An Interview with Charles Murray from Skeptic volume 3 number 2 1995

Posted: December 6, 2011 at 11:49 am

An Interview with Charles Murray from Skeptic volume 3, number 2, 1995

From Skeptic vol. 3, no. 2, 1995, pp. 34-41.

The following article is copyright © 1994 by the Skeptics Society, P.O. Box 338, Altadena, CA 91001, (818) 794-3119. Permission has been granted for noncommercial electronic circulation of this article in its entirety, including this notice. A special Internet introductory subscription rate to Skeptic is available. For more information, contact Jim Lippard (lippard@skeptic.com).

FOR WHOM THE BELL CURVE TOLLS

A Prelude to an Upcoming Special Issue of Skeptic (Volume 3, #3)

An Interview with the Author of The Bell Curve

CHARLES MURRAY

Interview by Frank Miele

Charles Murray has achieved the impossible, or at least the highly improbable. He has co-authored an 845-page book, filled with figures, tables, references, and appendices loaded with multiple regression analyses, that is also the most controversial book in America. It has been panned by many outside the intelligence testing community and by some within. Commentators from the left, right, and middle have taken their best shots, and leaders of both major political parties have rung in with scathing attacks, even while admitting they had not read the book. Despite the brouhaha, or perhaps because of it, The Bell Curve made it to the New York Times top-10 nonfiction best seller list. As Ted Koppel put it on Nightline, The Bell Curve will be like Clinton's health plan: no one will actually read it but everyone will form an opinion of it.

Charles Murray is no stranger to controversy. His previous book, Losing Ground: American Social Policy, 1950-1980, argued that the Great Society programs of the 1960s not only did not help the poor, they often made things worse. Arguing that the welfare system should be abolished, the New York Times called Losing Ground, "The [Reagan] Administration's new 'bible.'" Losing Ground was but a prelude.

In The Bell Curve: Intelligence and Class Structure in American Life (co-authored with the late Richard J. Herrnstein, author of the previously controversial IQ in the Meritocracy, and co-author with James Q. Wilson of Crime and Human Nature), Murray has pushed the envelope of public political discourse to its breaking point. He has now been dubbed by the New York Times Magazine "America's Most Dangerous Conservative." When editor Andrew Sullivan ran an excerpt from The Bell Curve in The New Republic, its entire editorial board rose in revolt. But a group of leading researchers in the field of human intelligence published a statement in the Wall Street Journal agreeing with the factual basis of The Bell Curve.

Herrnstein and Murray argue that IQ is real; that it matters (ever so much more as society becomes more equitable and technological); that it is somewhere between 40% and 80% heritable; and that it relates to not only school performance, but to jobs, income, crime, and illegitimacy; and that it cannot be ignored in any meaningful look at America's future. But the most explosive of The Bell Curve's arguments is that some of the difference in mean IQ scores between the white European population of the United States and the African-American population (one full standard deviation of 15 points) is probably attributable to genetic factors. No one in the field disputes this difference. The argument is over why the difference exists and, of course, whether and how it can be reduced. (Read the now-infamous Chapter 13 of The Bell Curve for yourself. It is a lot more tentative and nuanced than popular denunciations of the book may have led you to believe.)

Charles Murray is a graduate of Harvard with a Ph.D. in Political Science from MIT, and currently is a Bradley Fellow at the American Enterprise Institute in Washington, D.C. The late Richard J. Herrnstein received a Ph.D. in psychology from Harvard, where he taught from 1958 until his death last autumn. He held the Edgar Pierce Chair in Psychology (the oldest such chair in America).

Skeptic interviewed Charles Murray and found him to be as calm in explaining his positions as some of his critics have been apoplectic. If there was any trace of anger in Murray it was not for the underclass but for former colleagues who have, as he put it, "ran for the high hills," and for the Cognitive Elite, whom he feels have undermined the country that provided them the chance to rise to the top in the first place.

The next issue of Skeptic will be a Special Issue on intelligence, I.Q., race, and class, and will feature in-depth reviews of The Bell Curve by leaders in the field, moving past the rhetoric and getting to the heart of the science behind the claims. In the first of what promises to be a regular interview feature in Skeptic, Contributing Editor Frank Miele was more than prepared for the Murray interview, having been published himself on the subject in Intelligence, the leading journal in the field.

Here then is what "America's Most Dangerous Conservative" had to say to Skeptic.

Skeptic: In your book you present a summary of the current evidence on I.Q. on pages 22 and 23. Snyderman and Rothman's on The I.Q. Controversy in 1991 surveyed expert in the field, and just yesterday the Wall Street Journal contained a 25-point statement by experts in intelligence. Based on those it seems your summary represents the consensus of experts in this field, even on the controversial issue of the involvement of genetics and the black-white difference in intelligence. As skeptics, we are skeptics of everything, including psychology. If we get this great a controversy over what looks like consensus, is psychology really a science in the same sense as physics?

Murray: I'm not comfortable with a blanket statement saying yes or no. But I think we can talk specifically about the basis for those statements in the Wall Street Journal and the book, which is certainly based on the kinds of methods that fall under the scientific method-- falsifiable hypotheses, the use of predictions, etc. A test is valid in so far as it predicts something of interest, or criteria measure, to use the jargon of the trade. More than most of the other social sciences, psychologists and psychometricians are prepared to have their results tested against classical statistical criterion of validity, reliability, and reproduceability.

Skeptic: One of the arguments would be that most of the analyses you and psychometricians have done is correlation, as opposed to causal analysis that we do in physics. Does that mitigate against the strength of the scientific conclusions?

Murray: We do not have accessible to us the same kind of control over our phenomena that physicists often have. However, having said that, their remains a black box where the cause hides that we cannot open up and look at. But one can eliminate a number of alternative hypotheses and transform correlational statements into ones which certainly have some causal persuasiveness. Example: the use of regression analysis, which is the all-purpose tool of the behavioral and social sciences these days. Let's take an example from The Bell Curve. The dependent variable is whether a person is below the official poverty line. And you introduce as independent variables a variety of candidate causes. Chief among these being socio-economic background, education, race, occupation, and then you throw I.Q. into that. If after looking at a variety of these other things which both theory and common sense say should have some bering on whether a person ends up in poverty, but one ends up with a large, statistically independent role for I.Q., it seems to me to make a causal statement that I.Q. looks like its a cause of poverty, it is a reasonable thing to do.

By the way, when you actually read the book you will see that we typically word things in this cautious kind of way.

Skeptic: But could not someone say that in correlational analysis it is not really proper that you are not randomly assigning people to socio-economic status groups, or racial background, or whatever; are you and Herrnstein doing anything different than what is the common procedure used in regression analysis in, say, voting behavior, or anything of this sort?

Murray: The analyses we conduct in the book are garden-variety regression analyses. There is, however, also a body of work which does use randomly assigned experimental and control groups that reflects on a lot of the issues we talk about in the book, which then begins to look more like experiments done in the hard sciences.

Skeptic: Can you be more specific?

Murray: Art Jensen's work with regard to reaction time. This is a matter of eliminating a lot of alternatives and trying to understand what's going on with I.Q. scores. This is where you have a situation with an apparatus with six buttons, and you have your finger on a button, and when a light goes on one of the other buttons you move to that button and push it. The reason why this kind of experiment is useful is (A) it gives us insight into something that has no known relationship to I.Q., in so far as you are simply asking someone to move his or her thumb to push a button. But it turns out that this reaction time is not only correlated with I.Q. scores, it is correlated with the general intelligence factor, g.

The main point is this. You have now made some headway into trying to understand what is going on with this thing called an I.Q. score--does it have a physiological basis, etc.

I'm not going to apologize for our use of regression analysis in the book. Everyone uses regression in the social sciences. If you want to say that social scientists are the astrologers of the 20th century and that they don't have the methods of science open to them and we thus can't take them seriously, fine, but unless you are prepared to make that argument the science in The Bell Curve using regression analysis is very much in the middle of the mainstream.

Skeptic: Stephen Jay Gould, in his New Yorker review, gives a four-point summary of your argument about intelligence: (1) it is a single number; (2) it is capable of ranking people in a linear order; (3) it is genetically based; (4) it is effectively immutable. Gould goes on to say that if any of these premises is false, the entire argument of The Bell Curve collapses, and he concludes that: "The central premise is false and most of the foundations are." Now, how do you square what Gould has said about this with your own summary of the book on pages 22- 23. One of you has got to be wrong.

Murray: Stephen Jay Gould is recycling the same argument that appeared in The Mismeasure of Man in 1981. It was a very influential book in terms of the lay population and lots of people out there have their opinions formed about intelligence by The Mismeasure of Man, which included two types of information, one considerably more useful than the other. The first type was a history of intelligence measurements in the 19th and early 20th centuries, in which people made mistakes. (I seem to recall that physicists used to believe in something called aether.) There were phrenologists and others to whom we can now look back at and poke fun at. Fine, the problem is that in the same way that physicists are not criticized now for something other physicists did in 1910, so also has psychometrics made some strides since 1910 and 1900 and 1890.

The second thing is that Gould tried to make the same arguments for modern psychometrics, a lot of which were based on trying to demonstrate that the general mental factor g is a statistical artifact. The contrast I want to draw is between the attention that Gould's book got in the media and what happened in the scientific literature. Basically, there were a few perfunctory and rather derisive mentions of his treatment of factor analysis, and work went on without a break.

To put it more specifically, factor analysis is subject to a variety of kinds of problems because you can make different assumptions about how to create the factors. One can even, for example, demand of the algorithm that it produce factors which do not load on a single dominent factor. The problem with this is, as Dick Herrnstein use to say, you can make g hide but you cannot make it go away.

Skeptic: Sounds rather like Joe Louis. Let me go back to Gould's four points. Is there any one of those that you think is not a fair and accurate statement of what you said?

Murray: All four of them.

Skeptic: So you are not saying intelligence is a single number?

Murray: No. In The Bell Curve, we say of the I.Q. score, first, there have been a variety of ways to try to come up with independent mental factors. That has been a failure. On the other hand, there have been a variety of ways in which there are distinctions among different types of intelligence that are useful, such as the distinction between verbal, visual and spacial intelligence. And we talk about the different ways these different skills lead to success in occupations. And we talk, somewhat sympathetically, about the notion that there are, in Howard Gardner's words, multiple intelligences. We are a little dubious about applying the word "intelligence" to them, but we are very sympathetic that there are large domains of human talent that are not encompassed in the word "intelligence."

Skeptic: One of the complaints about the Snyderman and Rothman survey, the Wall Street Journal survey, and your own survey of the literature, is that you are working in that standard psychometric paradigm, but that is yesterday's news. The real forefront is Sternberg's approach to practical intelligence and Gardner's seven intelligences. You are sticking with something that is a very small portion of the discussion, so naturally you are going to get concensus.

Murray: Let me make a couple of other points about intelligence. One, the general mental factor, g, is very robust. You can take all kinds of different ways of creating your factors, and you will always get g. It doesn't matter whether rotate the matrix orthogonally, or obliquely, or whatever else, you always get the same thing. The second major point is that when you try with factor analysis to produce a situation where you do not have a general mental factor g, guess what? All the factors are correlated. Which goes back to Spearman's initial insight, which is why are the different measures of mental ability so consistently correlated with each other? What's going on here? The answer is: there is an underlying general factor. That does not mean that it blocks out a whole territory of human talents or intelligence, and we say so in the book.

Gardner has made a variety of assertions about intelligence which, if true, are falsifiable. He is not only saying there are different talents, which Dick Herrnstein and I would agree with, he is also saying they are independent. With something like kinesthetic talent, which is quite physical, this is easy to say. It gets harder to say when he talks about interpersonal skills, versus verbal skills. If you are going to make that kind of statement, the next logical step is to come up with measures of these different talents and demonstrate that they are, in fact, independent.

Skeptic: So you are saying that some of these disagreements are empirically testable?

Murray: Yes, and Gardner has consistently been unwilling to subject his own work to that kind of empirical defense. He has stood apart from quantitative attempts to describe what he is doing and to enable other researchers to replicate it. Of all the different types of intelligence that Gardner wants to treat as co-equals, there is only one kind that will put you in the retarded class--namely the plain old fashion general mental ability. If you are kinesthetically challenged, teachers and guidance counselors do not get real worried about you. If you are kinesthetically challenged you may be the last person chosen for the baseball team, but you can go out and make a success of yourself in any number of ways. If you are intellectually challenged, however, you have a general disability that is pervasive over all kinds of ways.

Skeptic: I read in a biography of Muhammad Ali that when he took his draft tests he had an I.Q. below 80. Now, if I make a mistake writing my spell-checker will fix it for me. If you make a mistake Stephen Jay Gould may beat you up in the press. If Muhammad Ali made a mistake he was flat on his back. This man was making split-second decisions of the first magnitude.

Murray: If you are five standard deviations out on the edge of the curve in kinesthetic ability to the right hand side, then certain possibilities open up to you. But if you are low in kinesthetic ability, it doesn't make much difference to you in life. If you are a Muhammad Ali and you possess extraordinary physical talent, you have other avenues that will open up to you. But this example illustrates another important point, which is that Muhammad Ali is not a blithering idiot. Yet there is nothing in his public utterances, his charm, his presence, his carisma, and all the rest of that, that is inconsistent with a measured I.Q. in the high 70s.

Skeptic: One of the things that Gould takes you to task for is that you do not report the scatter on your regression lines, and that the r squared values do not appear in the body of the book, but in the appendix. Can you tell us what those terms mean and why he thinks they are so important? And what is the usual practice here--have you and Herrnstein done something different than what would be done in a book on, say, political voting behavior.

Murray: Correlation is denoted by r, and in ordinary regression analysis r-square--the square of the correlation--reflects the proportion of the variation in the data that is explained by the set of independent variables that are in the regression analysis.

Two points. First, the book is exemplary in opening up the section in which we present these regression analyses by explicitly pointing out in the body of the book that the r-square is small. That is in the very first pages of the whole presentation. It is also exemplary in a book aimed at a general audience that we specifically include an appendix with a print-out of every single analysis presented in a graph in the body of the book. This is something you will not find in most books aimed at a general audience, including The Mismeasure of Man. Dick and I presented far more statistical information than is ordinarly presented in a book such as this.

Second, I don't know how much Gould knows about regression analysis. When you are using logistic regression analysis--in which the dependent variable is a nominal variable, in our case a dichotomous yes-no: Is the person below the poverty line, yes or no? Whenever you have a dichotomous dependent variable r-square becomes very difficult to interpret. Particularly with rare phenomena. For example, if you have a situation, as in the case of poverty where 87% of the population is not impoverished, you only permit two values in your dependent variable, you are going to get a lot of noise in the data, which means r-square becomes very difficult to interpret for technical reasons.

Skeptic: One of the criticisms, then, would be that the I.Q. isn't that effective. There is a lot of noise, so why are you saying it is so important?

Murray: Let's use poverty as an example. For poverty the r-square is .10. So we can explain only 10% of the poverty, so obviously I.Q. cannot be very important, can it? Well, if you then go back and take a look at the chapter on poverty, and then you take the probability that a person is going to be in poverty if he has low I.Q., you find out that among whites, the probability of being in poverty if you are in the bottom 5% of I.Q. is 30%. The probability of being in poverty if you are in the top 5% of I.Q. is 2%. Furthermore, when you take into account education and socio-economic status, the magnitude of the difference in probability of being in poverty is not much attenuated. How can this be if you can only explain 10% of the variance? It goes back to the ways in which logistic regression equations in which r-squared is not nearly as interesting as the magnitude of the effect that I.Q. has on the probability of being in poverty. And this applies across the range of the analyses we report.

Skeptic: Let's try to cut this another way. If you get so much predictive value from using intelligence just in I.Q. score, how much do you add by getting these other measures of socio-economic status or whatever. I.e., what's the sequence? What's the biggest predictor, and how much do you add by cranking in the others?

Murray: It depends on the phenomenon you are looking at. Once you introduce I.Q. the importance of socio-economic background is much attenuated. Often times the role of socio-economic background disappears altogether when I.Q. is in the equation. Conversely, introducing socio- economic background into the equation often times attenuates the role of I.Q. by only a very small amount.

Skeptic: Let's talk about cognitive stratification you discuss in Part I of your book. Secretary of Labor Robert Reisch gave a speech on the "Anxious Class," where he says: "Contrary to some theorists our destinies do not reside in our genes. Study after study show that skills can be learned. Every year of education or job training beyond high school, whenever it occurs in life, increases average future earnings by 6% to 12%. GNP is not simply a matter of DNA. Most Americans are on a downward slide not because of genetic deficiencies but because they lack the learnable skills to prosper in an economy convulsant with change."

But the picture Reisch paints is actually very similar to your own: a high-end cognitive class that is doing great, a bunch of worried people in the middle saying "where's that job my father had," and an underclass at the bottom that is falling downward in freefall.

Murray: Yes, and we make a reference to Reisch's work and we point out that he is more optimistic about the role of education than we are, but there is great similarities.

Skeptic: And on this subject of what society can do to bring up this underclass, you have been something of a godfather of the get-rid- of-welfare movement before you got into talking about I.Q. Conservatives like Newt Gingrich, Buchannon, Kemp, and Bennett, have also talked about getting welfare, but they have rejected The Bell Curve's analysis of I.Q. Does one of these follow from the other?

Murray: You know, in all 845 pages of The Bell Curve, we talk about getting rid of welfare in one sentence. We have a single sentence in Chapter 22 in which we talk about the ways government should get out of the business of encouraging any group of women to have babies, whether they be smart or dumb. And we generally urge that policies that subsidize birth be ended. It is one sentence. A single sentence. And one does not follow from the other.

Skeptic: Even if there was no inheritability to intelligence, no racial difference in any of these things, you would still be in favor of getting rid of welfare.

Murray: Yes, absolutely.

Skeptic: One question we might ask about your book is: why this book now, and why the controversy that surrounds it? Is this a case of a bad economy, an anxious public, and so we are blaming the victims and scapegoating those least capable of defending themselves?

Murray: Why is it published now? A better question is: why was it not published in the last 30 years? There has been a collective intellectual cowardice about understanding the role of intelligence in understanding social problems. Let's take one example. Child neglect is one of the most rapidly growing social problems we have. How many thousands of people make their living either writing about problems of child neglect and abuse, and so are advocating for new laws, etc.? Well, as every parent knows without reading anything about I.Q., there is a plausible relationship between intelligence and child abuse. Which is to say, any parent knows if a child has had a fever for 24 hours and hasn't been taking in liquids, you make a calculation that this has gone on too long and we've got to get this kid to a doctor, etc. Any parent knows that child-proofing a home takes foresight and thoughtfulness--it takes a certain amount of I.Q. With that plausible relationship in mind, the failure of social science and politicians alike to confront the possibility that low I.Q. is an important risk factor in child neglect is scandalous. Every single bit of evidence that does bare on this says that I.Q. is a great big factor in child neglect.

Skeptic: Couldn't someone take your arguments and say "we need more redistribution programs, not less, because these people cannot help themselves"? Haven't you knocked the bottom out from the conservative pull-yourself-up-by-your-own-bootstraps ideology?

Murray: You put your finger on something that Dick Herrnstein and I also thought about from the time we began working on the book. It is something that my friends on the right were concerned about. They said, "look, this type of material lends itself to all sorts of reasons to have a more interventionist state, and more welfare, and more redistribution, not less." We knew that. That is one of the major reasons for saying that it doesn't really hang together that this book was designed to foster a political agenda. It can be used by both sides.

The other point is that you have just described why Dick and I open up Chapter 22 with seven or eight pages of political philosophy. We say to the reader very explicitly "what we have just described for you could play out in any number of ways politically. Therefore what we are going to do is describe to you our own political predispositions, which have nothing to do with I.Q., why we hold them, and having told you those predispositions, then we will tell you our strategic view of what ought to be done."

Frank, I challenge readers of Skeptic magazine to go to any other book with policy recommendations by liberals which contain such an explicit, open, candid description of the author's political bias.

Skeptic: I'd like to go further on that limb. One might argue that The Bell Curve challenges the whole tradition that many people identify as American--namely equality. Do you find that your conclusions better fit a pagen vision of the universe that sees humanity as continuous with the rest of existence rather than as created in the image of God, and the Goddess fortuna working her wiles through DNA?

Murray: Our vision is Jeffersonian. Up until 30 years ago, in the early 1960s, Dick and I would have been describing a vision of America that everyone would have said, "of course." It is a vision in which we say that people bring different things to the table. The important thing is that everyone be given the opportunity to go as far as their own tempermanent, energy, characteristics, and intelligence will take them. The crucial factor in coming up with a harmonious society is not equal outcomes, it is abundance of opportunity. We are talking straight out of a tradition that until 30 years ago had virtually no intellectual challenge. It is only in the last 30 years that people have lost sight of those fundamental tenets of the American idea. And Thomas Jefferson would read The Bell Curve and, I like to think, nodding his head in approval. He believed there was a natural aristocracy that would make the republican experiment work. Personally I don't like the term "natural aristocracy" because I don't think the cognitive elite that we have now is all that great.

Skeptic: Along the these political lines, with your previous work in many circles you have been the intellectual darling of the conservative anti-welfare crowd. But now that your book has stirred things up, do you feel that your former allies and friends are now running away from you, and how do you feel about that behavior?

Murray: I assumed that when the book came out that a lot of people that used to think I was really neat would now say, "Charles Who?"

Skeptic: Has that happened?

Murray: I'm surprised at the extent that it has not. I thought that my political life would end. There seems to be a reflexive, almost deep inner panic, in an awful lot of people to be on the right side of The Bell Curve issue. And the right side is being perceived publically that you are shocked that these authors would suggest that intelligence has an important role in social problems; shock to think that anyone would still suggest there are differences among the races in intelligence. I've seen people, who I thought were both smart and honest, lie when it comes to the book.

Skeptic: Can you still be friends with these people?

Murray: No.

Skeptic: As I've said, at Skeptic were are skeptical of everything. Given your experience do you think that the American political process can deal with the fact that Homo sapiens is a biological species, subject to the same laws of evolution and genetics as other animals? Can a democracy deal with the information in The Bell Curve?

Murray: Actually, I'm optimistic on this score. This book has created in the news media a type of hysteria, where it has been denounced not just as wrong, but as evil and misguided. But there are now over 400,000 copies of the book in print, and as my wife points out, correctly I think, people do not plunk down $30.00 to buy a pseudoscience, racist track. They just don't do that. They are reading the book and talking about it.

I think what has happened to American intellectual life is that we have undergone a temporary aberation--30 years, short as these things go- -whereby we have tried to deny all sorts of realities about human biological characteristics. The best thing about this book is that these issues have been taken away from the ??chattering?? classes. They are now out there in public discourse in a way that is going to provide cover for a lot of good scholars who want to talk more openly about these issues but have been reluctant to. I'm Panglossian in my optimism.

Skeptic: What happened the last 30 years?

Murray: What happened in the 1960s, and now I'm citing from Losing Ground, was a fundamental change in the view of how society works and what individual responsiblity is, and this includes everything from education to law enforcement to the use of lawsuits, etc. It was a very widespread, but I think temporary, change that we are just now beginning to recover from and I think one of the lessons of this most recent election has nothing to do with people wanting a middle-class tax cut. It has to do with people wanting to return to a much more original view of how America is suppose to work.

Skeptic: Let me follow up on that. Hillary Rodham Clinton was in charge of the President's attempt to get a welfare reform, but it didn't go through. No one would say that was because she didn't have sufficient intelligence, energy, knowledge, whatever. When I see your idealistic vision of what you would like to have in America it doesn't seem realistic. You are being Panglossian.

Murray: I was Panglossian about these issues getting into the public dialogue. Now let me shift to being the pessimistic curmudgeon, of which I'm much more comfortable being! And that has to do with looking ahead to the long term. Hillary Rodham Clinton is the personification of what worried Dick Herrnstein and I about the cognitive elite. I'm sure she has a high I.Q. score. She, and for that matter her husband Bill, are both examples of people who by the age of 18 had been siphoned off into elite colleges and have spent the rest of their lives interacting with other people very much like them--the cognitive elite. And what happened in the Helm's bill [Frank: explain what this is] is a classic example of what happens when the cognitive elite has been talking to itself too long, and thinks it knows what's best for everyone. In this case, fortunately, they were derailled.

In the longer term what scares me is that the cognitive elite is, indeed, a powerful enough force to continue to rig the rules of the game. We are in favor of deregulation and decentralization, but I'm afraid the cognitive elite are going to make these things very difficult to carry out.

Skeptic: This bewilders me. You seem to say two different and possibly contradictory things. One, The Bell Curve finds the tremendous advantages that high I.Q. people have and which can be interpreted in a very elitist manner. Then, the libertarian Charles Murray emerges and says, well, the average Joe can run his life better than anyone else. How do you have it both ways?

Murray: Because running one's life is a matter of making all sorts of choices, and the satisfactions one achieves from running one's own life is inextricably linked with having been the person to make those choices. Someone with an income of $30,000 a year who made it himself I submit to you is a happier man than someone who got that same $30,000 unearned, whether it comes from welfare or trust funds. People running their own lives, taking responsibility for their own actions, that's the way human beings are wired to live satisfying lives.

Skeptic: So you really are a volunteerist on this. You are not a determinist. You are not saying everything is in the genes. You think free will is a meaningful concept.

Murray: Yes, and so did Dick Herrnstein, who was a student of B.F. Skinner.

Skeptic: Who didn't!

Murray: Yes, and Dick evolved a lot from his days as a behaviorist. One of the most difficult things to get across to people is that one may talk about genes playing an important role without being forced into anything resembling a determinist view of the world. But it is a contradiction only in this sense. The people who run their own lives are not necessarily going to make decisions that maximize anything in terms of some external source of comparison. In the Hillary Rodham Clinton world they might look at the things you have done or the choice you have made, and say, "no, no, if you would have done this other thing you would have had more money, you would have had more security, etc." I'm saying that a lot of the basis for deciding whether the decisions one makes in running one's life are right or wrong has nothing to do with these types of external criteria.

You've asked very difficult questions that are hard to answer in a few sentences . . . but they are good questions.

Skeptic: Well, that's what we tend to do at Skeptic. Is there anything you would like to add in conclusion?

Murray: I've enjoyed the interview. The only thing I would add is my own unhappiness at the way that Dick Herrnstein's name has been eclipsed. As I've said to Susan Herrnstein, she would not be pleased to have Dick being called all the names I have been called over this issue.

Skeptic: Yes, but he seemed to give more than he got in his lifetime.

Murray: I have confidence that in five years from now, and thereafter, this book will be seen as a major accomplishment. I also want it to be known that this collaboration between a political scientist and a psychologist is something I'm immensly proud of. Working with Dick was this wonderful experience of dealing with a man who loved and respected data, and respected the scholarly ideal of getting it right, absolutely right. And we think we did.

Bibliography

Gardner, Howard. 1993. Multiple Intelligences. New York: Basic Books.

Gould, Stephen Jay. 1981. The Mismeasure of Man. New York: Norton.

Herrnstein, Richard J. and Murray, Charles. The Bell Curve. New York: Free Press.

Kaus, Mickey. 1992. The End of Equality. New York: Basic Books.

Murray, Charles. 1984. Losing Ground. New York: Basic Books.

Reich, Robert. 1991. The Work of Nations. New York: Knopf.

Snyderman, Mark and Rothman, Stanley. 1990. The IQ Controversy. New Brunswick, N.J.: Transaction.

Sternberg, Robert. 1988. The Triarchic Mind. New York: Viking.

The Wall Street Journal. December 13, 1994. "Mainstream Science on Intelligence." P. A-17.