading

Fiction I've Finished Recently

Non-Fiction I've Finished Recently

Books I've Been Referring To

I've Been Listening To

Games I've Been Playing

Other Stuff


    thinking blogger
    thinking blogger

    Dr. Seuss Pro

    Search or read the Bible

    Example: John 1 or love one another (ESV)

  • Link Policy
Powered by Movable Type 5.04