Editore"s Note
Tilting at Windmills

Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

February 21, 2003
By: Kevin Drum

INTELLIGENCE....Liberals rightly castigate conservatives for denying the mounting evidence in favor of global warming. The problem, of course, is that conservatives are unhappy with the idea of reducing emissions of greenhouse gases, so for ideological reasons they are forced to deny the science itself. They are like tobacco company executives denying that smoking causes cancer.

But conservatives are not the only ones who fight passionately against science they find distasteful. As Charles Murtaugh pointed out yesterday, among liberals it's almost impossible to discuss the question of human intelligence a reasonable area of scientific study without being accused fairly quickly of racism. As Charlie points out, there is good reason for this: studies of intelligence invariably conclude that it is partly a result of genetic factors, and there's an unfortunately large and unsavory group of people who have dedicated their lives to insisting that this means that blacks, who score lower than whites on IQ tests, are genetically less intelligent than whites. This case has been made most recently in a long and execrable book called The Bell Curve, a collection of pages that probably set back research into intelligence by a decade.

But we shouldn't let these people drive the discussion or drive us away from it. Instead, we should talk about the science, we should talk about the results, and then we should talk about where it leads us. So let's do that.


Liberal disgust with racist misuse of research into intelligence is understandable especially given its long and miserable history but in fighting it we have a tendency to throw the baby out with the bathwater. In particular, liberals are forced to make the following arguments:

  • Intelligence isn't really a meaningful trait.

  • And even if it is meaningful, IQ tests don't measure it well. They are culturally and racially biased.

  • And even if intelligence is measurable, it doesn't have a significant genetic component. It's mostly based on environment and upbringing.

Well, is intelligence real? After all, there's no specific point in the brain we can point to and say, "That's where intelligence lies."

But, like athletic ability, intelligence is a useful, everyday umbrella term that has genuine meaning. It is, roughly, the ability to deal with analytic complexity, and let's face it: we all know it when we see it. When we say "she's a smart cookie" we generally agree that this mean she catches on quickly, she predicts the consequences of actions well, she can juggle a lot of balls at once, she can put two and two together, and so forth.

Technical discussions of intelligence often revolve around whether there is a mental trait called "g" (for "general intelligence") that underlies all the factors that make up intelligence. The current consensus is that g does exist, but for our purposes it doesn't really matter. What does matter is that from a practical perspective, whether g exists or not, the term "intelligence" has a useful meaning and describes something that's important. In fact, the main reason that intelligence is such a hot button is that it so obviously is important. In a complex society, high intelligence is an extremely valuable commodity, and this makes the politics of intelligence both contentious and ideological.

Still, "I know it when I see it" isn't good enough for scientific study: we need to measure intelligence, and the preferred method is via an IQ test. This forms the second line of liberal defense: IQ tests are no good. But once again, the evidence says otherwise. Most professionals in the field agree that IQ (and related) tests do exhibit some level of group bias, but the bias is quite modest on modern instruments. If you score low on an IQ test, the odds are very high that you are, in fact, not very intelligent.

The same is true of heritability. Evidence that intelligence is party heritable has been accumulating for years, and today it is a veritable mountain. There's still plenty to argue about, but the consensus opinion is that intelligence is approximately 50% heritable and 50% learned. In other words, about half of the variation in human intelligence is based on your genes.


There's really no reason that what I've said so far should be very controversial. But it is, and primarily for one reason: race. A regrettably large group of intelligence researchers have extended the results above to argue that blacks are genetically less intelligent than whites.

The best and most well-known recent example of this kind of tendentious reasoning can be found in The Bell Curve, a 500-page tome that is basically broken into three parts. In the first part the authors argue that intelligence is real, g exists, and that both are partly heritable. There are things to argue with, but it's a reasonable synopsis of current thinking.

Part 2 uses lengthy statistical analysis to argue that intelligence is correlated with lots of positive social values. Smart people make more money, commit fewer crimes, have fewer out of wedlock children, etc. There's more to argue with here since the authors overstate their statistical conclusions, but there's still nothing all that objectionable. After all, detailed statistics aside, I think we'd all be surprised if this weren't true. Our society values intelligence highly, and it's only reasonable that intelligent people do better in a society that values them.

Then there's Part 3, in which the authors argue that the 15-point IQ difference between blacks and whites is primarily caused by genetic differences. And suddenly the quality of their arguments falls off a cliff. They report tentative results as positive, they ignore most of the contrary evidence, and, like so many likeminded researchers, they vastly underestimate the power of environmental factors. Sure, intelligence is 50% hereditary, but that means it's also 50% environmental. And that 50% is more than enough to account for a 15-point difference but only if you take seriously the wretched conditions that blacks at all socioeconomic levels face in this country.

The problem is that human beings are not like stalks of corn planted in poor soil. You can replant seed corn in good soil and the resulting plants will be beautiful. But humans are cultural animals, and the effects of poor environment don't disappear overnight. It takes generations, and in our case it's been barely one generation or maybe none since we have started to treat blacks fairly.


None of this might matter too much if it weren't for the fact that ignoring the value of intelligence hurts one of liberalism's most cherished goals: equality of the races. But in our efforts to discredit the scientific racists, that's exactly what we've done.

Although the origins and effects of racism are long and complex, there's little question that lower average intelligence is one of the big reasons that blacks do poorly in American society. The fact is that the black-white gap does exist, and it's not merely a cultural artifact or the result of bias on standardized tests. It's a very real thing and it needs to be attacked head on.

This is the reason that I feel so strongly about primary education: it's our best hope for erasing the test score difference and truly bringing an end to racism. It's also the reason that I believe these efforts should be primarily aimed at blacks, not at poor people in general: while better education helps everyone, that 15-point gap indicates that it's blacks as a group who suffer the most and need the most help.

Intelligence is not a synonym for social worth, nor is it the only mental trait that's important. But denying that global warming exists or that smoking causes cancer does not make those problems any less real, and the same is true of research on intelligence. In an increasingly complex world, intelligence is increasingly important and increasingly valuable, and denying this does no one any good.

Nor does it do any good to scratch our heads and cast about for another line of defense: intelligence may be real, but race isn't and therefore it makes no sense to talk about racial differences in intelligence. In the real world, however, race, like intelligence, is a fuzzy but still useful concept, and this argument becomes little more than yet another transparent excuse to avoid distasteful truths.

What should be important to liberals is not the results of intelligence research themselves, but what we do with them. If we face the truth squarely, we are in a better position to argue for social programs that have a realistic chance of accomplishing what we want. That's what we should focus on, not on the remote and dwindling hope that the scientific community is wrong. It's a bad bet.

UPDATE: If you want to read more (much more!) about this, Christopher Jencks and Meredith Phillips wrote a 1998 book on the subject called The Black-White Test Score Gap. It is available online from the Brookings Institution here. They summarized their argument in The American Prospect in this article, and responses were published here.

UPDATE II: More comments from:

DeLong in particular is useful because he reprints this Thomas Sowell column (indirectly linked in my original post) that although more sympathetic to The Bell Curve than I am very clearly makes the point that intelligence is mutable and can be significantly changed via education and other environmental factors.

Kevin Drum 4:38 PM Permalink | Trackbacks

Bookmark and Share
 
Comments




 

 

Read Jonathan Rowe remembrance and articles
Email Newsletter icon, E-mail Newsletter icon, Email List icon, E-mail List icon Sign up for Free News & Updates

Advertise in WM



buy from Amazon and
support the Monthly