When science and ideology collide

By Greg Stevens on October 2nd, 2013

Not since the days of the civil rights movement have racial tensions been so fraught in America. One of the consequences of this is that deeply uncomfortable questions are being asked once again in public about the differences between different racial groups by some voices in our national debate.

Our first instinct when we hear ugly-sounding claims about, specifically, correlations between race and intelligence, is to shudder and look away in disgust. But what if there’s something in all this that we ought to know? And if there is, how do we find out for sure, and what do we do with that information?

Earlier this year, analyst Jason Richwine became the centre of controversy for his writings about race and IQ. He co-authored a report for the Heritage Foundation that predicted a cost of over $6 trillion for comprehensive immigration reform: a number that was arrived at, in part, based on the assumption that Hispanic immigrants would be a drag on the economy and would never rise above working class.

The assumption was seen as biased and partisan, but not particularly noteworthy, until the Washington Post uncovered numerous earlier talks and manuscripts in which Richwine had declared unambiguously that Hispanics can never be expected to be as successful as white people because they simply are “not as intelligent”.

His research, going all the way back to his dissertation at Harvard, targeted black people, native Americans and Latinos as being “low IQ” groups that he claimed could never be expected to reach parity with higher IQ groups, such as white people and Asians.

Although he resigned his position at the Heritage Foundation in the aftermath of the controversy, Richwine is reported to have been “amazed” that he was accused of being a racist “merely” for reporting quantitative scientific results.

As awkward and unpalatable as the controversy was when discussed by mainstream media outlets, the language used by the shock jocks of conservative talk radio was even more shocking. Over and over again, radio personalities invoked the spectres of “political correctness” and the “censorship of science”. “How very unfair it is,” they would proclaim, “that this poor scientist is being vilified simply for doing research on race and IQ”.

Nor was this the only time these drums have been beaten by pundits in American politics. There is a continual undercurrent in American right-wing media, made more visible from time to time by events such as the Richwine controversy, of the following type of argument: “Why shouldn’t we be allowed to ask whether white people are smarter than black people? Shouldn’t science be allowed to ask any question?”

Why shouldn’t scientists be able to research this question without facing ridicule?

For those in the political centre, and especially those who are not scientists by profession, this can seem like a reasonable complaint. Why shouldn’t scientists be able to research this question without facing ridicule?

And, setting aside the odious worldviews of many of these people, we have to admit that they have a point. If you are an idealist who believes that science is a quest for truth, then there has to be a better reason to reject a question out-of-hand than merely the creeping fear that the answer might be politically inconvenient, or simply very offensive or uncomfortable for some people to hear.

We are mapping the genome, discovering highly complex relationships between traits, skills and tendencies that we’ve never been able to measure before. And as our ability to delve into biological data continues to advance, we have to admit the possibility that eventually we just might come across some scientific fact that we find ideologically distasteful.

Putting aside even the specific issue of race and intelligence, it is possible that someday we may find some difference between two groups of people that we have been assuming, for political reasons, to be the same. If that happens, then what will it mean? And what will we do next?

Which questions are bad questions?

Let’s take what many readers will consider a deeply shocking question, one which, as racial tensions in America show no signs of slowing down, is increasingly discussed by talk show hosts and in whispers over the dinner table: “are white people ‘smarter’ than black people”?

Well, first of all, questions like that are truly unscientific questions, and the reason has nothing to do with censorship or political correctness. Unfortunately, in popular debates and casual conversations, the underlying problem with such questions is rarely pinned down.

What happens most of the time, instead, is people get caught up with the issue of motivation. People will say, “Why are you asking that question? Are you just some kind of racist, trying to justify your own prejudices?”

Indeed, this is exactly how the response to Richwine’s research played out in the media. The fact that he had spent years of his life asking “which races are smarter than which?” was construed as prima facie evidence that he was a racist who was merely seeking to justify his own deeply-seated prejudice.

Indeed, this conclusion might very well be true. But when it is the knee-jerk response to anyone trying to research issues of racial differences, we have to admit that it is presumptuous, petty, and partisan. To many people, scientists included, the accusation that anyone who asks “are white people smarter than black people?” must be racist can seem ridiculous. It can seem to justify the accusations that this is merely the censorship of science for political ends.

The problem with the question “Are white people smarter than black people?” is not motivation; the problem is with the way that the question is framed.

The motivation critique, however, is a red herring. The problem with the question “Are white people smarter than black people?” is not motivation; the problem is with the way that the question is framed.

Ultimately, “are white people smarter than black people?” is an unscientific question, in the same way that “are fancy cakes better than plain cakes?” is an unscientific question. There is nothing objective that you can actually measure to get a scientific result.

Why? Because categories like “white people” and “black people” are social constructs. Although there are some genetic markers that are somewhat correlated, and there are some physical features that are somewhat correlated, the distinction between “white people” and “black people” doesn’t completely correspond to any objective measure.

The same problem exists with the notion of “smarter”. After decades of scientific research and debate, psychologists have long given up the belief that there is even such a thing as a unitary measurement of intelligence.

Sure, as a practical matter, institutions still make use of IQ tests. But those who research “intelligence” know that the term is a kind of colloquial catch-all for a wide variety of different types of ability (mental speed, memory capacity, reasoning ability, and so on), all of which need to be measured separately if they are to have any scientific value at all.

So a scientist – a true scientist who is interested in researching a hypothesis that can be tested with measurable data – would never ask the question “are white people smarter than black people”. A true scientist, interested in objective fact and knowledge, would know that these terms are scientifically meaningless.

That is why Richwine’s research is roundly repudiated by the mainstream scientific community. The fact that Richwine’s research is scoffed at, or ignored, by scientists has nothing to do with “liberal bias” in science, and has nothing to do with “censorship”.

Richwine’s research is rejected by scientists simply because real scientists know that IQ is not a scientific measure, and groups such as “whites”, “blacks” and “hispanics” are not scientifically defined groups. Questions about the relationship between race and intelligence are quite simply not scientifically-framed questions.

What types of question might a responsible scientist ask, instead? A scientist might ask: “Do individuals with genetic markers X perform spatial rotation tasks more quickly than individuals with genetic markers Y?” where genetic markers X are more common among black people and genetic markers Y are more common among white people.

These are the questions that a scientist might ask, because these are questions that have concrete answers.

Or, a scientist might ask: “Do individuals who self-identify as ‘white’ on a survey have a larger working memory capacity than people who self-identify as ‘black’ on a survey?”

These are the questions that a scientist might ask, because these are questions that have concrete answers. These questions refer to very specific measurable experiments that can be carried out, and very specific well-defined methods for identifying the groups of people who are being compared.

Which answers are bad answers?

This still leaves open the possibility that science might find some conclusions that we don’t like. As an example, consider the hypothetical scientifically-framed question: “Do individuals with genetic markers X perform spatial rotation tasks more quickly than individuals with genetic markers Y?”

Leave aside, for the moment, the details of what these markers are. There is currently heated debate in the field of genetics as to what types of genetic markers are correlated with our casual conception of “race”. Just for the sake of argument, let’s take it as a given that some such marker exists that can be used to distinguish, frequently if not perfectly, between what we commonly think of as “white people” and “black people”.

It’s possible for people to do experiments to get a concrete numerical answer to this question. It’s also possible that if people perform those experiments, the results might show a difference between the two groups.

For example, it’s possible that they will find that, on average, people with genetic markers X are able to perform mental spatial rotation tasks 1.3 times faster than people with genetic markers Y, where that difference is statically significant.

I want to take a moment to emphasise a few points. First, to the best of my knowledge there are no studies that actually demonstrate such a difference. My choice of example, a difference in performance during spatial rotation tasks, is arbitrary and hypothetical.

Second, my choice of using genetic markers as an objective way to distinguish, even roughly, between “races” is also arbitrary and hypothetical. I could just as easily use some other measurement, including “self report” of racial classification on a survey.

Finally, I’m not entertaining this “what if” scenario because I want to push the idea that it’s true. Like many people, I actually would feel much more comfortable in a world where it were not true. But that is the very reason why it is important to consider the possibility.

There is a possibility that someday, someone will gather enough detailed data and perform a complex enough analysis, that they will be able to demonstrate, beyond a reasonable doubt, with no apparent flaws in experimental design or methodology, that there is some difference between two groups of people that we really wish, for political reasons, we hadn’t found.

To illustrate this, I’m suggesting the hypothetical example where a group of people with genetic marker X (which is present more in white people) performs spatial rotation tasks on average 1.3 times faster than people with genetic marker Y (which is present more in black people).

Hypothetically, from an ideological and political standpoint, what would that mean?

Would people use it to justify racism? Would it justify preemptively excluding black people from certain types of jobs, on the grounds that those jobs require speedy mental rotation skills? Would it justify setting up “separate but equal” education methods or public services for black people and white people, on the grounds that it would be unfair to have black people and white people compete with each other in an arena where one was at an actual genetic disadvantage?

All of these ideas seem creepy. The instinctive reaction of many people is to simply reject them out of hand, or say that we shouldn’t be talking about it. But what, then, is the correct answer? If a scientific result like this were to appear, should there be no policy consequences at all?

The merit of meritocracy

The standard progressive, socially-conscious response is that there should be no policy consequences at all. And it’s certainly easy to understand why.

The argument is simple: people should always be evaluated as individuals. In a true meritocracy, when a person applies for a job (or tries out for a sports team, or whatever) the decision about whether or not he is successful should be based on his own individual level of skill, not the average level of skill of some group that he belongs to.

Consider, for example, height or gender. It’s true that on average men are taller than women; however, there are plenty of specific individual men who are shorter than specific individual women. If a job requires a person who is six foot tall or taller, does it make sense to have a rule that says “no women allowed”?

Of course not. In a true meritocracy, each person will be judged on his or her own height, because height, not gender, is the thing that actually matters.

The same argument goes for the hypothetical result where genetic markers associated with white people are correlated with being better at performing mental spatial rotation tasks than genetic markers associated with black people.

Does it make sense to say that fighter pilots (who need to be exceptionally quick with their mental rotation abilities) should therefore be “whites only”? Of course not. A rule like that would arbitrarily exclude a great number of black people who have exceptionally good mental rotation abilities, and would conceivably lead to white applicants with lesser mental rotation abilities being accepted.

People should be judged purely as individuals. This idea of the “perfect meritocracy” is uncomplicated, and it sounds like the perfect solution.

However, as with most simple answers, there are potential problems. There are a number of cases where known biological inequalities are the justification for not having everybody compete together in a pure “meritocracy”. Consider, for example, women’s sports teams.

In many games, most notably American football, the real and significant differences in average physical size and strength between men and women mean that if both competed “equally” for positions on the same teams, women would almost never be able to play football in the NFL.

Allow me to be clear: there is no doubt that there are women out there who are just as fast and strong as male professional football players. Moreover, a good argument could be made that those women should, in principle, be allowed to play on the same field with the men who are their athletic peers and equals.

But the number of women who would be able to compete at that level would be much smaller than the number of men. There would forever be a large disparity, and a great number of excellent female athletes would never have the pleasure of being able to play professional football, if they were only able to compete for positions on mixed teams.

As a result, the progressive position in this case would be that a genetic inequality between two groups should lead to policy differences. The progressive position is to have both men’s and women’s professional football teams.

Even though it falls under the rubric of “separate but equal”, it is actually more socially conscious to create separate teams for women. This allows those players to experience the joy of playing professional ball, and their fans to experience the joy of watching those games.

In the end, there may not be a single cut-and-dry answer to this type of “what if” question. If science comes at us with an answer that we don’t like, it will take careful consideration. Is a purely individual meritocracy the correct policy response? Sometimes it will be, but sometimes it might not be.

The most responsible way to approach such issues is simply to be aware of what questions are scientific questions, what questions are not, and to know that even with the scientific questions there is no cookie-cutter policy solution that will apply across the board. And, of course, to resist firmly the entreaties of bigots who seek to manipulate scientific enquiry to support racist prejudice.