Monday, November 28, 2011

narrative is distorting/the mechanism is what matters

Via Ta-Nehisi Coates, I see that Andrew Sullivan is lamenting a purported blackout of research regarding the race-IQ connection. This is not new territory for Andrew; he helped bring the issue into the public consciousness back during his tenure as editor at The New Republic.

Now: the first issue here is the claim that such a blackout exists. As Coates points out, the evidence for such a blackout that is presented amounts to the complaint of a single researcher, Dennis Garlick. The researcher is someone who could reasonably claim expertise on the issue, and he appears from my limited vantage to have an impressive resume. His claim, though, seems disturbingly unsupported. Whether or not research into IQ and heredity is being squashed is an empirical question. Coates links to a blogger who attempts to answer that question empirically, and while we couldn't call it scientific, I find it a sober and constructive attempt to find out the truth. The results seem to speak for themselves, but I can't know what goes on behind the scenes. I'm unqualified to say if Garlick is right, but as a consumer of research I also don't find his case compelling.

The fact that empirical inquiry cuts against the grain of what Sullivan and Garlick are claiming is resonant in the context of the race-IQ question. The "race realist" movement has always pushed a narrative where politics corrupts empiricism, but the movement's failures have primarily been empirical failures. When you strip away the endless paranoid conspiracy theorizing and the relentless flogging of the narrative, you get down to a robust set of data demonstrating differences in performance on IQ tests, then some fairly wild speculation about genetic causes. Over and over again, assertions about the genetic undesirability of black people involve making massive leaps from an observed phenomenon to a particular mechanism to explain that phenomenon, with dubious or nonexistent evidence to support those leaps.

It's true: a broad swath of research demonstrates that black Americans tend to perform less well on standardized tests of intelligence. This racial achievement gap is not adequately explained merely by controlling for socioeconomic status, as is commonly assumed, although adjusting for poverty does shrink it. There's no need to hide from that data, as those positing genetic determinism constantly accuse others of doing. If a connection between heredity and IQ can be discovered, it should be. (Measure what is measurable, etc.) But what empiricism requires-- not political correctness, not bleeding heart compassion, not even basic human decorum and civility, but cold-blooded rational inquiry-- is far more than the racial determinists have show us. The narrative they present is seductive, which is precisely why their insistence on narrative over the complicated and limited claims of science is disturbing. From my perspective, most people who assert racial genetic deficiencies seem remarkably disinterested in identifying specific mechanisms for the observed phenomenon. They instead seem primarily interested in flogging crude and reductive visions of our society and what ails it.

The rush to find genetic origins for any and all human phenomena has become so popular, particularly with the press, that the standards of evidence have eroded everywhere. Genetic or evolutionary speculation has become an obsession of our media, frequently undertaken without a shred of scientific credibility, and defined by faddishness and imprecision. Take homosexuality and genetics. I find it remarkable the number of educated people who I meet who assume, quite confidently, the homosexuality (in both men and women) is purely and straightforwardly the product of genetic predisposition. This is a politically palatable idea-- one might call it PC-- but it can't yet be proven, even conditionally. There are complications, such as the (controversial) older brothers hypothesis, which is important because it posits a mechanism that is non-genetic and yet nonetheless physiological in origin (and thus not "chosen"), as well as other evidence contrary to the assumed genetic origins of homosexuality.

But as it became politically important for people to insist that homosexuality is genetic, that insistence became more and more prevalent. Never mind that the dichotomy between "homosexuality is purely genetic" and "homosexuality is a choice" is flagrantly false, or that "they can't help it" is not a stirring cry for equality. Politics made the genetic origin necessary, so people believed in it. Indeed, it's hard for me to imagine a scenario where politics has more directly corrupted popular understandings of empirical questions than the widespread belief that we know for a fact that homosexuality is genetic. Yet curiously, I don't find Andrew railing against that assumption, or insisting on the supremacy of disinterested research, or leading the battle for more open-mindedness in the attempt to explain the origins of homosexuality. Perhaps we will identify specific alleles that determine sexual orientation; I wouldn't be remotely surprised. But jumping from observed phenomenon to the assumption of genetic origins of that phenomenon with limited direct evidence of a specific mechanism is unhelpful.

Compare these speculative genetic causes of low IQ or homosexuality to, say, genetics and sickle cell anemia. We don't have the presence of a human condition and vague talk such as "it's genetic." We have identified the particular gene, in a particular chromosome, that causes the condition. We know how the mutation changes protein structure, which leads directly to specific consequences in gestation that cause the negative health effects we see in people with sickle cell anemia. We identified the alleles responsible for specific phenotypic traits and demonstrated the connection scientifically. At every step, we have gone beyond "it's genetic," in regards to sickle cell anemia, specifically and constructively. We have identified the mechanism which causes the condition. That's the job of those who are dedicated to racial determinism: find the mechanism. Do your work. Show me the data. Nobody is going to feel sorry for you when you fail to prove your assertions.

My question for Andrew and others is whether my dissatisfaction with the assumption of genetic origins for the racial achievement gap is necessarily "PC," particularly when placed in context with our knowledge about genetic phenomena like sickle cell. Is the narrative so powerful that we couldn't merely be unpersuaded by the evidence?

Speaking as someone who is involved, for 14+ hours out of a typical day, in reading, researching, and learning about education and pedagogy, let me engage in understatement and point out that education and intelligence are remarkably mulitivariate phenomena. My continuing frustration with the ed reform crowd is how relentlessly reductive they are in discussing the origins of poor educational performance. Saying "it's those damned unions!" and accusing any dissenters of obstruction isn't just politically unfair. It's an incredible failure to soberly assess the depth of our problems and the complexities of their origins. Trying to isolate specific variables in education and intelligence research is incredibly hard. That's not politics. It's reality. To ascribe genetic origins without greater explanation of mechanism or the exploration of environmental factors which shape IQ is to engage in wild-ass speculation.

My own wild-ass speculation? The question of race and IQ will be answered in a way that is complex, rather dull, and totally useless for providing headline fodder for sensationalist publications like The New Republic or Slate. I wouldn't be surprised if a whole slew of factors, including poverty, exposure to lead, poor diets, parent educational background, the idiom tests are written in, neonatal health care, learning disorders, dyslexia and dyscalculia, lack of exposure to educational toys and games, low childhood reading loads, the persistence of syntactic immaturity due to parental modeling (my own academic obsession), and other environmental factors played a role. That doesn't even begin to untangle the web of what "black" means in terms of specific linear heritage, particularly since we are talking about a truly unique genetic history that has been conditioned by the rape and forced breeding programs that are common to chattel slavery. If I'm right and the origins of the racial achievement gap are revealed to be a stew of competing factors, it will make our job of closing the gap harder, but it will also hopefully blunt the words of those who ascribe vast social problems to the supposed inherent inferiority of our most oppressed group.

I chose the example of sickle cell anemia purposefully, of course. It's a condition that is generally found in those with sub-Saharan African lineage. What does that mean for American blackness? Is sickle cell anemia "inherent" in black people? Is there something essential about the disorder in black people? They're absurd questions. Yet they are of exactly the same character as claims routinely made about black people and intelligence. As I said, my unsupported speculation is that a large number of factors contribute to the racial achievement gap. It's possible that one of them is a genetic predisposition. If so, we'll need to know what genes are actually producing this trait, and how. Then what? If we find such a predisposition, does that make low intelligence "inherent" to blackness? Does it mean we send black people off on a barge? Is this somehow an insurmountable challenge to liberalism, or to our social policies?

Racial determinists say that they want rationality and then engage in hysteria. The first step in assessing these issues is getting to the truth of the matter, and their dogged insistence about what we know exceeds their evidence and thus hinders that pursuit of truth. They then dig deeper, insisting on a slew of negative social conditions that stem from these supposed genetic deficiencies. It isn't surprising where the conversation next turns, although those who embrace these ideas continue to feign shock when they find racists involved in race science. (I want to loathe Stephen Metcalf, but his prudence, intelligence, and fairness in this piece makes it impossible, I'm afraid. Seriously, read it.)

I find the case for racial determinism currently unpersuasive. I find the notion of a research blackout unsupported. I find the discussion of racial lineage and genetic diversity reductive. I find the description of a specific genetic mechanism nonexistent. I find the idea of essentialized blackness offensive. I find the suggested consequences unsupportable and the supposed policy responses laughable. And I find the case for egalitarianism, equal protection before the law, and the assumption of equivalent human dignity totally unchallenged, whatever the reality about the racial achievement gap.

Of course, I'm not without considerable biases, and I couldn't tell you that I have inhabited a space of pure rationality when confronting this question. But this narrative of a refusal to learn the truth due to political correctness, so self-aggrandizing to those who push it, is not credible and does not serve the cause of empiricism. The pursuit of the controversial for its own ends is as distorting as the avoidance of it, and nowhere is that more true than here. Many people have attempted to marshal the evidence for the race-IQ connection for quite some time. Rather than evidence, they keep bringing us the narrative. Remember that.

17 comments:

Anonymous said...

Hear, hear.

Michael23 said...

Am I the only one wondering about the limited range of this debate? I mean just the assumption that it is genetics OR environment seems a bit odd to me - i thought the idea of epigentics, the nterplay between genes and environment and the on- and off-switch of some genes etc. might have trickled down into the mainstream media and yet the debate still lives on - thinly disguised as a debate about about political correctness and the freedom of research...

I'm totally with Freddie's "wild guess" - the dull, complicated and no silver bullet offering "solution" will be wildly ignored. And yet every media outlet will pick the one factor that is supportive of its pre-exisiting view. And there will be headlines. And there will be debates and they will tune out the scientific results in favor of a belief. And that is not how good science is promoted. That's how it is ignored and, ultimately, killed - as only people who cater to beliefs will get funding for their research.

Fake Herzog said...

Intelligence and the brain itself is too complicated (right now) to reduce to simple (relatively speaking) genetic mechanisms. But given the advances we are making, are you really convinced of your just so story above? Let's face it, we have plenty of evidence from history and adoption studies -- the reason black folks have lower IQs than white folks is partially genetic:

http://unamusementpark.com/2011/05/there-are-innate-race-differences-in-intelligence-a-flyer/

End affirmative action and help blacks excel at what they can do well (and stop freaking out about all the gaps because they ain't going away)!

Anonymous said...

What's odd about this topic is that the people who take the finding and run with it go straight to race. They never advance a race neutral policy of setting a school-or-gas-chambers cutoff for everyone. The race finding, not the IQ finding is what's important to them. Of course, they also ignore the fact that Koreans, Japanese, Ashkenazi and urbanites tend to be their betters by this metric.

Joe said...

Don't forget that all of this is predicated on a very narrow conception of intelligence in the first place. The notion that an IQ test is THE metric seems remarkably parochial. But then again, I use sociocultural theory to explain intelligence, so this whole conversation just troubles me in a Foucauldian kinda way.

Paul said...

The whole piece is good; the last 3 sentences are a knockout.

sweet tooth said...

Definable and measurable mental aptitudes may possibly exist, and if they do, genes probably help define their broad parameters, though not by 'expressing information' but through taking part in a developmental process. However, 'intelligence' most certainly does not exist. The concept is an illusion, one of our most cherished, right up there with the arbitrariness of the sound of words, and the culture:biology dichotomy. It's hardly unsurprising that, when a dominant and predominantly white social class seeks to discover who has the 'most' of it,they find themselves wearing the gold crown. If you can't bring yourself to lose the word, read one of the great neglected masterpieces of our age, The Perception of the Environment, by the Galileo of anthropology, Tim Ingold.

I like your polemical heat, Freddie, you are getting more concise and pungent with every post, but I would recommend you spend a few years reading out-of-print ethnographies, far from the buzz of the hive mind. You will come away infinitely enriched, and much better prepared to do battle against the endless spurious bullshit spouted by our prize-winning morons (i.e. paid-up intellectuals). I would do it myself but I have other fish to fry! Arm thyself with KNOWLEDGE, homie.

Megatron said...

Don't forget that all of this is predicated on a very narrow conception of intelligence in the first place. The notion that an IQ test is THE metric seems remarkably parochial. But then again, I use sociocultural theory to explain intelligence, so this whole conversation just troubles me in a Foucauldian kinda way.

Foucauldian? Sheesh! Hey, the 1990s called, they want their French theorists back.

Look, the argument your are making is a very common one, but it shows a lack of understanding of what IQ tests are measuring. And I don't blame you, because most people conflate IQ with intelligence. No psychometrician, even Charles Murray, would claim that IQ tests are the sole measure of what we commonly understand as "intelligence," which encompasses a great variety of human abilities. I mean, we all know that a brilliant sculptor might not be able to solve a simple math problem, or that a brilliant mathematician may be too lazy to get a job.


What IQ tests are designed to measure is called "g", or "general intelligence." G rates your ability to solve abstract problems. The higher G score you have, the more complex problems you can solve. It turns out that G so consistently correlates with so many important outcomes in academics, military, jobs, economics, ect... and can be measured reliably across so many different kinds of tests (even tests with no words or numbers), and is so persistent throughout the human lifespan, that it is indeed a very useful metric to predict someone's ability to perform certain mental tasks.

The SAT, GRE, MCAT, LSAT, and ASVAB are all essentially g loaded IQ tests. Even the Wonderlic (which is used by the NFL) is an IQ test. If IQ weren't important, if it was just some arbitrary statistical artifact, then it wouldn't be so widely used throughout every phase of the educational system. There needs to be a reliable way to screen people's mental abilities, and g loaded IQ tests are very good at doing this.

As far as race & IQ, I just don't know much about it to make a judgement. The existence of a gap between Asian, white, and black IQ is not really up for debate. The etiology of that gap, however, is. The social psychologist Richard Nisbett ("Intelligence and How to Get It") acknowledges that IQ is a relevant and accurate metric of cognitive function, and that it is at least a partially heritable trait, but argues that there is no racial difference in heritability.

L'Hote gives the issue a very fair treatment. IQ probably results from a very complex mix of genetics and environment. In my opinion, it is just not worth obsessing over. Different people have different abilities. Get over it!

Anonymous said...

Sullivan is so overwrought, sometimes it's just unbearable. And I love that Mr. DeBoer retains an open mind on the subject, because not taking seriously pseudo-scholarly race bait authored by self-admitted cross-burners and built upon the 'work' of 'academics' that also represent institutions the Southern Poverty Law Center consider hate groups would be incredibly unscientific.

Anonymous said...

Some background on the fringe hate gr... err... research organization that has underwritten 'scientific inquiry' such as The Bell Curve and its source material:

"[Current President of the Pioneer Fund, Phillipe] Rushton and Jensen came to my attention when [Bell Curve author, Charles] Murray fingered them, along with Lawrence Summers, as the impetus for his new Commentary article. The two published a "comprehensive survey" of evidence supporting The Bell Curve this past June in the journal Psychology, Public Policy, and Law. Murray—who leans heavily on Rushton and Jensen's work both here and in The Bell Curve—identifies this survey as being the "strongest argument" yet made by race realists.

Rushton has been retailing the idea of black inferiority for decades, though in two distinct styles: In pseudo-legitimate journal articles, he sounds a very Murray-like note of scholarly disinterest; at avowedly racist conventions, in front of the likes of David Duke, he argues that white women's birth canals are larger than black women's, allowing white women to give birth to larger-brained babies. In his 1995 book Race, Evolution and Behavior—now a race-realist classic—Rushton argued that "Negroids" are underevolved in comparison with "Caucosoids," because Caucosoids, having abandoned Africa for colder climates 110,000 years ago, were forced to develop their "intelligence, forward planning, sexual and personal restraint." Negroids, meanwhile, are characterized by smaller brains, larger genitals, sexual license, and lower IQs.

Rushton's work reads like a parody of 19th-century race phobia dressed up as 20th-century science...

What is the Pioneer Fund? The Pioneer Fund was founded in 1937 by Harry Laughlin and Wickliffe Draper...

...As superintendent of something called the Eugenics Record Office, Laughlin's testimony before Congress helped pass the Immigration Restriction Act of 1924. ("The Jew is doubtless here to stay," Laughlin confided to his associate Madison Grant, "and the Nordic's job is to prevent more of them from coming.") Meanwhile, as editor of the Eugenical News, Laughlin was an avid admirer of the German racial and sterilization policies being pioneeredunder Adolf Hitler. After Hitler signed the Law for the Prevention of Defective Progeny, Laughlin wrote an editorial praising Germany as one of the "great nations of the world" for its recognition of the "biological foundations of national character."...

...Draper obsessed over the "dysgenic" degeneration of America's Nordic stock. In late August 1935, Draper traveled to Berlin to attend the International Congress for the Scientific Investigation of Population Problems. Presiding over the conference was Wilhelm Frick, the reichminister of the interior who had administered the Nuremburg Laws. (Frick was hanged in 1946 for his crimes against humanity.) There Draper's travel companion and Laughlin's colleague and official conference surrogate Clarence Campbell delivered an oration that concluded with the words: "The difference between the Jew and the Aryan is as unsurmountable [sic] as that between black and white. … Germany has set a pattern which other nations must follow. … To that great leader, Adolf Hitler!" Three years later, when Draper paid to print and disseminate a book titled White America, a personal copy was delivered to Reichminister Frick.

Draper remained on the board of the Pioneer Fund until he died in 1972. A recent official history of the organization, produced in-house, describes him glowingly as "distinguished or aristocratic, tending toward the chivalrous," and as a disinterested funder of disinterested science. The book also praises Laughlin"

Fake Herzog said...

Last Anonymous,

Oh no, we all might get cooties from the Pioneer Fund!

Megatron,

You rock. Another excellent treatment of the subject can be found here:

http://www.lagriffedulion.f2s.com/g.htm

Anonymous said...

An excellent post, as usual. But I find a logical flaw. Your reasoning seems to be as follows:

(1) There is some evidence that there are social pressures on scientists not to research particular questions.

(2) Some scientists are using incomplete data on precisely these questions to make unwarranted claims (intelligence/homosexuality is heritable).

So, if the problem is that the current state of research doesn't justify the claims, shouldn't you be calling for much more research? You challenge the theorists you disagree with to prove the specific mechanism for heredity and intelligence, but surely you must know that doing so requires large amounts of funding. Which, according to many sources is not forthcoming.

Put more simply, you can hardly blame the heritable-intelligence people for not being able to prove a proposition that they are not being given the resources to prove. Can you?

Megatron said...

Fake Herzog,

Thanks, that's a great primer on g, though I have to confess some of the math is over my head.

One thing I think people don't understand is that, person to person, a few IQ points may not make a huge difference. For instance, the difference between a neurosurgeon with 130 IQ vs a 140 IQ may come down to traits like fine motor skills, emotional intelligence, and work ethic. The surgeon with a lower IQ may perform better.

But statistically, a few IQ points difference *at the population level* can make a HUGE difference. This is where it is worth exploring where these IQ difference come from. If we can raise average IQ a few points by reducing the parasite load in say, Cameroon, that may have a profound impact on the quality of life for the average Cameroonian (although this would be a very, very long experiment).

The way I see it, there are 2 main hypothesis, in a nutshell:

Hypothesis A:
Human Biodiversity: "IQ measures cognitive ability, which is a biological function. The distribution of IQ varies between populations, and it is a partially hereditable trait."

Hypothesis B:
Human Neuronal Uniformity: "IQ measures cognitive ability, which is biological function. The distribution of IQ varies between populations, but the distribution is entirely a function of the environment."

Given what I know about biology and statistics, Hypothesis A seems more likely, although the "race realists" could conceivably use it to push a dangerous political agenda.

Careless said...

It's hardly unsurprising that, when a dominant and predominantly white social class seeks to discover who has the 'most' of it,they find themselves wearing the gold crown

... But that didn't happen.

JL said...

But what empiricism requires-- not political correctness, not bleeding heart compassion, not even basic human decorum and civility, but cold-blooded rational inquiry-- is far more than the racial determinists have show us. The narrative they present is seductive, which is precisely why their insistence on narrative over the complicated and limited claims of science is disturbing. From my perspective, most people who assert racial genetic deficiencies seem remarkably disinterested in identifying specific mechanisms for the observed phenomenon.

Which researchers are you referring to? Which papers? What "narrative"? I think Rushton and Jensen's 2005 review, for example, is an exemplary piece of empirical research. They present ten different lines of evidence in favor of the thesis that the black-white IQ gap is largely genetic.

In fact, it is Jensen and his collaborators who have always been the ones interested in empirical research, whereas their detractors have relied on lachrymose narratives about racism, slavery, Nazism, etc. Similarly, when Sullivan pointed out that there are racial differences in IQ and they should be studied, everyone's attacking him as a racist and Nazi, while you claim that he is an irrational hysteric. You are proving his case.

There are very few researchers engaged in research along Jensen's lines, and there's very little funding available for this research. There have been attempts to shut down even the meager sources of funding they have, mainly the tiny Pioneer Fund. In light of this, it's remarkable how strong a case the hereditarians can make, and how the anti-hereditarians have failed to pinpoint the environmental factors that supposedly explain the gap despite all the political support and funding they have.

There should be more research on race and IQ, and there are methods that could potentially give a definitive answer. But the sort of research that would further the debate requires large sample sizes and massive amounts of genetic sequencing. Who's going to sponsor such research?

Is sickle cell anemia "inherent" in black people? Is there something essential about the disorder in black people? They're absurd questions. Yet they are of exactly the same character as claims routinely made about black people and intelligence... we'll need to know what genes are actually producing this trait, and how.

No, they're not comparable traits. Sickle cell anemia is a simple monogenic disorder, whereas intelligence is a complex trait influenced by hundreds or more genes. It's absurd to claim that nothing can be said about the genetic basis of intelligence differences before we know all the causal variants and how they function. See Rushton and Jensen's paper above for details.

And I find the case for egalitarianism, equal protection before the law, and the assumption of equivalent human dignity totally unchallenged, whatever the reality about the racial achievement gap.

Who doesn't?

keatssycamore said...

As long as folks are posting explanations of g, here's a debunking:

The Myth of G:

"... the case for g rests on a statistical technique, factor analysis, which works solely on correlations between tests. Factor analysis is handy for summarizing data, but can't tell us where the correlations came from; it always says that there is a general factor whenever there are only positive correlations. The appearance of g is a trivial reflection of that correlation structure. A clear example, known since 1916, shows that factor analysis can give the appearance of a general factor when there are actually many thousands of completely independent and equally strong causes at work. Heritability doesn't distinguish these alternatives either. Exploratory factor analysis being no good at discovering causal structure, it provides no support for the reality of g."

R. Jones said...

Here's a 2010 point-counterpoint paper between Nisbett and Rushton/Jensen. You can decide for yourself whether the pure environmentalist or the genes+environment model makes more sense.

Greg Cochran:

"Like the Chinese and Indians, some ethnic groups show mediocre results in their benighted homelands and better results in Western countries. On the other hand, other groups do poorly everywhere. People whose ancestors are from sub-Saharan Africa produce very few scientific papers – on average – no matter where they live. This is also the case (to a lesser extent) for mestizos from central and South America, for Filipinos and Malays, and most of the Indian castes. You don’t see much out of the Middle East, either - although Armenians are an exception, and there may be others. Interestingly, Armenians were generally thought (by themselves and others) to be smarter than the average bear back in Byzantine times."