1. What is intelligence?
One simple, straightforward definition of intelligence is is “problem-solving ability.” Another definition is “that which IQ tests measure.” Egalitarians will object, “Since we can’t all agree on a definition, it’s a useless concept.” Not true! Intelligence is like heat. We know the difference between hot and cold, and we can measure fine gradations of heat. Some people will say, “It’s too hot in here!” while others say, “It’s too cold!” Does this mean we must discard the concept of heat? No. Almost any definition of any word could give rise to disagreement. We don’t have unanimity on definitions of many important constructs which we use every day, but we carry on nevertheless, and we are much better off with them, than without them.
Egalitarians also love to say, “But IQ isn’t everything!” That’s true. (Is there anything which is everything?) But IQ clearly is something very important. Those who pooh-pooh it have an impossible task explaining why IQ is the single best predictor of success in school and in life. How could anything which measures nothing – or even something trivial – predict success so well?
2. There are many admirable human qualities that aren’t measured by IQ tests. There will never be consensus on what all of those qualities are. What gives any of us the right to decide which ones to phase out?
There’s already a consensus on the fundamental traits we value – for example, what traits would you want to see in your children? Most people want their children to be healthy, intelligent, sane, law-abiding, and conscientious – meaning possessing good character (honest, hard-working, concerned for well-being of others). These are universally valued traits. Have any parents, anywhere, ever said, “We’re hoping our son will grow up to be a psychopath”? Or, “We hope our daughter will be retarded”? These values were exactly the same 100 years ago, and 1000 years ago.
Another way this consensus is expressed is in government expenditures on hospitals, research on diseases and mental illness, prisons, police, etc. We as a society are already very clearly trying to change people, using environmental engineering in a marginally-effective attempt to make people smart, law-abiding, sane, and healthy. Why not do something that really works?
A “right” implies there’s something in it for us, when in reality, there’s nothing in it for us. I believe that we have a responsibility to future generations, and a great and unique opportunity to help them. We already agree on what is good, and what is not. There’s absolutely no doubt about it – we are quite sure that we wouldn’t want to be diseased, retarded, a criminal, a psychopath, or insane – so it’s no great leap of faith to assume people of the future don’t want that, either.
But it’s not as if a “Eugenics Court” will dictate each individual who can and cannot be born! A likely scenario is that legislators, in response to public opinion, will form a new Eugenics Department that will provide attractive incentives for criminals and the mentally deficient to be sterilized, and incentives for bright, healthy couples to have more children, and medical professionals to help prospective parents make decisions on how best to utilize the new reproductive technologies.
3. Everyone knows that IQ tests are biased – what makes you think they’re not biased?
“Everyone knows” that IQ tests are biased because the media keep telling us this, but it’s an outright lie. Here’s an example of real bias: Say an IQ test is created and standardized in England, and the vocabulary section includes words like “lorry” and “scones.” If this same test were given to American kids, these items would stand out rather conspicuously. When you looked at the data, you would recognize immediately that: (1) answers to these questions were merely random guesses, (2) kids who scored high on the test as a whole weren’t any more likely to get them right than those who scored low, and (3) older kids didn’t do any better than younger kids. This means they’re worthless questions with no predictive value for the American kids, because all they do is add “noise,” thereby reducing the reliability and validity of the test. Furthermore, if nobody ever bothered to look at the data and delete these questions from the American version, they could legitimately be said to be “biased” against American kids in relation to the English kids.
By analyzing the data this way, it’s possible to determine definitively whether a test is, or is not, biased against any group, or whether particular items are biased. If a test doesn’t satisfy the criteria for bias, it’s not biased. People’s feelings, and what may appear on the surface to be bias, have nothing to do with making this determination. Also, there’s the crucial question of whether the test predicts success equally well for all groups. The fact is that IQ tests and other standardized tests predict success in college and in career in blacks as well as whites.
In Arthur Jensen’s authoritative work on the subject, Bias in Mental Testing, he found that IQ tests are not biased (using statistical criteria), except that the tiny unreliability of the tests slightly favors low-scoring groups. Also, it’s hard to imagine how the argument of bias in favor of Caucasians could be refuted any more effectively than by the finding that American kids of Japanese ancestry score higher on average.