Monday, October 03, 2005

Crucified for the Truth

O.K., this is going to be the most controversial post I've ever done, but since I have so few regular readers, I'm not thinking that that is going to matter much. If you have been following the scandal news of the US of late than you have been unable to escape the story of William Bennett. It seems this former Secretary of Education known for making hateful and bigoted comments was caught yet again spreading racism and prejudice on his nationally syndicated radio talk show. Bennett is absolutely being ripped apart in the press here, by politicians (both Democrat and Republican), commentators, talk show hosts, editorials, you name it, everyone wants to get in on this band wagon and rip this guy a new asshole. Now don't get me wrong here, I am NOT a fan of this man. He HAS said some very hateful, hurtful things in the past. But I think we need to take a closer look at what Mr. Bennett said in this specific case. Here are the comments that have spawned such a virulent backlash:
"If you wanted to reduce crime, you could -- if that were your sole purpose -- you could abort every black baby in this country and your crime rate would go down.
"That would be an impossibly ridiculous and morally reprehensible thing to do, but your crime rate would go down."

Look at the whole quote, look at the qualifying remarks, and now understand the context of the comments. Bennett was having a conversation with a listener of his show about the social and economic effects of abortion in this country. During the conversation, Bennett brought up some results published by an economist (Steven D. Levitt) with the help of a freelance writer (Stephen J. Dubner) in the book Freakonomics. Levitt's analysis of the data suggest that the most strongly correlated factor to the significant drop in crime during the decade of the 90's was the legalization of abortion in 1973 approximately 20 years earlier. Levitt doesn't stop there, he analyzes the data further to determine if the correlation suggests causality. To do this, he compared the crime rates of the 5 States that had legalized abortion prior to Roe v. Wade to those of the other 45 States of the union and discovered that indeed, the crime rates in those States had begun falling years earlier than the crime rates in the rest of the U.S. Then he compared the crime rates in the States that had the highest abortion rates in the decade after Roe v. Wade with those in the States with the lowest abortion rates, and once again discovered that crime rates fell faster in those States where abortions rates were highest. Other statistical analyses confirm these results.

Using this as launching point, Bennett makes the above "controversial" statement. So now let's consider his statement. Bennett suggests that if every black fetus were aborted in this country, we would see a drop in the crime rate. I think we have to assume that Bennett means first, that such a thing would have to happen on a sustained basis for many years, and second that we would see that drop in crime rates in approximately 15 to 20 years after such a scenario started. So, here's the question: is Bennett wrong? No, he isn't. Bennett is absolutely correct in his analysis, and I challenge anyone to produce a sound mathematical/statistical argument to prove otherwise. At the same time, I also think it is quite clear that Bennett is attempting to forward this notion as an absurd hypothetical and that he does not believe that the abortion of all black babies is a reasonable strategy for reducing crime. If Bennett were actually advocating such an idea as a solution to crime in this country, then I could completely understand the indignation and angry responses his statements have engendered. But clearly he isn't. Look what Bennett says next. He acknowledges that such an idea is "impossibly ridiculous and morally reprehensible." Bennett wasn't calling for the abortion of all black children, he was attempting to make a point with his listener. And while admittedly, he did so quite clumsily, the claim he makes is undeniably true. Further, he qualifies his remarks sufficiently to make it clear that he isn't advocating a policy of forced abortion for black women.

So why is this man being crucified in the public arena in this country? This is political correctness at its absolute worst - a monumental case of the ostrich sticking its head in the sand in order to deny eminent catastrophe. Rather than vilifying Bennett over his comments, we ought to examine what circumstances in this country account for the fact that Bennett's analysis is absolutely correct and reflect on the steps we need to take in this country to remedy those circumstances. It is a national disgrace and disaster that we have created a society and culture that has resulted in Bennett's words being true. Attacking the messenger doesn't do a damn thing to illuminate solutions to the problem.


At 6:08 AM, Blogger Malik said...

I think your analysis is faulty in several ways. First people aren't outraged because they think Bennett was advocating aborting black babies. They're outraged because he implied that black people and crime are intrinsically linked, and he refuses to admit that.

Second, even from a statistical point of view, the hypothesis that eliminating black people from the population will reduce the crime rate is unsound. Take a hypothetical population of 1000 people. Of those 1000 people, 120 are black (approx. 12% of the population is black). Among that population about 23 violent crimes occur that year . That gives us a violent crime rate of 2.3% (the actual rate of total violent crimes for the most recent year that complete data is available, 2003) Of that total of 23 crimes, 21.3% were committed by black people (again, actual data from 2003). (The actual term used by the Department of Justice for racial categories is "perceived race of offender", which, in the case of the "black" category, can mean anything from Pakistani to Brazilian, but that's another discussion.) So, of our 23 violent crimes, (rape/sexual assualt, murder, robbery and assault) about 5 were committed by black folks.

Now let's say we take your suggestion, and eliminate black people from the population. That means that we've reduced the population by 120 folks, leaving 880 people. We've also eliminated their 5 crimes, leaving 18. That gives us a new crime rate of 2.0% That's a difference of .3%, which, as I understand it, isn't even statistically significant. The statistics I used are from the Department of Justice National Crime Victimization Survey. If you come to a different conclusion using the 2003 numbers, I'd like to see how.

At 8:48 AM, Anonymous treacletrade said...

Hmm . . . a tricky subject. Racism being the new witchcraft you only have to be accused to be found guilty.

Years ago the Metropolitan Police used to profile crimes by race/demographic group - this was discontinued for the sake of "race relations". The uncomfortable fact was that 96% was commited by Afro-carribean males.

Hiding these facts doesnt do anyone any favours, until we are honest with ourselves we wont be able to try and solve these problems.

At 8:50 AM, Anonymous treacletrade said...

Oops! just read my last comment and the crime I was referring to was Mugging NOT 96% of all recorded crime. Sorry!

At 11:44 AM, Blogger Faltanus said...

malik, you're wrong. People aren't outraged by Bennett "implying" that black people and crime are "intrinsically" linked (most people don't even begin to understand the subtlety of what you are trying to get at in that statement), they are outraged that he is linking black people and crime...period. because in this hyper politically correct climate we live in, "polite" people don't make that link.

Next, you've made several significant errors and oversights in your rebuttal. First of all, a change in violent crime from 2.3% to 2.0% isn't a 0.3% reduction - that is simply the difference in percents. Think about it - if we could reduce crime to 0% would you argue that we had only reduced crime by 2.3% and wow - what a small percentage that is. Nope, you would say that crime had been completely reduced! What a major event! What you need to do with you result is to find the percent that 0.3 is of 2.3 - it turns out to be 13%. That is, a 13% reduction in violent crime. Not significant?

Next, you've missed one of the most significant implications of Bennett's scenario. That is, in 15 to 25 years, Bennett's scenario doesn't eliminate all Black people, it eliminiates Black people between 15 and 25 years of age. I don't have the exact statistics at my fingertips, but how many of the 5 crimes you mentioned would have been committed by that specific age subgroup? 4 of them? So let's reduce the population by 40 (a significant over estimate), rather than 120, but still reduce the number of violent crimes by 4 and now what is the change? A reduction from 2.3% to 1.8% or a drop of nearly 22%.

My point in ALL of this is that the real implication of Bennett's comments, what we should all be walking away from those comments with, is an indictment of a social, political, economic system that perpetuates a cycle of almost institutionalized poverty that preys disproportionately on a specific minority group in this country. So let's focus on the problem illuminated by the comments, and not on the commentator himself.

At 3:33 PM, Blogger Malik said...

Interesting, but I still have to stand by my analysis. I both listened to and read Bennett's comments in full. I just went back and reread them to assure myself that I wasn't being selective in my reading, and I wasn't. First, let's deal with the substance of his remarks. The scenario that he proposed was in no way as reasoned and nuanced as the one you put forth on his behalf. He stated, simply and bluntly:

"I do know that it's true that if you wanted to reduce crime, you could -- if that were your sole purpose, you could abort every black baby in this country, and your crime rate would go down. That would be an impossible, ridiculous, and morally reprehensible thing to do, but your crime rate would go down."

Note the last phrase. He paused, dropped his voice a register, and repeated, for emphasis "your crime rate would go down." Nothing is being misconstrued. He deliberately chose to use an inflammatory hypothesis to draw attention to his assertion that black people and crime are intrinsically linked.

As for your rebuttal to my statistical argument, you're drawing comparisons between different sets of data. If the population in this example remained constant (i.e. an entire race of people isn't eliminated as in Bill Bennett's example) then a year over year reduction of %13 in the overall incidence of crime among that total population would indeed be significant. However, that's not what we're measuring, because that's not what he suggested. His hypothesis involves wholesale genocide, which would both reduce the population and the absolute crime rate among the population. That change in the absolute crime rate among our now reduced population is only .3%, which, to my understanding, is not statistically significant.

Secondly, there's a problem with the mathematical reasoning in the detailed scenario that you yourself hypothesize. You say that, hypothetically, 4 of our 23 violent crimes are committed by the 15-25 age group. You suggest that by eliminating that age group, you might reduce the population by 40 individuals. In other words, 15-25 year olds are 33% of the black population of 120 people, but they are responsible for 80% of the crimes committed by black people, and approximately 17.5%of crime overall, according to your example. Now, let's say you eliminate those 40 individuals, remembering that you're reducing the total population, removing both perpetrators and victims of crime. You're left with a total of 19 crimes among 960 people. That leaves you with an absolute crime rate 1.98%, not 1.8%, and a year over year change (if we're examining yearly change rather than the absolute crime rate) of, you guessed it, 17.5%.

However, in both of the above scenarios you would get better crime reduction results just by eliminating any random group of people from the population, rather than concentrating solely on black people. Indeed, the Authors of Freakonomics, whose arguments Bill Bennett claims he was rebutting, have said that this is exactly what has resulted from the increase in abortions, but they didn't attribute the crime reduction to race. In fact, Steven D. Levitt, one of the co-authors of Freakonomics said, "None of our analysis is race-based because the crime data by race is generally not deemed reliable."

All the intellectual contortions in the world cannot make Bill Bennett's remarks defensible.

At 4:57 PM, Blogger Malik said...

Upon reflection, I realized that you're quite correct to point out that what we're concerned with in these examples is the change in the incidence of crime versus a change in the absolute crime rate. Still, the logic behind both your hypothesis and Bennett's hypothesis does not stand, for the reason that I pointed out in the last paragraph: a random reduction in population serves to reduce your crime rate more effectively than concentrating on black people alone, which is the topic that the authors of Freakonomics explored, contrary to Bennett's assertion. So, we're not talking about a studied reflection on the facts, as you assert. His comments simply reveal bald prejudice.

At 6:31 AM, Blogger Faltanus said...

malik, I don't want you or anyone to take away from my post or this discussion that I am attempting to defend William Bennett as a person. I started out the post acknowledging that Bennett is a bigot and spreader of hate. I am trying to make 2 points here. First that the statement Bennett made here, as a statement of mathematics and statistics is in fact true. On that point, you are wrong about Levitt's analysis, it is not about a random reduction in population, because as Levitt points out, abortion is, for numerous reasons, used disproportionately by women from lower socio-economic classes. In fact Levitt makes it clear that it is this very fact that brings about the strong statistical correlation between abortion and reduction in violent crime. Why? Because violent crimes are committed in greater proportion by young adults from the lower socio-economic classes. So the effect of abortion was not a random reduction in population as you assert, it was (and continues to be) a selective harvesting of a particular segment of our population. In particular, abortion serves to reduce a very specific phenomenon - unwanted children. Unwanted children tend to grow up abused and neglected and are much more likely to become violent criminals as a result.

Now, of course, it is not true that we can extend that idea to a scenario in which all black children are forcible aborted. We have to believe that the majority of children born to black parents are wanted, are loved, are not neglected and abused. But acknowledging that fact, it doesn't change the fact that from a strictly mathematical/statistical standpoint, Bennett is correct, violent crime rates would be reduced.

The second point - the really important point - and I tried to make this point in both my original post (I don't think I made it forcefully enough) and in my first rebuttal to you (I think I made the point more strongly there) - is that despite the fact that Bennett is a hateful bigot, in attacking Bennett and his comments, his attackers are deflecting attention from the real problem. Poverty preys disproporionately on certain minority groups in this country and considering the fact that we are the richest country in the world, the fact that we have a social, economic and political system that allows this to continue to happen, generation after generation is deplorable, and even more than that, it is dangerous. More than any other nation in the world, the United States has the resources to battle this problem head on, and the simple truth of the matter is, we don't. Rather we spend our time and energy attacking ignorant radio talk show hosts for perceived violations of political correctness.

At 4:04 AM, Blogger Malik said...

I agree that attacking Bennett is pretty useless, but I don't think you can dismiss his comments so easily, because they raise larger issues. Here's something I wrote earlier about the portrayal of black Katrina victims:

"This raises the larger issue of the pervasiveness of negative portrayals of black people, in popular culture, in academia, and in the news media. To a large extent, black people are used as a scapegoat for and a symbol of society's ills (just ask William J. Bennett). Black people have problems like everyone else of course, but our problems are discussed as if we are a problem people. There's crime, and then there's black crime. There's poverty, and then there's black poverty. Our problems are held to be endemic to our culture, and there's a strongly held suspicion, acknowledged only in whispers, that they're quite possibly a result of our genes."

That's the phenomenon that I think Bennett's comments reflect, and that's why I think they deserve reflection and discussion. I don't think that attitudes towards black people and their social status can be treated as separate issues. Perhaps you're right about the statistical issue, but as I said earlier, Bennett spoke as if black people and crime are instrinsically linked, and judging from the responses of his defenders, there are scores of people who believe that. I think it's important to talk about why that's so, and to also attempt to counter that myth. I did a poor job of that with my statistical example, but the effort has to be made.

At 7:00 AM, Blogger Faltanus said...

i definitely understand what you are saying malik, and i agree that the issues you raise need raising, need discussion, need to understood and addressed. in fact i don't think that the point i was trying to make is too terribly different from your own. maybe i just didn't do a very good job in making clear that a statistical link between two things does not in any way, shape or form "prove" or even really imply a causal or intrinsic connection between those two things. but a lot of people try to use statistics to make that very argument, and you are right, it is inappropriate to do so.

At 10:34 PM, Blogger TheDevilIsInTheDetails said...

Another abortion alternative organization Resource... . A discussion forum for all that deals with such hot-button issues as abortion alternative organization .


Post a Comment

Links to this post:

Create a Link

<< Home

Free Website Counter
Web Counters