Shades of Grey in a Black & White Issue.
September 5, 2005 1:29 AM   Subscribe

The Inequality Taboo - Charles Murray defends his ideas, published in the controversial book The Bell Curve.
posted by Gyan (71 comments total)

 

We're all eeeeeeequal.
posted by uncanny hengeman at 1:40 AM on September 5, 2005


If we are equal, then why does your "equal" have more 'E's than mine? :(
posted by TwelveTwo at 1:48 AM on September 5, 2005


Taboo is too stronge a word. People do research gender diffrences, but they publish them in a non-inflamatory way, see the Pinker vs. Spelke debate.

To apply a few of their facts: Murry may be correct that race based affirmative action does little good, as objective highering criteria lead usualy to equality. When they don't, its reasonable to loook for innate diffrences or prejudices; and these could be combated by social pressure in the case of race discrimination. However, when assessing potential, people are "intractably" prejudiced towards males over females. Here an exprement which should be done:

Take some mathematics and linguistics (or any female dominated field) papers, across the spectrum, ask professors to rate them 1-10 butrandomly set the name to male or female. You will see an average 1 point advantage for the male it mathematics, but I expect the same 1 point advantage will occur for males in the female dominated field. Which would suggest that this discrimination, while "wrong", is not the source of the real diffrence.

The single biggest enviromental facters controlling female presence in math and physics are likely to be single sex education (which significantly improves the preformance of women), and how those fields present themselves to the public. In particular, maths does not present itself as being a safe career choice (males are far more likely to take risks than females; most female matehmaticians have an older realtive who is a mathematician).
posted by jeffburdges at 2:19 AM on September 5, 2005


Anyway, my point is that there are lots of people talking about sex diffrences, but they are saying "Hey, lets try single sex education" or "Hey, lets try making the subject more friendly". They are not denying that gender discrimination exists and they are not antagonizing the (lefty) moralists by saying affirmative action can't work. They are trying to support ideas which can work.
posted by jeffburdges at 2:24 AM on September 5, 2005


Just wondering: how is it possible, when working on human populations, to differentiate in a statistically sound way between genetic causes and environmental/cultural ones, particularly on traits that are definitely linked to environmental and cultural circumstances (like social achievement) ?

This can be done on animals and plants because one can design experiments to do just that, and it's already not that simple even on "technical" traits such as milk production or crop yields). But for humans one would need some alternate history, for instance one without slavery or colonisation. I have not read the Bell Curve, but this question is always on my mind every time a scientist comes up with "genetic" explanations of this or that cultural/behavioural human trait.
posted by elgilito at 2:27 AM on September 5, 2005


I agree in principle that a more rational discussion of group differences would be beneficial.

However, he seems to lapse heavily into 'just so' stories, and blur the distinction with cultural differences. For example he seems to imply that the reason some Arab immigrants don't fit in could be genetic. I would have thought that was a classic case of cultural differences. And the bit about Italians being more 'vivacious' than Scots - try a Scottish winter..

The arguments on both sides seem to split into 'genetic' and 'cultural/racism blah blah', but I haven't seen anything about individual parenting. I think mental capacities can be inherited, but I'm sure it makes a difference what you do with kids when they're young - what you give them to play with and read. I'm not completely convinced that 'genius will out'
posted by lunkfish at 2:41 AM on September 5, 2005


In the humanities, the most abstract field is philosophy—and no woman has been a significant original thinker in any of the world’s great philosophical traditions.

You know, he probably shouldn't say that around my roommate, who considers one of her female professors to be one of the most original thinkers in the department. He might soon be missing body parts.

Also, Teresa D'Avila and Catherine of Siena are both considered to have been major Catholic thinkers - they have been made Doctors of the Church. I guess they must have been man in drag.
posted by jb at 2:45 AM on September 5, 2005


To add to my previous point: in my field of work, agricultural engineering, the number of female students in the university where I work has grown from 0% in the 60s to 40% when I was a student to close to 60% now, without any sort of affirmative action (quite the opposive in fact, since the field was predominantly male and intended to stay that way). Other fields in engineering didn't follow this pattern, so what made agricultural engineering (that requires both maths and physics) attractive to these young women? No matter what, I doubt that the answer has any sort of genetic component.
posted by elgilito at 2:48 AM on September 5, 2005


(males are far more likely to take risks than females; most female matehmaticians have an older realtive who is a mathematician).

This logic doesn't really hold. Careerwise, going to graduate school for literature is much more risky, and those departments have more women.

This is purely anecdotal, but I was once discussing this in a class, where many of the young men said they felt under society pressure to go into "useful" science and engineering fields, to support a family. Whereas, as a female, I was never felt that way and went where my aptitude and interests were. That said, BAs actually have a better chance of getting a job in Ontario than Engineering majors, so maybe that wasn't such a bad choice.

Men are not genderless - they have an experience that is shaped by gender expectations and constraints just as much as women.
posted by jb at 2:53 AM on September 5, 2005


What is Murray's operational definition of "Black"? Po-mo gender relativism aside, it's not difficult to assign the vast majority of people on this Earth to either the "male" or the "female" group according to objective biological criteria. But what is the objective biological criteria for the category "Black"?

He talks about "genetic markers" that can accurately classify people according to their self-identified ethnic category, but DNA is mix-and-match; is he suggesting that these genetic markers are inextricably linked to the (no doubt dozens of) genes that determine innate intelligence potential, in a way that keeps the genetic markers and the intelligence connected through the DNA recombination process?
posted by luckywanderboy at 2:55 AM on September 5, 2005


Guns, Germs and Steel is well worth a read while we're on this topic. Looking at everything from available crops to domesticable animals to geographic factors, the author shows how, on almost every front, the Eurasians held nearly all the cards. As he speculates, had been possible to domesticate the rhino - invoking the intriguing image of Bantu tribesmen riding into the middle east on rhinoback - then history could have been very different.
posted by rhymer at 3:05 AM on September 5, 2005


lunkfish : "For example he seems to imply that the reason some Arab immigrants don't fit in could be genetic."

He doesn't. He says, "Let us start talking about group differences openly—all sorts of group differences, from the visuospatial skills of men and women to the vivaciousness of Italians and Scots. Let us talk about the nature of the manly versus the womanly virtues. About differences between Russians and Chinese that might affect their adoption of capitalism. About differences between Arabs and Europeans that might affect the assimilation of Arab immigrants into European democracies. About differences between the poor and non-poor that could inform policy for reducing poverty.". This part of the essay isn't implying a genetic basis to the difference. It's just a call for discussion.

jb : "You know, he probably shouldn't say that around my roommate, who considers one of her female professors to be one of the most original thinkers in the department. He might soon be missing body parts."

Well, he's referring to the 'world's great traditions'. That lack may be pegged to social factors. He clearly states that individuals occupy a range - from high to low. Group differences relate to the mean.

luckywanderboy : "What is Murray's operational definition of 'Black'?"

Self-identification, I would guess. He uses this study as proof that such identification is reliable. That study relates genetic clusters of 350 microsatellites to self-identification. The black human skin color seems tied to mutations in a single gene.
posted by Gyan at 3:16 AM on September 5, 2005


I think there's a lot to critcise about the article, but you have to admit that there is a taboo on discussing any sort of racial difference, more so than gender difference.

I accept that there's no clear cut line between 'White' and 'Black' say, but that doesn't mean we can't ever say anything useful about race and genetics.

He makes a good point here

We would be free to talk about other sexual and racial differences as well, many of which favor women and blacks, and none of which is large enough to frighten anyone who looks at them dispassionately.


The last part is important - a difference in 0.5% over a big group shouldn't make a difference to how you treat any one person.

I think we have to have a balanced view - this kind of article needs to be picked apart carefully - but I don't like the knee jerk lock down on looking at racial differences that predominates.
posted by lunkfish at 3:31 AM on September 5, 2005


jb: yes "risk" is the wrong word. But their is some "perhaps humanizing" effect of having a realitive who is a mathematician, which has a major impact on a girl's probability of becoming a mathematician. It might not be a "risk assessment" as there are many other gender diffrences, but life choices are far more complex than your graduate school comment takes into account. In particular, an undergraduate math major has very goood job prospects outside of mathematics, meaning they can quit easily. At the same time, discussion of mathematics tends to either (a) be iinclusive of people who have no chance in the field, or (b) focuses on the extreme difficulty at the high end.
posted by jeffburdges at 3:46 AM on September 5, 2005


This part of the essay isn't implying a genetic basis to the difference. It's just a call for discussion.

Fair enough, but he could be more clear about the difference between national stereotypes etc. and genetic difference. If it gets muddied, it will make it easier for people to confirm their racial/national prejudices, and will get the debate shut down.
posted by lunkfish at 3:46 AM on September 5, 2005


Just wondering: how is it possible, when working on human populations, to differentiate in a statistically sound way between genetic causes and environmental/cultural ones, particularly on traits that are definitely linked to environmental and cultural circumstances (like social achievement) ?

Answer: it isn't. Any serious behavioral scientist will confirm that socialization is directly implicated in neurological development, from before birth (we are born ready to recognize the phonemic distinctions in our mother's language; if our fathers speak a different language, we aren't nearly as far along; this is because we hear in utero). So this is in an excellent and basic point, which immediately refutes 90 percent of the nonsense that people like Murray spout.

The man is an utter fake.
posted by realcountrymusic at 4:58 AM on September 5, 2005


In the humanities, the most abstract field is philosophy—and no woman has been a significant original thinker in any of the world’s great philosophical traditions.

Does Any Rand count? She did write 'the Virtue of Selfishness'.
posted by Alison at 4:59 AM on September 5, 2005


no rand whatever counts. That's the rules.
posted by alloneword at 5:05 AM on September 5, 2005


Furthermore, colloquial "racial" distinctions ("blacks," "latinos," etc.) have very little basis in genetics. A huge problem even with legitimate (as opposed to Murray) sociobiology is the tendency for its proponents (and even more, its readers and critics) to map colloquial (and largely socially constructed, as for example the days when the Irish and Italians were not considered "white" in the US) "racial" categories onto some faint understanding of the naturalistic bases for group (or individual) differences. Most population biologists don't even think of "race" as a useful category.

Every few years, these clowns resurface. What a strange conincidence that Murray reappears right on the heels of Katrina, and an emerging and very ugly debate about the human worth of Katrina's victims, especially her African American victims. How convenient to remind us all that they weren't really worth saving after all, right?

I smell a rat. And in Commentary, no less.
posted by realcountrymusic at 5:06 AM on September 5, 2005


PS - is Derek Jeter "black" or "white" for the purposes of assessing his intelligence and ability?
posted by realcountrymusic at 5:08 AM on September 5, 2005


I certainlyh do not have the background to make any intelligent statement about this post. However, in any and every sensible book I have read on the Nature versus Nurture (enviornment/genetics) issue, I note that all the writers mention that these things are never isolated but act together and also interact at the same time.
posted by Postroad at 5:15 AM on September 5, 2005


Without p-values or confidence intervals, the assertions he makes are meaningless.

To make such sweeping statements he had better have had unimpeachable statistics -- instead, he ignored them altogether.

Move along, nothing to see here.
posted by NucleophilicAttack at 5:48 AM on September 5, 2005


I think there's a lot to critcise about the article, but you have to admit that there is a taboo on discussing any sort of racial difference, more so than gender difference
Yeah, right, it's so "taboo" that Murray has been able to sell millions of his "Bell Curve" book and received tons of press coverage for his theories: what a martyr, languishing in penniless obscurity!

I doubt any of those who insist on pushing this ridiculous "taboo" line would be quite so eager to step up to the plate if Murray were propagandizing instead for the "Jews are genetically [insert negative epithet here]", but when it comes to engaging in speculation about the supposedly "innate" shortcomings of black people, it's suddenly time to start crying about the mythical PC Police.
posted by Goedel at 6:08 AM on September 5, 2005


Postroad: I note that all the writers mention that these things are never isolated but act together and also interact at the same time.

I used to work in (farm) animal genetics and for us, the basic equation was like this : P = G + E where P is the phenotype (or the observable, measurable traits), G the genotype (the part driven by the genes) and E the environment (sensu largo). Estimations of G and E (and of heritability h² = variance (G)/variance(P) that measures how much variation of P is due to variation of G) are obtained either by specific experiments of by calculations carried out on huge populations (for instance on national data from dairy or meat production control that also include all the genealogical information on sires, dams, siblings etc.).

These estimations work ok on relatively simple traits where the biological basis is well established. Even then artifacts/skewering in the population distribution or bad experimental design or accidental circumstances can ruin the calculations and make them useless and nonsensical. Whenever the traits become more behavioural or generally when environment becomes really important, the technique tends to become murky and error-prone. Applying these techniques to humans (where trait measurement itself is an issue because it's often socially constructed) is asking for trouble and frankly, interpretation of such "experiments" seems more like palm-reading than like real science. I'm not saying that it's theoretically impossible, but from an experimental point of view, people are just not good subjects. Similar criticism can be done to regular medical studies carried out on large populations, but at least in these cases the traits (weight etc.) are often more objectively defined.
posted by elgilito at 6:19 AM on September 5, 2005


You know, he probably shouldn't say that around my roommate, who considers one of her female professors to be one of the most original thinkers in the department

You can't throw a teaser like that out there without giving us a name!
posted by IndigoJones at 6:25 AM on September 5, 2005


(Well, I guess you can, as you did. But really, you oughtn't. Who's the prof?)
posted by IndigoJones at 6:26 AM on September 5, 2005


The man who cannot occasionally imagine events and conditions of existence that are contrary to the causal principle as he knows it will never enrich his science by the addition of a new idea. (Max Planck)

Isn't the context of how we view Murray's science dependent on the viewing framework we bring to bear upon it, much as Nijinsky's leap through the window would be experienced differently by a physicist and a choreographer?

The most exciting phrase to hear in science - the one that heralds new discoveries - is not "Eureka!" but "That's funny..." (Isaac Asimov)

If we turn off our curiosity for fear of being called misogynist, racist or anything else, how will we learn anything new?
posted by Enron Hubbard at 6:32 AM on September 5, 2005


Charles Murray is a nincompoop.
posted by OmieWise at 6:36 AM on September 5, 2005


And a nincompoop who is bound and determined to get everybody to agree with him that blacks are dumber than whites. Not that there's anything wrong with that! Dumb people have their place in the scheme of things!

Also, everything realcountrymusic said.
posted by languagehat at 6:45 AM on September 5, 2005


"Original" thinkers are a dime a dozen; those that have a wide and lasting influence are a bit more scarce.
posted by Wolfdog at 6:51 AM on September 5, 2005


"In the humanities, the most abstract field is philosophy—and no woman has been a significant original thinker in any of the world’s great philosophical traditions."
Except for Hannah Arendt, who's, like, THE motherfucker in studies of totalitarianism.

"In the arts, literature is the least abstract and by far the most rooted in human interaction; visual art incorporates a greater admixture of the abstract; musical composition is the most abstract of all the arts, using neither words nor images."

Really? How about this formulation: literature is the most abstract and least rooted in human experience, because it involves using high-level abstracted symbols to communicate meaning, which requires a tremendous amount of deviation from direct sense experience.
I mean, as long as we're shovelling bullshit, why's it have to be your bullshit?

"The role of women has varied accordingly. Women have been represented among great writers virtually from the beginning of literature, in East Asia and South Asia as well as in the West."

But are outnumbered about 1000-to-one. Do any of his points stand upon reflection?

"Women have produced a smaller number of important visual artists, and none that is clearly in the first rank."

Really? Georgia O'Keefe. Sally Mann. Diane Arbus.
Oh, it's none that his art-ignorant ass considers first rank! I see now!

"Social restrictions undoubtedly damped down women’s contributions in all of the arts, but the pattern of accomplishment that did break through is strikingly consistent with what we know about the respective strengths of male and female cognitive repertoires."

Uh... You mean that aside from a very few women who held social rank and resources to become art educated, that the vast, vast majority of trades and educations went to men didn't have anything to do with the preponderance?

God, does he set up anything where he's not just begging the question? I couldn't read anymore after this was SOOOOO DUMB!
posted by klangklangston at 7:07 AM on September 5, 2005


What took him so long to defend himself? Was he waiting for Stephen Jay gould to die?
posted by dial-tone at 7:31 AM on September 5, 2005


Just wondering: how is it possible, when working on human populations, to differentiate in a statistically sound way between genetic causes and environmental/cultural ones, particularly on traits that are definitely linked to environmental and cultural circumstances (like social achievement) ?

Well, Murray didn't even try with the bell curve. In fact, he just assumed everything was genetic. Seriously, there is no weight given to environmental factors at all. This is the fundamental weakness of his book.

What is Murray's operational definition of "Black"? Po-mo gender relativism aside, it's not difficult to assign the vast majority of people on this Earth to either the "male" or the "female" group according to objective biological criteria. But what is the objective biological criteria for the category "Black"?

Self identification. He uses data from a military aptitude test.

From a quote of Murray posted by Gyan

[let us talk about] About differences between Russians and Chinese that might affect their adoption of capitalism.

Holy crap, is he kidding. Is he seriously going to attribute the actions of a handful of people over the last 50 years to the genetic 'predisposition' of the nation they governed during that time? This guy is a nut.

It's sad mefi is giving space to this idiotic racist.
posted by delmoi at 7:35 AM on September 5, 2005


Researcher publishes research that goes against mainstream opinion, gets hounded by mob. Film at 11.
posted by spazzm at 7:58 AM on September 5, 2005


From The Great Gatsby:

“Civilization’s going to pieces,” broke out Tom violently. “I’ve gotten to be a terrible pessimist about things. Have you read ‘The Rise of the Colored Empires’ by this man Goddard?”

“Why, no,” I answered, rather surprised by his tone.

“Well, it’s a fine book, and everybody ought to read it. The idea is if we don’t look out the white race will be—will be utterly submerged. It’s all scientific stuff; it’s been proved.”

“Tom’s getting very profound,” said Daisy, with an expression of unthoughtful sadness. “He reads deep books with long words in them. What was that word we——”

“Well, these books are all scientific,” insisted Tom, glancing at her impatiently. “This fellow has worked out the whole thing. It’s up to us, who are the dominant race, to watch out or these other races will have control of things.”

“We’ve got to beat them down,” whispered Daisy, winking ferociously toward the fervent sun.
posted by johngoren at 8:14 AM on September 5, 2005


Title IX of the Educational Amendments of 1972 assumes that women are no different from men in their attraction to sports. Affirmative action in all its forms assumes there are no innate differences between any of the groups it seeks to help and everyone else.

Um, no. Such assumptions are not required to support either program. One can "assume" as weak a claim as "obseved differences in athletic participation by sex are influenced by institutional inequities" and fully support Title IX. Similarly, one need only believe that the non-innate differences have a more significant effect on the content of résumés in an applicant pool than do any innate differences to think that affirmative action of some sort might be warranted. Murray has set up a mighty flimsy straw man to knock down, it seems to me.
posted by dilettanti at 8:16 AM on September 5, 2005


Researcher publishes research that goes against mainstream opinion, gets hounded by mob

Um, "research" publishes "research" that goes against the vast preponderance of real research on the subject, gets laughed out of science. Do we really need to revisit this? Murray's work has been shown to be shallow pseudoscience by scores of top scientists, and that was a decade ago. That he hasn't modified his methods, his opinions, or his interpretation even slightly shows science really has nothing to do with it. He's a racist ideologue, and enjoys the accolades of his fellow racists who find their own curdled beliefs confirmed by his "science." It's a lot like the "intelligent design" nonsense or the AIDS-denial hucksters. The fame and notriety of being a "crusading anti-establishment figure" with fervent supporters willing to confront and abuse your legitimate detractors goes to these guys' heads.
posted by realcountrymusic at 8:22 AM on September 5, 2005


Race is attributed culturally. There are genetic components that influence what we perceive to be race, like the color of skin and type of hair. In the end, race is not tied to biology. It is not a subspecies of Homo Sapiens.

Ethnicity is cultural and often infers your country of origin. A person from Russia isn't Russian by race, they are Russian by ethnicity. It appears that Murray (and some of you) have confused race and ethnicity. The distinctions are important.

Murray appears to have ignored an entire body of work and several branches of science dedicated to the scientific study of these issues. For starters, class, not race, ethnicity or gender, has been repeatedly shown to be the single largest contributor to whether or not one is successful in school or life.

Second, the author doesn't appear to measure the affects of institutional racism on what he perceives to be race. It's a critical, measurable factor in the outcome of his field of study. Is he comparing individuals from the same class? Do they have the same income and the same wealth? Studies show that if you compare a group of self-identified blacks and whites with similar incomes and similar wealth that there is no measurable difference in things like education, class and other life rewards. If you were to compare wealthy self-identified blacks with poor, self-identified whites, you see the same difference The Bell Curve attempts to attribute to genetics, except entirely in reverse. Maybe Murray could benefit from an introductory course in sociology.

The Bell Curve is an apologia for racism. It's the Intelligent Design of social sciences. It's not a book bore out by science, it's a book that markets a preconceived idea to people who want some sort of reassurance that their racism is well founded.
posted by sequential at 8:29 AM on September 5, 2005


According to my fantabulous undergrad advisor (Harriet Lyons; she discusses these points among others in the book she published with her husband), The Bell Curve heavily relies on Phillipe Rushton for many of it's references. Rushton published a study suggesting that genital size was inversely proportional to intelligence, and that gential size was distributed racially: from largest to least we have black, white, asian. If you trace Rushton's sources, you come up with a piece of Victorian pornography published as a "scientific" account to avoid anti-pornography laws.

Gould pretty much destroyed the entire premise as well as the statistical work of The Bell Curve in The Mismeasure of Man. The groundwork for Gould's work was laid in the early 20th century by anthropologist Frans Boas, who (interestingly to this discussion) trained some of the major female minds in anthropology: Zora Neale Hurston; Margaret Mead; and Ruth Benedict for a few.

I bring all this up because I think no discussion of either Murray or Rushton is complete without pointing out that they use a fictional work of Victorian pornography as a scientific source to back their positions (either directly or indirectly). I think that once you do that, the discussion is over.
posted by carmen at 8:41 AM on September 5, 2005


Just wondering: how is it possible, when working on human populations, to differentiate in a statistically sound way between genetic causes and environmental/cultural ones, particularly on traits that are definitely linked to environmental and cultural circumstances (like social achievement) ?

There are plenty of statistical approaches scientists use to reliably disentangle the relative contributions of (a) genes, (b) shared environment (e.g., parental influences), and (c) non-shared (i.e., unique) environment to a particular trait. Typically these rely on twin or adoption samples. For example, the contribution of shared environment can be estimated by comparing the similarity between identical twins raised together to identical twins raised apart. It turns out, that for most aspects of personality and cognition, the variance is attributable about 50-50 to genetic and non-shared environmental causes. Familial environment has little or no detectable effect on personality or intelligence, counter-intuitive as it may seem.

That, by the way, is not controversial within behavioral genetics. It's a finding that's been appreciated for several decades and has been replicated hundreds of times. Unfortunately, most lay people simply refuse to believe it because it flies in the face of common intuition.

For a nice overview of behavioral genetics methods and applications, see here.f

Answer: it isn't. Any serious behavioral scientist will confirm that socialization is directly implicated in neurological development, from before birth (we are born ready to recognize the phonemic distinctions in our mother's language; if our fathers speak a different language, we aren't nearly as far along; this is because we hear in utero). So this is in an excellent and basic point, which immediately refutes 90 percent of the nonsense that people like Murray spout.

Wrong. As I noted above, socialization (at least in the sense of shared environmental influences such as parental environment) has almost no detectable effect on 'neurological development'. Again, that's counter-intuitive, but not controversial. I resent your appeal to the authority of 'serious behavioral scientists' when you'd have a hard time finding behavioral geneticists who think shared environmental influences have any meaningful impact on intelligence after childhood.

Also, it's been established that up to about age 1, infants can recognize differences between all phoneme distinctions humans are capable of making. That ability appears to dry up just as infants begin to learn the language(s) they're exposed to.
posted by heavy water at 8:59 AM on September 5, 2005


Title IX of the Educational Amendments of 1972 assumes that women are no different from men in their attraction to sports.

No, it assumes that state-sponsored, state-funded university scholarships have to be sex-neutral.
posted by ROU_Xenophobe at 9:07 AM on September 5, 2005


Murray may have been able to find proof that boys are better on math tests than girls, race has influence as well, and social class doesn't matter [though that one's a biggy, because social class is such a major factor in explaining differences it is often not even taken into account]. That still doesn't explain why all American students do so piss poor on international math tests compared to those from other countries.

Statistical models are only a simplified way to look at reality.
posted by ijsbrand at 9:13 AM on September 5, 2005


heavy water, one problem with twin/adoption studies is that the children are raised in essentially the same broader social context. I've never heard of a twin study where one was raised in a really significantly different social context (like a Western verses a non-Western country)*. So, while twin studies may show that individual family dynamics don't have a huge impact on personality, they aren't able to address how broader socialization (for instance, from school, media, social values) interacts with genetics.

It's certainly possible for environment to affect intelligence: in the absence of proper nutrition, which is an entirely environmental factor, children can experience various degrees of both physical and mental delays and eventually permanent damage. This is an easily observable and highly testable environmental interaction with genetics. It would be premature to assume there are no others simply because it is difficult/impossible to test for them.


*This is not my area of expertise, so I could be wrong. I'd love to hear about cross-cultural twin studies.
posted by carmen at 9:37 AM on September 5, 2005


There are plenty of statistical approaches scientists use to reliably disentangle the relative contributions of (a) genes, (b) shared environment (e.g., parental influences), and (c) non-shared (i.e., unique) environment to a particular trait. Typically these rely on twin or adoption samples. For example, the contribution of shared environment can be estimated by comparing the similarity between identical twins raised together to identical twins raised apart.
Sure, except that these "reliable" statistical methods systematically underestimate the influence of environment in favor of "genetic" factors - even, no, especially the twin studies so beloved of the "scientists" you name.
It turns out, that for most aspects of personality and cognition, the variance is attributable about 50-50 to genetic and non-shared environmental causes. Familial environment has little or no detectable effect on personality or intelligence, counter-intuitive as it may seem.
This is nonsense from start to finish. Do you realize that the concept of "heritability" as employed in population-genetics is extremely context-dependent?

The sad thing about your comments is that Charles Murray and his ilk display no more appreciation of such obvious statistical phenomena as I've pointed out here than you do, despite their claims to scientific respectability. People who confidently state that "heritability" is "50-50 genetic" without qualifying it to any particular group of individuals in any particular set of environments are totally clueless, and nothing else they have to say about population genetics is worth taking seriously.
posted by Goedel at 9:46 AM on September 5, 2005


heavy water, one problem with twin/adoption studies is that the children are raised in essentially the same broader social context. I've never heard of a twin study where one was raised in a really significantly different social context (like a Western verses a non-Western country)*. So, while twin studies may show that individual family dynamics don't have a huge impact on personality, they aren't able to address how broader socialization (for instance, from school, media, social values) interacts with genetics.

You're absolutely right; people usually don't realize that estimates of heredity are themselves culture-bound. So let me qualify what I said. In principle, it's possible for cultural differences at the broad societal level to have a meaningful impact on emotional and cognitive traits. But barring extreme deprivation, there's reason to think the effect isn't very large. While you're right that there are very few examples of genuinely cross-cultural behavioral genetic studies, studies within different cultures generally converge very strongly on the same estimates of heredity. E.g., it's the case that personality dimensions are equally heritable in virtually every sample they've been studied in.

With respect to intelligence, you're right that environmental factors such as nutrition could play a role. Similarly, we know that keeping children in the closet and never speaking to them (an environmental factor) is a pretty good way to induce retardation and prevent language acquisition. And certainly, if many people behaved in such a way, estimates of environmental influence would skyrocket. But in point of fact, most people don't behave in such a way, and we take it for granted that the interesting influences on cognitive ability aren't the extreme ones (because, as you point out, no one's surprised if starving a child will prevent normal development). No parent sits around wondering whether they should starve their children or not in order to ensure good development. Lots of parents wonder whether they should play their babies Bach every night. The former sort of environmental influence would have a major impact, but is a non-issue for the most part; conversely, the latter has no discernable impact but is something you hear a lot about in parenting books.
posted by heavy water at 10:00 AM on September 5, 2005


Sure, except that these "reliable" statistical methods systematically underestimate the influence of environment in favor of "genetic" factors - even, no, especially the twin studies so beloved of the "scientists" you name.

Actually, that's incorrect. Statistical methods can over- or underestimate any of the contributions. Where the bias falls is generally a function of experimental design, not of statistical procedure. It's incumbent on a researcher to use a design appropriate to his or her question. For example, if you were specifically probing the putative difference between white and black intelligence (and you seem to be assuming I come down on Murray's side--I said nothing to that effect anywhere), you wouldn't want to throw black twins and white twins together in the same sample. A more appropriate strategy would be to, say, compare black twin pairs in which one child was raised in a white household with black twins in which both were raised in black households, controlling for other relevant variables such as parental socio-economic status.

Aside from which, to the degree that there is a systematic bias in these studies, it tends to be an inflation of the unique environmental influence, because it includes measurement error (i.e., unreliable variance). If you take measurement error into account, the contribution of both genetics and shared environment increase somewhat.

This is nonsense from start to finish. Do you realize that the concept of "heritability" as employed in population-genetics is extremely context-dependent?

It's certainly context-dependent (see my previous post). But I think there's very little evidence to suggest it's 'extremely' context-dependent, unless you include highly atypical extremes. To the contrary, estimates of heritability of dimensions of personality such as extraversion are almost identical across just about every culture they've been investigated in.

People who confidently state that "heritability" is "50-50 genetic" without qualifying it to any particular group of individuals in any particular set of environments are totally clueless, and nothing else they have to say about population genetics is worth taking seriously.

Well look, there are tacit assumptions behind every statement. If you'd like me to qualify my statement by saying "personality dimensions such as extraversion display about 50% heredity and amost no shared environmental influence across virtually all human populations they've been investigated in to date," I have no qualms about that. On the other hand, I think you're going too far in the opposite extreme. The fact that estimates of heredity are context-dependent does not make them worthless. The fact of the matter is that there seems to be relatively little perturbation of the estimates as a function of culture. In fact, not only are the gross contributions of genetic/environmental roughly the same, but some groups of researchers have even replicated the same genetic structure of personality (i.e., the same patterns of covariance in traits) across multiple samples (see recent work by Jang and Livesley, for example).
posted by heavy water at 10:17 AM on September 5, 2005


heavy water: Familial environment has little or no detectable effect on personality or intelligence, counter-intuitive as it may seem

But how many of these studies provide sufficient variability in the first place, since they cannot be part of a planned experimental design? I don't think, for instance, that there are experiments where 30 kids would be systematically beaten and abused or at least subject to parental neglect in Group 1 while their 30 twin brothers/sisters would be raised in a more ideal fashion in Group 2, all other factors being equal. Or where Twins 1 would be brought up "white" in a "white" environment and Twins 2 would be brought up "black" in a "black" environment (whatever that means). We can do this with animals, not with people (fortunately).

I don't doubt that twin studies (great paper btw) can be extremely valid and useful for the purposes indicated in the paper, but the range of human experience is just not there. After all, adoptive families that agree to participate in such studies are by definition already self-selected... This seems to be a fundamental, unavoidable bias.

If there's little environmental variability in the first place, then all that's left is the genetic part, which may appear to be more important and therefore provide high heritabilities (a typical problem iirc).

(on preview, I read your answers after writing this)
posted by elgilito at 10:20 AM on September 5, 2005


But how many of these studies provide sufficient variability in the first place, since they cannot be part of a planned experimental design? I don't think, for instance, that there are experiments where 30 kids would be systematically beaten and abused or at least subject to parental neglect in Group 1 while their 30 twin brothers/sisters would be raised in a more ideal fashion in Group 2, all other factors being equal.

You're right, and that's certainly a concern worth keeping in mind. But it's also easy to underestimate the variability in these samples. Just to give you a sense of how broad they can be: several European countries maintain registries of all twins; researchers (who often offer pretty good compensation, so if anything poorer people are more likely to participate) then target broad swaths of the population. Many behavioral genetic studies (and certainly molecular genetic studies) have sample sizes in the thousands or even tens of thousands. And researchers take pains to ensure that reasonable variability exists in such dimensions as socioeconomic status, age, etc.

Also, a lot of work has been done in clinical populations who are being tracked for other reasons. For example, there are dozens of studies of personality in alcoholics and drug addicts, as well as in individuals with personality disorders. The popular thinking is often that such samples should show higher estimates of shared environment, since we tend to think of such diseases or disorders as stemming from traumatic childhood experience or early exposure to influences. But in fact, there turns out to be very little difference in estimates of personality in normal vs. personality-disordered or substance abuse populations. In fact, if anything, estimates of heritability are higher for most personality disorders than for continuous personality dimensions. So your point stands, but be careful not to underestimate the amount of variability in these samples, or the extent to which they generalize to abnormal populations.
posted by heavy water at 10:34 AM on September 5, 2005


No parent sits around wondering whether they should starve their children or not in order to ensure good development.

Of course not :)

But, given how many children in the world endure periodic starvation, the role of environmental factors like nutrition should not be overlooked in "intelligence" studies, especially considering that access to food is unevenly distributed in virtually all societies. Unequal access to the necessities of life that fall along the lines of socially defined "race" might create disabilities that follow the same lines, making it appear as though these disabilities were inherited traits.

This is the issue that Frans Boas (whom I mentioned in my first comment) addressed in the early 20th century. He showed that head shape (thought at the time to be related to intelligence) was linked to nutrition, and that in populations of immigrants from certain countries, head shape changed within the first generation of living in the US. He also showed that birth order affected height.

heavy water, you make several good points, but I feel like you are losing sight of what Murray and his ilk suggest. They are not suggesting simply that traits are inheritable, but that traits which affect ability (and potential) are inherited predominately within racial and gender boundaries. They suggest that by knowing the race/gender of a person you can know something about the intelligence/ability of that person, and they imply that inequality is a result of genetics. I don't think you were trying to support that position with twin studies, and I don't think that twin studies can actually support that position. However, I do think that twin studies in this context can be a little misleading because they do not address the constraints on genetic potential that social inequality can create.
posted by carmen at 11:10 AM on September 5, 2005


I just want to thank johngoren for that great (and apposite) Fitzgerald quote.
posted by languagehat at 11:11 AM on September 5, 2005


Actually, that's incorrect.
It's clear you didn't even bother to read the Ned Block essay, or if you did, you didn't understand it.
A more appropriate strategy would be to, say, compare black twin pairs in which one child was raised in a white household with black twins in which both were raised in black households, controlling for other relevant variables such as parental socioeconomic status.
Again, if you'd actually bothered to read the Block essay I linked to, you'd realize why this would still lead you to an inflated estimate of the degree to which IQ is determined by "race", but you clearly didn't bother. If all red-headed children had their brains bashed in from childhood on because of some social stigma, would that mean that the gene for red-headedness "caused" retardation? According to the textbook definition of heredity, yes!
It's certainly context-dependent (see my previous post). But I think there's very little evidence to suggest it's 'extremely' context-dependent, unless you include highly atypical extremes.
What you "think" and what is true are very different things; not only are there recent studies showing the heritability of IQ to be as low as 0.10 amongst poor Americans and as high as 0.72 amongst the rich - in perfect keeping with what we'd expect given the formula for heritability - but even a heritability figure of 1.0 for both groups would not imply that any difference in IQ which existed between them was genetic - read the second link I provided.
Well look, there are tacit assumptions behind every statement. If you'd like me to qualify my statement by saying "personality dimensions such as extraversion display about 50% heredity and almost no shared environmental influence across virtually all human populations they've been investigated in to date," I have no qualms about that.
1 - We aren't talking about extraversion but IQ.

2 - If you're going to make such a claim, you better have sources at hand to back it. "Extraversion" and "IQ" aren't some easily determined physical trait like height which can be relatively uncontroversially measured, so to claim that studies done "across virtually all human populations" unequivocally say something or other is almost certainly a blatant falsehood - unless the "human populations" you refer to all happen to be middle-class white Americans or something. Heck, even height has measurably varying heritability across populations, so I'm 100% certain you're talking nonsense.
On the other hand, I think you're going too far in the opposite extreme.
"Think" what you want: the fact is that your claims display nothing more than a facile understanding of the subject you're discussing.
The fact that estimates of heredity are context-dependent does not make them worthless.
Straw man: who ever said they were "worthless"? I specifically said that people who confidently claim that "heritability of X is Y" without any qualifiers are talking rubbish.
The fact of the matter is that there seems to be relatively little perturbation of the estimates as a function of culture.
A blatantly false claim. Try getting that one over on someone less aware of the literature or the underlying mathematics than I am. Give me any varying trait you want, and I can easily put together a population group with the right parameters to hit any number I please. In fact, Eric Turkheimer's study blows to pieces what you're trying to pass off here as knowledge.
In fact, not only are the gross contributions of genetic/environmental roughly the same, but some groups of researchers have even replicated the same genetic structure of personality (i.e., the same patterns of covariance in traits) across multiple samples
Utter and total bollocks, backed up by nothing more than vague allusions to "some researchers", all conveniently unnamed. Not only is it factually impossible for anyone to claim to be able to give a definitive answer to such a thing as a matter of sheer logistics - who has the resources to examine all the ethnic groups, castes, clans and income strata on this planet of ours? - but any quack whatsoever can pick scores of subject groups to study which only differ by dialect and geographic location, and then claim that his or her work can stand in for humanity as a whole.

Grandiose claims like yours just reinforce for me the impression that your knowledge of just what "heritability" is extends no further than the extremely superficial. Try actually reading the links I provided and then consulting an actual population genetics textbook sometime rather than pretending that I'm just making fussy distinctions: 2 + 2 does not make 5, however much prolixity you expend trying to make it seem so.
posted by Goedel at 11:38 AM on September 5, 2005


"In a large sample of mathematically gifted youths, for example, seven times as many males as females scored in the top percentile of the SAT mathematics test.6 We do not have good test data on the male-female ratio at the top one-hundredth or top one-thousandth of a percentile, where first-rate mathematicians are most likely to be found, but collateral evidence suggests that the male advantage there continues to increase, perhaps exponentially.7"

Having just attained a math degree from a University renowned for the rigor of its program, and having known many kids that will likely go on to be "first-rate mathematicians," I just have to say that this is bullshit.

Your SAT score is a measure of your ability to do a certain kind of math quickly, but the work that mathematicians do is rather different and people with imperfect scores often turn out to be great at it.

What's more, many graduate programs in math put a greater weight on your score in the verbal portion of the general GRE than on your score in the math portion, because they've found it a better correlate. Standardized tests should not be assumed to measuring anything other than one's ability to take them.
posted by mai at 12:13 PM on September 5, 2005


Wrong. As I noted above, socialization (at least in the sense of shared environmental influences such as parental environment) has almost no detectable effect on 'neurological development'. Again, that's counter-intuitive, but not controversial.

And you are utterly wrong about this. As a linguist, I can assure you the "uncontroversial position" among members of my profession is that the brain requires specific environmental conditions, triggers, and inputs to undergo its "normal" development in infancy and toddlerhood. And the fact that infants are *capable* of recognizing the the entire phonemic inventory of human Language at 1 does not obviate the point that they are predisposed to respond with preference to the distinctions they have heard while in utero. This is relatively old news. The brain is shaped by interactions with the environment. It is genetically predisposed to do so.

Twin studies are the very thin soup from which many generalizations about the distinctive (or intertwined) roles of genes and environment in development are drawn. Very thin. Twins raised apart tend to be raised in relatively similar socio-economic and linguistic settings. And very few twins are raised separately from birth, certainly not enough to provide a basis for generalizing about such a broad topic.

Take a baby born in Cambodia, and raise her in Paris. She will grow up French for all intents and purposes, and vice versa. But take a baby born anywhere and deprive her of normal social interactions, some sort of linguistic input, etc., and she will grow up neurologically damaged. I am not at all saying genetic factors don't determine development. I'm saying, short of a very unethical experiment (see Russ Rymer's book on the "Genie" case from the early 70s for example), there is no way to isolate genetic and social factors in normal human development well enough to approach the generality of Murray's claims, which further rely on a reduction of genetic influence to phenotypic variation.

If you want to get down into the science, let's start a new thread. I don't want to be constrained by the possibility od dignifying Murray's nonsense with serious discussion of real research.
posted by realcountrymusic at 12:33 PM on September 5, 2005


And what Goedel and Carmen said.
posted by realcountrymusic at 12:35 PM on September 5, 2005


I thought The Bell Curve was a ridiculous book, because not only did it not prove its point, but it couldn't have proven its point even if its conclusion was correct. There's simply no way to have a suitable control for such studies.
posted by spira at 1:24 PM on September 5, 2005


heavy water, you make several good points, but I feel like you are losing sight of what Murray and his ilk suggest. They are not suggesting simply that traits are inheritable, but that traits which affect ability (and potential) are inherited predominately within racial and gender boundaries.

I wasn't addressing Murray's claims, nor do I have strong feelings about them one way or the other. I was responding to methodological concerns about the use of twin designs.
posted by heavy water at 1:56 PM on September 5, 2005


Again, if you'd actually bothered to read the Block essay I linked to, you'd realize why this would still lead you to an inflated estimate of the degree to which IQ is determined by "race", but you clearly didn't bother. If all red-headed children had their brains bashed in from childhood on because of some social stigma, would that mean that the gene for red-headedness "caused" retardation? According to the textbook definition of heredity, yes!

You're right that the design I proposed wouldn't get at this particular effect, but there are ways to get at this. For instance, you could conduct studies of heredity in genetically-similar populations (e.g., African populations in the case of black americans), where whites are a tiny minority and presumably discrimination is not an issue. Alternatively, you could quantify discrimination (e.g., by sampling from racially segregated vs. racially de-segregated areas). Is there a perfect solution? No, of course not. Nor was I suggesting there was.

Note that my original response to your point still stands: twin studies do not systematically underestimate environmental contribution as opposed to the genetic contribution. They could just as easily underestimate the genetic contribution, because no sample is perfectly genetically heterogeneous. If you were to throw people of more diverse genetic backgrounds living in the same culture, you'd raise your genetic estimates. But there are no perfect samples, much as there are no perfectly balanced environments.


What you "think" and what is true are very different things; not only are there recent studies showing the heritability of IQ to be as low as 0.10 amongst poor Americans and as high as 0.72 amongst the rich - in perfect keeping with what we'd expect given the formula for heritability - but even a heritability figure of 1.0 for both groups would not imply that any difference in IQ which existed between them was genetic - read the second link I provided.

I'm familiar with the Turkheimer study, and it's perfectly consistent with plenty of other work (see the Boomsma review I linked to above) indicating that the shared environmental contribution to IQ is very large in early childhood and decreases to almost nothing by age 18. (Notice I said 'after childhood' in my first post for a reason.) Essentially, programs like headstart work for a little while, then their effect dwindles away as other kids catch up. The trend towards decreasing shared environmental contribution as a function of age is fairly general and you find it for a range of traits (did you read the Boomsma article I linked to? Apparently not. Then we're even). If you can point me to a study showing the same effect in an adult population, I'd buy your argument. But I follow this literature and have not seen anything of the sort.

1 - We aren't talking about extraversion but IQ.

Well, you might be talking about IQ. I was responding to methodological concerns. Notice I haven't said anything to support or detract from Murray's position at any point.

2 - If you're going to make such a claim, you better have sources at hand to back it. "Extraversion" and "IQ" aren't some easily determined physical trait like height which can be relatively uncontroversially measured.

They certainly can be uncontroversially measured. There are a number of highly reliable measures of extraversion which all intercorrelate substantially and have good external validity. Similarly, virtually all putative intelligence tests load highly on a single factor that's indexed closely by Ravens Matrices or the WAIS, and these scores predict a variety of real-world outcomes better than any other single psychometric factor.

So to claim that studies done "across virtually all human populations" unequivocally say something or other is almost certainly a blatant falsehood - unless the "human populations" you refer to all happen to be middle-class white Americans or something.

You're mangling (deliberately?) what I said. I'm not claiming that studies have been conducted in virtually all populations, I'm saying that in virtually all populations in which studies have been conducted, you get the same picture. Do a search on Google Scholar or Pubmed for 'extraversion heredity' and you'll find dozens of samples drawn not only from just about every western culture there is but also a number of Asian and Native American samples. The coverage of studies isn't exhaustive, of course. But I know of no large studies that have failed to find a large component of heredity (averaging around 50%) in any culture. Is it possible that you might find cultures in which there's little or no heritable contribution to extraversion? Sure. But I think it's exceedingly unlikely given existing evidence, and isn't a reasonable position at this point.

Straw man: who ever said they were "worthless"? I specifically said that people who confidently claim that "heritability of X is Y" without any qualifiers are talking rubbish.

I followed up my initial post with two qualifiers to the same effect.

A blatantly false claim. Try getting that one over on someone less aware of the literature or the underlying mathematics than I am. Give me any varying trait you want, and I can easily put together a population group with the right parameters to hit any number I please. In fact, Eric Turkheimer's study blows to pieces what you're trying to pass off here as knowledge.

As I pointed out, Turkheimer's study is perfectly consistent with everything I've said (I even went out of my way to acknowledge that the genetic contribution is only apparent after childhood, precisely because I've read this paper and others). And I'm not sure if you've ever conducted an empirical study involving human subjects, but one does not go about tweaking the correlations between one's samples until one is satisfied prior to conducting analysis. To suggest that the high estimates of heritability some researchers get are due to data fudging is ridiculous and, frankly, an affront to scientists who collect empirical data (including myself). If you're going to go down this road, you could just as easily include every other study every conducted and conclude no one can conclusively determine anything of interest because everyone could be cheating.

Utter and total bollocks, backed up by nothing more than vague allusions to "some researchers", all conveniently unnamed.

This comment seems rather disingenuous considering you omit the very next parentheses in which I refer you to recent papers from the Jang and Livesley group supporting my point.

Not only is it factually impossible for anyone to claim to be able to give a definitive answer to such a thing as a matter of sheer logistics - who has the resources to examine all the ethnic groups, castes, clans and income strata on this planet of ours? - but any quack whatsoever can pick scores of subject groups to study which only differ by dialect and geographic location, and then claim that his or her work can stand in for humanity as a whole.

But no one that I know of is making this claim. I'll be the first to say (and have said above, repeatedly), that there is no way to get an 'objective' estimate of these contributions taking all factors into account. What I have said is that across a wide range of cultures, you get the same estimates of heredity and shared environment for virtually all dimensions of personality that have been measured. And I would like to think that Native Americans, European Americans, Finns, and Japanese differ in more than just dialect and geographic location.

Grandiose claims like yours just reinforce for me the impression that your knowledge of just what "heritability" is extends no further than the extremely superficial. Try actually reading the links I provided and then consulting an actual population genetics textbook sometime rather than pretending that I'm just making fussy distinctions: 2 + 2 does not make 5, however much prolixity you expend trying to make it seem so.

Well, we could try to have a constructive discussion, or we could name-call. My background in genetics is limited to a couple of courses and extensive reading of the behavioral genetics literature, and my statistical background extends to several graduate level statistics classes. If you'd like, we can discuss the mathematical details of these procedures, and I'll do my best to follow. Otherwise, I don't see how that's relevant here. Do you hold a position in population genetics somewhere? And if so, am I supposed to simply defer to your wondrous knowledge of the field, in the absence of argument for your position?
posted by heavy water at 4:17 PM on September 5, 2005


And you are utterly wrong about this. As a linguist, I can assure you the "uncontroversial position" among members of my profession is that the brain requires specific environmental conditions, triggers, and inputs to undergo its "normal" development in infancy and toddlerhood. And the fact that infants are *capable* of recognizing the the entire phonemic inventory of human Language at 1 does not obviate the point that they are predisposed to respond with preference to the distinctions they have heard while in utero. This is relatively old news. The brain is shaped by interactions with the environment. It is genetically predisposed to do so.

I agree with all of this (and alluded to cases of extreme language deprivation earlier), so I suspect our definitions of 'environmental' carry different emphases. There's no question that some input from the environment is necessary in order to ensure 'normal' development of each and every trait. The question is whether those extreme conditions are worth worrying about. I'm familiar with the long history of the nativist-empiricist debate in linguistics, so I apologize for being flippant; I realize it's an important observation from a linguistic perspective that exposure to language during the critical period is a necessary component of language development.

On the other hand, from a behavioral genetic standpoint, I think the kind of influences you point to--while equally uncontroversial--are not that interesting, because they're virtually universal. In the context of twin studies, environment is taken to connote relative environment, in contrast to your focus on absolute environment. But I don't think we're in substantive disagreement here.

I am not at all saying genetic factors don't determine development. I'm saying, short of a very unethical experiment (see Russ Rymer's book on the "Genie" case from the early 70s for example), there is no way to isolate genetic and social factors in normal human development well enough to approach the generality of Murray's claims, which further rely on a reduction of genetic influence to phenotypic variation.

Here I think our disagreement may be real. Personally I do think we can isolate these factors with enough generality--if by enough generality you mean 'to a degree that has important social and political implications'. The fact that personality and IQ aren't susceptible to shared environmental influences--even within a single culture!--is hugely counterintuitive and has all sorts of practical implications. For example, if you know your behavior as a parent isn't going to have much of an effect on your children's eventual personality, I think to some degree it behooves you to worry more about making them happy while they're living with you and maybe focus less on traditional notions of 'molding' children. So these kinds of findings are important whether or not they're universal to all human cultures (though at least for personality, that seems likely).

As far as Murray's claims go, I don't have strong feelings one way or the other about whether he's right or not. But I do think most people who reject his position out of hand do so for emotional and not rational reasons. It's also worth keeping in mind that, because of the sensitive topic he's chosen, his arguments have received much greater scrutiny than almost any other scientific hypotheses are subjected to. The social sciences are, unfortunately, imprecise; I don't know of many (or even any) models of human nature or behavior in psychology that wouldn't show cracks under the same level of examination. Given the political implications, the double-standard may be justified, of course; but it's worth keeping in mind that it's there.
posted by heavy water at 4:36 PM on September 5, 2005


heavy water, I reserve the right to dismiss anyone who uses Victorian pornography as support for a claim about genetics out of hand. Seriously. Once you cite Rushton, the jig is up.
posted by carmen at 5:40 PM on September 5, 2005


heavy water, I reserve the right to dismiss anyone who uses Victorian pornography as support for a claim about genetics out of hand. Seriously. Once you cite Rushton, the jig is up.

Well, you don't have to agree with many of Rushton's conclusions (and I don't either). But I think it's important to separate the man from his science. Rushton has published quite a few articles in good peer-reviewed journals. Presumably the editors of those journals and the reviewers they solicited are in a better position to judge the merit of his work than a second-hand report from your advisor. Also, it's virtually impossible for a serious academic to avoid occasionally citing questionable research; when you're reading hundreds of papers and distilling them, it's very difficult to be meticulous about everything. A paper shouldn't be made or broken on the basis of one data point (unless it's critical to the argument, obviously).

It's interesting you mentioned Gould in this context above; one could argue that Gould was a master of the careful omission of contradictory data, or even of deliberately misconstruing others' work. Actually, Rushton himself published a harsh peer-reviewed critique of The Mismeasure of Man (ignore the main website; unfortunately it's the only place that reproduces the copyright-protected material on the web). I haven't read the book myself, so I'm not in a position to pass judgment one way or the other. But I do think the fact that Murray cites Rushton is any grounds to dismiss Murray out of hand; Rushton may be wrong and even a racist, but he's a very smart man and certainly not the crackpot you make him out to be.


posted by heavy water at 7:44 PM on September 5, 2005


The pursuit of this information is itself suspect, as nothing civilized can be gained from it.
Perhaps there is a subset of the human species that is, for example, intellectually inferior. By labelling the entire group you do a disservice to individuals in that group that demonstrate exceptionally high intelligence. One does not deny the possibility of the science being sound, but insists that the application of it is unacceptable.
Following that philosophy, we now find ourselves in a world where women can vote, black people are free, and jews can own property. Discard that philosophy and see what happens.
posted by CynicalKnight at 8:02 PM on September 5, 2005


"...but that doesn't mean we can't ever say anything useful about race and genetics"

Yes it does, because race and genetics have almost nothing to do with another. You cannot define race genetically. You cannot find a genetic variance which correlates to a specified "race" across the entire human population. There is no common usage today I'm aware of, at least in the US, where "race" wouldn't define all very dark skinned peoples as "black" and a "race". But there is not a genetic correlation to this designation. There's not. It's been looked for.

I, myself, dislike the backlash against sociobiology and evolutionary psychology and regret that there is almost a taboo against considering the possibility of a genetic basis of certain kinds of behaviors. But the reason that backlash exists is because of people like Murray and his ilk who use arguments from nature to support a prejudice. Indeed, the very notion of "race" as it exists in modern, western thought is the product of the discovery of genetics and evolution and the subsequent attempt to "rationalize" and justify bigotry. This has a long and ignoble tradition and Murray is only one of the more recent practitioners.

The Bell Curve is crank science and it's been soundly proven to be crank science.
posted by Ethereal Bligh at 9:22 PM on September 5, 2005


"The pursuit of this information is itself suspect, as nothing civilized can be gained from it."

And, by the way, I disagree with this statement.
posted by Ethereal Bligh at 9:23 PM on September 5, 2005


I, myself, dislike the backlash against sociobiology and evolutionary psychology and regret that there is almost a taboo against considering the possibility of a genetic basis of certain kinds of behaviors. But the reason that backlash exists is because of people like Murray and his ilk who use arguments from nature to support a prejudice.

Incredibly well put, Ethereal. As a linguist, I too cannot dismiss the genetic basis for the most culturally significant behavior that characterizes our species without respect to clades of phenotypic adaptations. I studied as an undergraduate with Gould, whom I revered. He made that point over an over again: tolerating racist pseudoscience means more than keeping an open mind. It damages our pursuit of the most interesting of all questions (to generations of humans), one that unifies the natural, social, and humanistic disciplines. Evolutionary Psychology has come a long way in reforming sociobiology, even if there's still a long way to go. And the excesses of the behaviorist and cultural relativist camps are apparent (though of couse these can't be reduced to one another). The real mystery is so much more profound, and its social implications much more radical when understood, as Chomsky has argued for decades.

I am also, by graduate training, a member of the Boas lineage, and find my first principles in *Race, Language, and Culture.* That doesn't mean I wasn't convinced by Chomsky's famous damning review of Skinner's *Verbal Behavior,* or Derek Bickerton's brilliant thinking about language and evolution, or Anne Senghas' discovery of Nicaraguan Sign Language at its moment of birth. It's trite but always necessary to say that the interplay of nature and nurture is where the most powerful explanations lie for almost anything we experience as significant in our encounters with our fellow humans: intelligence, beauty, aggression, altruism, etc. It strikes me that people like Murray aren't interested in explaining mysteries, just confirming biases. And hatin' on folks. Fuck them.
posted by realcountrymusic at 10:48 PM on September 5, 2005


It's interesting that you'd defend EP at all given your association with Gould. I side with the adaptationists myself. Ah, but then you're a linguist, which brings us to Pinker.

I must confess I've not read Pinker, not even The Language Instinct (which I very much think I ought), but like many have absorbed some familiarity with his arguments in The Blank Slate. My own impression from my experience defending his side of the argument from the relativists is that perhaps he's done more harm than good. Although I side against Gould, his essential point that there is a tendency to "just so" stories is correct.

But, anyway. Sigh. The Science Wars. They make me tired.
posted by Ethereal Bligh at 1:52 AM on September 6, 2005


Well, you don't have to agree with many of Rushton's conclusions (and I don't either). But I think it's important to separate the man from his science.

I am. I'm not saying Rushton is a bad person. I'm saying that he did bad science. I gave two peer-reviewed sources. And, on a personal note, I have never met a social scientist or statistician who takes Rushton's work seriously.

Presumably the editors of those journals and the reviewers they solicited are in a better position to judge the merit of his work than a second-hand report from your advisor.

This is kind of a wierd statement. I cited a published source for the comment, because mine was a pretty strong claim for anyone to just take on faith from me, and I'm not sure if it is general knowledge.

Also, it's virtually impossible for a serious academic to avoid occasionally citing questionable research; when you're reading hundreds of papers and distilling them, it's very difficult to be meticulous about everything. A paper shouldn't be made or broken on the basis of one data point (unless it's critical to the argument, obviously).

Two things:
1) it wasn't "research." It was a fictional book about a surgeon who traveled the world and told tales of strange and exotic sexual exploits.

2) There has to be a standard that we require of people who are making scientific claims. While it is true that it is difficult to be sure of the quality of every source, there needs to be a basic standard. Requiring that all scientific sources be works of fact and not fiction is a pretty minimal standard. If you can not distinguish fictional from non-fictional accounts, then you're ability to conduct good research is suspect. If you cite Pipi Longstocking as a bibliographical account of childhood in early 20th century Sweden, (particularly if you go on to make claims about how parental abandonment and association with pirates leads to antisocial behaviours) I'm going to dismiss you out of hand. Even if you only do it once.
posted by carmen at 6:25 AM on September 6, 2005


realcountrymusic, I like you more and more. Sorry I bit your ankles in some thread I no longer remember. We linguists have to stick together.
posted by languagehat at 8:10 AM on September 6, 2005


right on, languagehat. all is forgotten and forgiven here -- i don't recall the argument either anyway.

ethereal -- i am not "defending" EP. i think Pinker is a brilliant ass. and his broadsides against linguistic anthropology are ill-informed, though the rest of Language Instinct is very useful for beginners and i've taught it a few times profitably. i simply think EP has done more than put makeup on the pig of sociobiology, by and large. i still think it's 90 percent full of shit. jerry fodor had a brilliant takedown of EP in TLS a couple of weeks back, but i can't find the reference online at the moment. but stephen pinker is to charles murray as pedro martinez is to john rocker. no comparison. (except that murray and rocker are both racist pigs).
posted by realcountrymusic at 12:26 PM on September 6, 2005


Fodor's TLS essay
He argues that behavior cannot be explained without motivation and insists that "ultimate cause" must resemble our own motivations. His two ostensibly opposite motives for copulation seem inextricably linked, and to separate them seems a bit disengenuous.
posted by Treeline at 7:42 PM on September 6, 2005


Treeline, awesome essay. Thanks for pointing it out.
posted by carmen at 8:12 PM on September 6, 2005


(Well, I guess you can, as you did. But really, you oughtn't. Who's the prof?)
posted by IndigoJones at 6:26 AM PST on September 5 [!]


I don't know, I'm not in that department. My roommate just really respects her (and I don't want to say her name in case I get it totally wrong, because I don't know if I remember right. Also, it would be gossiping). I said she was an original thinker in the department, but I don't know how she's seen world wide, though from what my roommate said, she had some papers that were pretty seminal for some subject - only she changed her name at marriage and thus most people don't know all that she's written. (It was in the context of a discussion on women in academia.)

My point is that not only have there been some very fine female theological thinkers in the past (Teresa D'Avila, Catherine of Siena), but there are many talented female philosophers today. My roommate wipes the floor with my, analytical thinking wise. Don't argue with a philosopher is my new motto.
posted by jb at 8:13 PM on September 6, 2005


« Older A little local color...  |  "It's as if the entire Gulf Co... Newer »


This thread has been archived and is closed to new comments