Talk:Race and intelligence/Archive 40

From Wikipedia, the free encyclopedia

Archive This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Contents

Keith Booker

here's the text from the news piece being quoted:

While acknowledging a need to respect the two professors' academic freedom, Keith Booker, president of the Wilmington, Del., chapter of the National Association for the Advancement of Colored People, says that "this research is being done in the name of white supremacy." He says the Pioneer Fund supports only research that "tends to come out with results that further the division between races...by justifying the superiority of one race and the inferiority of another."

The problem with this is that quoting a leader of a political organization with vested interests in this issue on a matter of research is a bad idea. --W. D. Hamilton 19:20, 2 November 2006 (UTC)

Seconded. It's preferable to the previous version though, which came across as the statement being a general opinion. --Zero g 19:32, 2 November 2006 (UTC)
There is also the part that accuses the PF of funding most British and American race researchers, which has been leveled by the SPLC, not by Booker. And the problem with wanting to censor the Booker (or the SPLC for that matter) quote is that even though they may have vested interests, the researchers themselves (Rushton foremost, but also Lynn and several others) have all been accused of bias. Sure, we can remove the two accusations - if we also remove all information related to Lynn's and Rushton's findings and theories, for the same reason. But the reasonable thing to do here, I think is to censor nobody, but to disclose all pertinent info that may help the reader to assess whether these people have any kind of bias, and if so in which direction. It seems only fair that way. The research has been decried far and large as also furthering a political agenda, or at least a specific worldview. I don't think you can cleanly separate the science from the politics in this matter.--Ramdrake 19:40, 2 November 2006 (UTC)
Going a bit over the top there, no? I don't think anyone has removed anything here. The implied suggestion is that a substitution should be made. It is an open question yet as to whether Bookers comments aren't amongst the most strongly put comments you will find about the PF, clearly stronger than what you get from Tucker or other academics. The goal is to describe the most common views and indicate their commonality where possible. This can be done best when someone spells out what the most common views are. If that's not available, we're in a hard situation to give a balanced presentation. Presenting the strongest possible criticisms -- e.g., white supremacy -- is probably a bad idea in this case, unless it can be seen coming from someone other than a political figure. --W. D. Hamilton 19:49, 2 November 2006 (UTC)
I sincerely apologize if you thought I was seriously suggesting we censor that much of the article's data. And about the comment regarding white supremacists, I'm not the one who added it originally and I wouldn't be sorry to see it go. However, the ones about the PF funding most American & British race researchers, and about the PF researchers results always coming up on the same side are to me much more important and should stay. We also need to make it clear who makes those comments (I think it's already clear enough, but a second opinion wouldn't hurt).--Ramdrake 19:59, 2 November 2006 (UTC)


I'm not entirely sure how the SPLC is making an accusation. To me it sounds more like an observation, unless 'funding racial research' is something that is considered immoral and automagically implicates racist motives.
While I think it's definitely worth mentioning in the PF section I'm not sure of it's relevance in the bias section. It's a bit of a wasp nest subject because most of the criticism seems to be based on moral grounds, proof that the PF tries to manipulate the researchers, or stops funding of researchers who deliver results that aren't to their liking would be preferable.
Regarding the white supremacy part, I think selective quoting is a bad idea. --Zero g 20:03, 2 November 2006 (UTC)
Whether you call it an observation or an accusation, it implies that somehow the PF skews the research field in some way. Personnally, would call it a bias, so it can go either in the bias or the PF section. Not sure why we have two distinct sectons at that point, as the accusations against the PF basically amount to accusations of biasing the research field.
Regarding the white supremacy part, while I don't lke the idea of censoring objections, I must admit it's the least specific of all the accusations (preferential funding and result skewing are, IMHO much more specific). I can live with or without the criticism, but since it's in there with a pertinent source, I'd like a more persuasive reason to remove it than "it comes from someone who's biased". I believe we all have our biases, in any case.--Ramdrake 20:32, 2 November 2006 (UTC)
"Skews the research field" is ambiguous. NIH skews the biomedical research field by what they fund -- not ESC research. It doesn't follow necessarily that they skew what gets determined to be "truth" in a way that "untruth" prevails. The problem with a politician isn't bias but accountability. Criticisms published in the scholarly literature can be responded to effectively and debate, but not those in news reports. Booker is unaccountable for what he says when it's published in the news. --W. D. Hamilton 20:49, 2 November 2006 (UTC)
The reason for mentioning truth/untruth is that's at the heart of the "bias" idea. "Bias" keeps you from getting at the truth. --W. D. Hamilton 22:58, 2 November 2006 (UTC)
Last time I checked, verifiabiility was a guideline for Wikipedia, not Truth, not Accountability. If you want a scholarly criticism of a PF fundee like Rushton, just take a look at Lieberman, 2001 (How Caucasoids got such big crania and why they shrunk).--Ramdrake 21:07, 2 November 2006 (UTC)
Verifiability is a necessary but not sufficient condition for inclusion. Neutrality is the problem here. I suggested Tucker was one of the best examples of a critic of PF, and that his views along with other scholars should be given precedence over the opinions of political figures who are by virtue of their position and venue are less likely to be presenting a balanced view. Neutrality isn't easy. The question we need to consider is what material is important for this article -- as compared to the PF article itself -- and in what form should it be presented. --W. D. Hamilton 21:50, 2 November 2006 (UTC)
NPOV requires neutrality in presentation. It was never meant to say we should remove a cited, pertinent view if we felt it was non-neutral. If such was the case, we would never be able to describe anything which is controversial in the least. I would suggest if you look for a good scientific critic of the PF work, that Lieberman is also a good candidate, and I'm sure the next editor also has an opinion. My point is, you can't decide to include or exclude a viewpoint on the basis of what you alone feel is appropriate. Only an editorial consensus can do that. I submit that this view is important because of the importance of PF funding in this specific type of research. If you include PF-sponsored research in there, you need to be prepared to include whatever criticism has been leveled at it and is cited. --Ramdrake 22:10, 2 November 2006 (UTC)
So you're committing to the opinion that it's a good idea to cite Keith Booker as an example of a notable opinion about the PF that is pertinent to understanding R&I? Is there any evidence that Keith Booker's view -- that the aim of R&I researchers supported by the PF is to "further the division between races" -- is shared by opinion leaders in this subject. This doesn't appear to be the view of Tucker. Moreover, we have to doubt the care with which Booker came to his conclusions. How much of this stems from commiting the error pointed out by Coren 1993:
The conclusions that can be reached based on these data should be obvious--and somewhat frightening. It is quite clear that many students, especially the freshmen tested here, cannot separate the scientific evidence presented by an instructor from the instructors own opinions. Also, they make one variety of the "fundamental attribution" error so well-known to social psychologists./4/ In this case, the error involves the belief that the conclusions reached by the lecturer are the conclusions desired by the lecturer. In other words, the observer (here the student) believes that the lecturer must be driven by internal motives consonant with the data he presents. [1]
Should we believe that Booker has resisted this cognitive bias in making his conclusions? Should we assume that his comments are well thought thru? --W. D. Hamilton 22:30, 2 November 2006 (UTC)

Which is not to say that I'm advocating the censorship of Booker. Just that I'm uneasy with the flat presentation of his opinion. It makes it seem as if he is representative of a major POV. Surely Booker is representative of the range of opinions, but a scholar would be better suited to express a major POV. --W. D. Hamilton 22:58, 2 November 2006 (UTC)

Given the ongoing controversy I'd say Booker, to a certain degree, expresses a major POV of the general public. It's not unlikely that the intentions of PF are exactly what he claims them to be. Intention and morality aren't relevant in science however, so regarding R&I being biased we'd have to establish that 1) PF only selects researchers that they believe will produce results they desire. 2) PF is the main sponsor of R&I research. 3) See if researchers with clear egalitarian believes have major benefactors. 4) If either of these research groups is biased.
It would seem to me that it's obvious on the face of it that this is an extreme opinion considering that research which essentially invariably ranks Askhenazi Jews and East Asians above Whites in IQ is not exactly white supremacist in nature and not what white supremacists would want to hear. Usually consensus opinions are not that manifestly irrational, but maybe with this issue they are... --70.91.235.10 02:59, 6 December 2006 (UTC)
One weak part of the article is that the culture only researchers seem to be above criticism, which in itself harms NPOV. I guess I should add that I've come across a couple of eyebrow raising actions from culture only researchers in the past. --Zero g 23:29, 2 November 2006 (UTC)


Zero g, for the points you raise, I can find you quotes for points 1 and 2 if you don't already have them (there are some in the article). If you can find a solid, scientific refutation of the culture researchers, please share with us. However, what concerns me the most (and I know I share this concern at least with JK) is that the debate is presented as culture-only vs partly-genetic causes debate, whereas if you listen to the theories of many "partly-genetic" researchers (take Rushton and Lynn, for example), you can't help but get the feeling that what they really mean is "mostly-to-totally genetic". Conversely, many researchers who oppose the kind of conclusions of a Rushton or a Lynn don't necessarily claim zero genetic involvement; many of them claim that the genetic involvement in the issue accounts for only a minor part of the observed B-W IQ discrepancy. In this respect, it's almost as if the debate is misrepresented to start with. That troubles me.--Ramdrake 23:48, 2 November 2006 (UTC)
It would be nice if that's the only thing JK meant. Jensen has suggested 50% genetic prior to 2005 (7.5 points of the BW gap), and Rushton/Jensen suggested up to 80% genetic in 2005 (12 points). They specifically say 100% genetic can be ruled out based on the available data. You don't have to look for the find quotes claiming that any genetic cause can be ruled out, but Reynolds (2000) suggests this really means a genetic contribution of <20% (<3 IQ points). --W. D. Hamilton 00:10, 3 November 2006 (UTC)
If we can put % numbers on the argument, we should. 80% genetic vs. <20% genetic are more informative than "partially genetic" vs. "culture only". --JereKrischel 00:21, 3 November 2006 (UTC)
What I wrote in my comment is contained within the explanations section of this article. Perhaps they are too spread out right now, but feel free to make that clearer. BTW, I don't know how seriously they take the 80% number. It was certainly meant as speculation. The context was that 50% didn't quite fit one of David Rowe's models, but was close. So they suggested that the heritabilty of g among adults (~80%) might fit. --00:41, 3 November 2006 (UTC)
Regarding the refutation of the culture researchers, I think I only ran into a couple from some far right sources which has the same problem as the Booker quote, though they weren't expressed as bluntly. One of the criticisms was that though the follow up of the minnesota transracial adoption study was in 1986 it wasn't till 1992 that Weinberg et al. published their findings. It's clearly a sign of bias when this happens, especially when the results are interpreted in a clearly controversial manner, or when studies are published silently with minimal media attention. --Zero g 11:09, 3 November 2006 (UTC)
In academia, a long time fom the conclusion of a study to its publication may not imply any bias: it may have to do with controversiality, with poor writing of the paper, with poor methodology, well you get the picture... And I must say that most studies are published with a minimum of media attention, otherwise it'd fill up all the newspapers. The amount of attention it gets may depend on the width of its appeal, on its significance in science, on its controversiality too. I'd just be a bit cautious of launching accusations of bias based on publication delays and lack of media attention alone (that being said, you might be totally right). However, if someone else made them, and its reliable and verifiable, you can certainly quote it.--Ramdrake 22:26, 3 November 2006 (UTC)
Guess I'll try finding that article sometime, for NPOV's sake. It's also worth checking if controversial researchers can find funding outside the PF, if that's not the case it might be alarming as well. --Zero g 23:17, 3 November 2006 (UTC)
We've already held this discussion, which is by now somewhere in the archives of Race and Intelligence, and we came to the conclusion that at the very least the vast majority of R&I researchers with an "hereditarian" stance (what you called the "controversial" ones, mainly) were PF fundees. We were hard pressed to find more than maybe one or two who weren't. That's where the bit about the Pioneer Fund being accused of biasing the field came from. Please feel free to reexamine the issue if you so feel.--Ramdrake 23:25, 3 November 2006 (UTC)

65.13.98.41

is 65.13.98.41 vandalizing? it's uncited material. --W. D. Hamilton 23:13, 2 November 2006 (UTC)

I agree, it should be cited, or removed. --JereKrischel 00:18, 3 November 2006 (UTC)

Vandalizing deals with the main article; this is a discussion area. I have seen some very good points erased from this discussion (better points then the main article). A moderator should look into whos editing this page because they have bias which obviously doesnt belong in science. Jon Dec. 18 2006

Pioner Fund

How about we just delete the Pioneer Fund subsection entirely, merge all its content into the Pioneer Fund article, and write a new concluding paragraph. Maybe start with:

On the other side, the Pioneer Fund has often been criticized for bias while funding "most American and British race scientists" [2]. Prominent race science critic Ulric Neisser states that the fund's contribution has overall been "a weak plus".

Main article: Pioneer Fund

Simple fact is, Pioneer Fund may or may not be biased now, not really sure, but any power they might have can only be to governments not funding contraversial research which might get diffrent (more objective) results. And that story fits much better with Pinkers quote about stuff being surpressed. JeffBurdges 20:54, 9 November 2006 (UTC)

There has been strong consensus to put this section in this article as a strong indication of the problems riddling race and intelligence studies. I don't think the section should be removed; I'd even dare say it's a very bad idea.--Ramdrake 22:39, 9 November 2006 (UTC)
I'm sertonly not suggesting ignoring the Pioneer Fund! Just eliminating the subsection break and leaving more to that article's introduction, which is much better than this subection. I was trying to ask what else should be included here. Linda Gottfredson, SPLC, NAACP, and Neisser all seem directly relevant. But most of the middle paragraph can be folded into the Pioneer Fund's main article. So maybe it can be reduced to 2/3rds or 3/4ths its current length. Well maybe I'll try to rewrite it sometime. JeffBurdges 07:10, 3 December 2006 (UTC)

Intelligent people in history with small brains?

So if one has a small brain he is incapable of becoming a physicist or mechanical engineer? Are there any documented accounts of inventors or scientists that may have had smaller than average brains? There are a few African Americans who were philosophers I remember, but they may have had some genetic material from Europeans that increased brain size. If their IQ were measured today, what would their score be? --Apathy 14:50, 16 November 2006 (UTC)

I believe Einstein had a below average sized brain. The lower IQ measured is on average however, so people with large brains can have low IQs and visa versa since brain size isn't the only factor that influences intelligence. --Zero g 20:43, 14 November 2006 (UTC)
Einstein did have a below average sized brain, but apparently, his brain was much more neuron-dense than the average brain. —The preceding unsigned comment was added by 129.97.242.58 (talk) 03:18, 10 December 2006 (UTC).
Is it impossible for someone with a smaller brain to truly excel in engineering ? Apathy 09:11, 12 December 2006 (UTC)

boas vs sparks and jantz

http://www.pnas.org/cgi/content/abstract/99/23/14636

which of boas' findings does this 2002 paper refute? it seems to be saying that boas was wrong about the environmental plasticity of cranial features. it mentions "cranial vault" size. after a careful reading, the statements in this article about boas' findings need to be reassessed. --W. D. Hamilton 06:32, 18 November 2006 (UTC)

Hey Rikurzhen, where exactly in the article is this cited? And do you have the full text of the paper available? --JereKrischel 07:40, 18 November 2006 (UTC)
PMC: http://www.pubmedcentral.nih.gov/articlerender.fcgi?tool=pubmed&pubmedid=12374854 --W. D. Hamilton 10:29, 18 November 2006 (UTC)
Thank you. It seems what they've done is carefully craft their analysis to build a model that fits their conclusions, and contradicts Boas. Apparently, there has been similar work supporting Boas ("Coincidentally, as of this writing, another paper, by anthropologists at the University of Michigan, the University of Florida, and Northwestern University, is scheduled for publication in American Anthropologist; it concludes that Boas correctly interpreted his head-form data." [3] I assume the article is this one: http://www.aaanet.org/aa/105-1_gravleeetal.pdf.
Sounds like an interesting direction, but nothing particularly new to see yet - they'll have to probably construct modern studies to prove or disprove Boas' theories, rather than do spreadsheet analyses. It would have been very helpful of them to provide the data tables and calculations they chose - without that, it is difficult to accept uncritically their findings (as they ask us not to accept uncritically Boas'). --JereKrischel 08:14, 19 November 2006 (UTC)

Fellas, it is time to do some archiving. Again, I have not been recently active. But surely much of the discussion on this page either will not lead to edits to the article, or has already led to edits - all such discussion can and should be archived. Please, someone, archive! If any of you find this difficult, consider this a valuable opportunity to take stock of the recent discussions, and prioritize unresolved issues that need further discussion, and then come up with a plan to work through and act on distinct trheads/issues one at a time, and start clearing the plate, as it were. Best, Slrubenstein | Talk 06:48, 26 November 2006 (UTC)

IQ scores adjusted for delayed ontogenesis?

It is well known that physical and mental development between races are, on average, different. Blacks mature the fastest, whites second, and Asians the slowest. In one of Rushton's works he stated that "The average age of walking is 13 months in East Asian children, 12 in white children.." etc. It is also well-documented that East Asians in developed countries have higher life-expectancies (see: life expectancy by country).

This could result in bias against populations that mature slower (South and East Asians, American Natives, perhaps Hispanics) in age-controlled IQ tests; where if, for example, the 13/12 ratio were applied to east asians their average IQ would be recorded as 5-6 points higher than whites.

I have also seen studies showing that the black-white gap increases as the groups age (I'll look for citations); which, if indeed as a result of differing ontogeny, would support the above theory.

Does anyone have more information on delayed ontogenesis and how it impacts mental age of maturity (again, I have heard sources say that blacks reach their peak fastest at around age 18-22 and asians the slowest at 23-25, which is similar to body/bone growth statistics). Also, average verbal and visuo-spatial IQs seem to rise to around 110-113 in east asian populations, depending on the age of the test takers. (derived from Arthur Hu's tables; need verification), with verbal IQs in age-controlled subtests increasing as people age (on average; fluid-IQ data gathered by Hu), and especially as they mature.

Similarly, when controlled for IQ, those with more testosterone (read: blacks, and men) perform better on visuospatial tasks while those with less (asians, women) have better memories and verbal skills, which is contrary to the popular opinion that asians are inferior in terms of "verbal IQ". Based on most studies for sexual dimorphism in the brain, asians tend to be less left-brained in sub-test results; athough average IQs blur out some of the distinction. It should be noted that verbal IQ on subtests (often sentence forming) are largely memory and logic based.

I am not making any assertions, as I do not have citations (I have to track them all down and verify them). Rather, I'm curious if anyone else has data that either denies or confirms this.

Sounds interesting, you should add it when you find the sources, assuming that your sources talk about brain development and not just body development. JeffBurdges 07:14, 3 December 2006 (UTC)
I have some data now, a few examples (2000+ tests each). Each age group/test set corresponds to the
sets of test scores below it.
(Country/Age Range/Test Numbers/Test Name)
(Country/General Score/Verbal Score if Present/Visuospatial Score if Present/Testers)
African populations:
South Africa 8-16 1,220 Progressive Matrices
South Africa adults 703 Progressive Matrices
South Africa 81 Notcutt, 1950
South Africa 75 Notcutt, 1950
This set shows a trend downwards for black South Africans past the "peak age" (14-16) and supports
claims that black children test higher at earlier ages due to age-norming.
North American Native score (shown to have similar ontogenic patterns and a "visuo/verbal gap").
Does trend toward a higher score (3-4 points over 7 years, I don't know if that is significant or not)

Ojibwa & Crees,

 Canada            6-7      33      WISC
   "               9-10     31      WISC
   "               14-15    36      WISC

Ojibwa & Crees,

 Canada            86       70      101
   "               90       81      100
   "               89       78      101

Ojibwa & Crees,

 Canada            St John, Krichev & Bauman, 1976


Navajo 5-8 44 WISC
Navajo 6-7 26 WISC
Navajo 79 65 93
Navajo 78 66 96
Navajo Cundick, 1970
Navajo Cundick, 1970


In the first study the hypotheses is supported; in the second, it is unclear as the average age is the
same. However, there was a one point increase with the presence of 8 year-olds, so it may a clue later.
For East Asia:


Japan 6 240 Vocabulary-spatial
Japan 11 240 Vocabulary-spatial
Japan 97 89 105
Japan 102 98 107
Japan Stevenson, Stigler, Lee, Lucker
                   Kitamura and Hsu, 1985


Another one (China vs. Singapore; not necessarily a good comparison but this is the only other set of
data with the same test and tester)


P R China 101
Singapore 110
P R China 6-16 5,108 Progressive Matrices
Singapore 13 147 Progressive Matrices
P R China Lynn,1991
Singapore Lynn, 1977b


In Japan, verbal scores had a major jump (9 points) between ages 6 and 11; and a 2 point increase in ::::visuospatial.
In the China/Singapore comparison, there was a 9 point increase in general scores over a 2.5 average
age difference; however, China has a vast spread of economic status (malnutrition) so this only weakly
supports my hypothesis.
|Whites|


Belgium adults 247 Culture Fair
Belgium 10-16 920 Culture Fair
Belgium 98
Belgium 104
Belgium Buj, 1981
Belgium Goosens, 1962
Unfortunately the data on Belgium is the only info I can find that has consistent tests with differing
age groups. The testers were not the same; though I believe the test is. However, it follows the trend
of european scores dropping towards 100 at maturity; with higher scores at a younger age (comparatively)
Germany 7-11 454 Coloured PM
Germany 6 3,607 Coloured PM
Germany 100
Germany 102
Germany Kurth, 1969
Germany Schmidtke, Schaller & Becker, 1978
Here is another with the same test; however, different testers. This set of scores also follows
the aforementioned trend. But they were tested in different regions.
All of Buj's 1981 tests:
Britain adults 1,405 Culture Fair
Bulgaria adults 215 Culture Fair
Czechoslovakia adults 363 Culture Fair
Britain 100
Bulgaria 94
Czechoslovakia 98
Britain Buj, 1981
Bulgaria Buj, 1981
Czechoslovakia Buj, 1981


It does appear that Africans tend to trend towards a lower score at a white age standard for maturity,
that europeans decline slightly (but not as significantly) towards 100, and asians trend upwards towards
108~ at maturity. There were no adult scores for Asians, so unfortunately I don't have data on that either.
It should be noted that Cattell Culture fair scores have a higher SD, iirc. Does anyone else have additional test sets for comparison? Sorry for the terrible format, fixing it as best I can

FYI - murray and flynn debate

link -- i would draw the attention of my fellow editors to the last word from James Flynn where he compliments Arthur Jensen and decries those who call Jensen a racist, a POV that I would encourage you to adopt. --W. D. Hamilton 07:51, 30 November 2006 (UTC)

Can you provide the quote from Flynn? It didn't seem to be on the page you linked to. Of course as editors we should respect the other POV, but if we report accusations of racism it does not mean that we hold those same perspectives - just as it is possible to report on racialist research without agreeing with its conclusions or the viability of its basis. The difficulty is of course when people assume that one needs to be racist in order to be mistaken - it can very well be the case that Jensen is honestly mistaken, and not driven by conscious or sub-conscious racism. --JereKrischel 18:59, 30 November 2006 (UTC)

there's no transcript. here's my attempt to transcribe:

we'll i would just like to say... compliment someone who isn't here, and that is arthur jensen. much of my research has been because of problems set by jensen. and how can a man who sincerely speaks what he believes to be the truth be called racist. if what he said was so absurdly ignorant that you knew that someone of his education couldn't really believe it, then you would question his motives. but he is a competent scholar, he's aways given me access to his data...
You're not protecting blacks or Chinese or Irish if you make their plight un-discussable. Anything that's undiscussable you then leave to prejudice and opinion and ignorance. They are the only gainers when you ban discussions of this sort.

--W. D. Hamilton 02:02, 1 December 2006 (UTC)

Major hat tip to Rikurzhen for the link - the mp3 should be required listening for anyone editing R&I articles. --JereKrischel 05:42, 1 December 2006 (UTC)
The video isn't much more informative, but Murray's slides are available. Flynn draws on a board, but I can't make it out. AFAIK the new papers they promise aren't available yet. --W. D. Hamilton 08:56, 1 December 2006 (UTC)
Possibly offtopic and probably not appropriate, but was it my imagination, or did it seem that Murray was in fact denying the relevance of the question as to whether the difference was genetic or environmental, and simply stating that it was unreasonable to attempt to close the gap implying that further environmental modification was not cost effective? One of the questions in the Q&A period seemed to imply that Flynn and Murray simply looked at the same data and came up with rationales to match their politics (Flynn the socialist, Murray the libertarian). Given that position, it strikes me that perhaps the entire idea of R&I explanations needs to include some sort of category for "yes, environment may be the cause, but it is too costly to attempt to remediate". All in all it was very informative to see how much they did agree on, and how their disagreements mostly stood on policy implications. --JereKrischel 09:07, 1 December 2006 (UTC)
Murray makes the point more clearly in his 2005(?) Commentary article. While the theory attractively fits the two data points, it's worth noting that Murray is a rather unusual libertarian (see In Our Hands). --W. D. Hamilton 09:19, 1 December 2006 (UTC)


Average gaps among races

I believe this image used in the article is incorrect, although I haven't looked up the source because I no longer have access to a science direct account (but maybe it's wrong too). One should be able to differentiate along the y-axis and get a bell curve characteristic to an IQ distribution, but instead you would get an inverted bell curve. In other words, there shouldn't be a large spike of population with IQ's on the fringe, that should be in the middle. Perhaps this curve is correct but the labels and units on x- and y-axes have been flipped, although I don't know how that could have actually happened if this plot was actually graphed. Perhaps someone was trying to eyeball it and drew the curves in.Ajnosek 22:39, 2 December 2006 (UTC)

I'm not sure that IQ is relevant to the article topic, but that is not your query. Are you differentiating with respect to the IQ parameter? If so, the maximum value of the derivative occurs at IQ=100, as you would expect. What am I missing? --Walter Siegmund (talk) 01:51, 3 December 2006 (UTC)
This is simply a plot of the Excel NORMINV function -- the inverse normal. --W. D. Hamilton 02:10, 3 December 2006 (UTC)
It's a *cumulative* total. The spike at the end isn't an increase in frequency, it's showing the asymptotic curve as you get to the top of the distribution. --03:07, 6 December 2006 (UTC)

JK edits of 4 Dec 2006

Outlined chronologically

1

Average gaps among races - - removing irrelevant mention of other characterisitics besides intelligence

  • Nectar previously made a compelling case that we should note that cognitive ability is not the only psychological variable that shows race differences.
irrelevant. We're not here to argue the case for race and intelligence differences based on original research, and bring in information we feel might be suggestive to support one side or the other. Reverted delete. --JereKrischel 02:07, 5 December 2006 (UTC)
  • The public is highly misinformed about intelligence research, and so we have high hurdles to overcome in explaining R&I. thus, we need to present the full spectrum of evidence.
Connecting tangential data is arguably original research. It would be equivalent to posting research regarding the tremendous height gains by the Japanese post WWII as evidence that certain racial differences thought to be constant were in fact quite variable. --JereKrischel 02:07, 5 December 2006 (UTC)

2

U.S. Black-White gap - - explicitly noting conclusions are not findings of fact)

  • Who says that achievement tests narrowed during the 1990s?
  • Charles Murray claims that -- does anyone disagree? If not, inline citation is unnecessary.
  • "There is disagreement about whether the results of IQ tests show a narrowing of the IQ gap." - You removed b/c of the agreement between Murray and Flynn in the debate about children up to the 70s, but they don't agree about what has happened to adults (and thus to the scores of people for most of their lives), and they don't agree about the net shrinking at present for all ages. This is a big disagreement.
  • "Roth et al. 2001 found". "found" is the right word here. Flynn agrees that that's what they found and he also agrees that their finding is true (see his 2006 rebuttal paper).
Murray explicitly drew trend lines that were arbitrary (see his powerpoint), and included. There is hardly universal agreement that the 90s present a clear trending. We should also be more specific in the note "1990s to 2005", otherwise, there is room for interpretation there (Murray's decision on when the 1990s started, for example). --JereKrischel 02:16, 5 December 2006 (UTC)
I think given the fact that the trend lines "during the 1990s" are ambiguous, we should either specifically state a year range for which things narrowed/widened, or make clear that there is some ambiguity as to where the trend line is. It also seems to be important which age group is being discussed, as they have distinctly different data points and trends. --JereKrischel 03:04, 5 December 2006 (UTC)
The term "achievement gap" specifics a narrow range of ages and kinds of tests -- school age and tests that focus on school-learned material. We can reexamine the relevant review papers, but Murray and Flynn (in the debate) are not talking about these kinds of tests. The Gottfredson/Nisbett papers discuss achievement tests. The NAEP is the premier achievement test along with the SAT and GRE. You can see the trendlines yourself here for the NAEP. --W. D. Hamilton 04:18, 5 December 2006 (UTC)
Yes, I saw those trend lines, and it seems like the global warming data - depending on exactly what subset of the data you look at, you can claim a trend one way or the other...at the very least, we should specify the exact start and end range being claimed for a trend, rather than just "the 1990s". --JereKrischel 05:02, 5 December 2006 (UTC)
I believe Gottfredson merely referred to decades in her review paper. I don't see the ambiguity that you claim exists in the NAEP data. In zero of the six charts does the BW gap shrink from 1990 to 1999. In 2/6 of the Hispanic-White gap there is shrinking of a couple points. --W. D. Hamilton 16:37, 5 December 2006 (UTC)
I believe in all if not most of the charts, the year with the biggest gap is not 1999, so an arbitrary choice of year within the 1990s could lead the trend line one way or another. It's probably better to be specific with the year here. --JereKrischel 18:18, 5 December 2006 (UTC)
Nothing we do here can get a more specific number/time than what's in the papers. AFAIK, all that's mentioned is the time frame of decades. In the 90s means "during" the 90s -- probably a linear regression of gap against time stratified by decades. --W. D. Hamilton 18:35, 5 December 2006 (UTC)

here's what Gottfredson wrote: --W. D. Hamilton 19:01, 5 December 2006 (UTC)

Black–White achievement gaps in the 1971–1999 NAEP Trend Series were no larger or smaller than g theory would predict. The maximum expected is 1.20 standard deviation (the size of the Black–White g gap itself), and the minimum is 0.80 � 0.04 standard deviation (1.20 multiplied by the IQ– achievement correlations in core subjects). NAEP gaps narrowed from 1.07 standard deviation in the 1970s to 0.89 in the 1990s when averaged over all three subjects and ages. Degree of narrowing stalled by the mid-1980s and differed by subject: 25% in reading (1.06–0.79), 20% in math (1.07– 0.87), and 15% in science (1.22–1.04). As of 1999, all gaps for 9-, 13-, and 17-year-old students were still near or above the minimum expected (reading—0.80, 0.73, 0.73; math—0.82, 0.93, 1.06; and science—0.97, 1.06, 1.07).

3

U.S. Black-White gap - - explicitly noting conclusions are not findings of fact

  • tried to reinstate this edit with my own words

4

Environmental explanations - - "partly-genetic" probably poor wording, since it doesn't denote the amount of genetic contribution with any specificity

  • as a term of art, "partly" is more precise than "significantly". "partly" implies a range like 20-80% heritability to a behavioral geneticist (an effect size of more or less is essentially nothing or all). "significantly" implies that an effect is statistically significant, which is an even larger range depending on sample size.

Can we find a more appropriate layman's term? Or perhaps mention the 20-80%? And doesn't heritability not imply genetic causality? --JereKrischel 02:23, 5 December 2006 (UTC)

In most publications the proponents of the "partly genetic" view claim to be agnositc about the exact size of the gap. I believe we cite Reynolds on 20% being a lower bound. I'll think on it, but we're obliged to use the most commonly used terms in the literature. --W. D. Hamilton 04:21, 5 December 2006 (UTC)
I've explicitly cited the figure you've posited as a general range (20-80%)...I think this helps differentiate between specific references better (for example, when someone argues against "culture-only", and they really mean 0% genetic/100% environment, rather than simply <20% genetic. If "partly-genetic" is as much of a commonly used term (for both the pro-hereditarian and the anti-hereditarian POV) as you state, I have no objection to using it with explicit labels. I'd appreciate further citations from anti-hereditarian sources using the same term - all too often, I see loaded terms in this series of articles that hew to the Rushton/Jensen camp bias. --JereKrischel 04:59, 5 December 2006 (UTC)
Herrnstein and Murray in The Bell Curve use r= .4 - .8 and state that the trend is more towards .8 IIRC --70.91.235.10 03:14, 6 December 2006 (UTC)

5

again, removing culture-only/partly-genetic as improper strawmen

  • you cannot treat the published and cited conclusions of a large scientific survey as an "improper" strawman. on the explanations page, you were unable to cite anything like the required direct evidence to try to overturn the body of literature that identifies the debate as between two camps: one which emphasizes a role for both genes plus environment and anther camp that says environment alone is sufficient.
  • the actual content of this is edit is relatively innocuous (see #4 above). culture --> environment is of little relevance, but culture is better b/c that captures the connotation that is most commonly implied (cultural effects were suggested by Flynn), but "environment" is a term of art for behavioral genetics.

I quite clearly demonstrated that the argument is not simply culture-only/partly-genetic, in a host of points listed on R&I Explanations talk page. --JereKrischel 02:30, 5 December 2006 (UTC)

For example:

[1] http://taxa.epi.umn.edu/~mbmiller/journals/pppl/200504/2/320-2.html

However, their either-or method of scoring the evidence between the genetic versus culture-only data implies a misleading dichotomy (Deary, 2000). There are clear interactions among genetic factors, anatomical structures, culture, and environment. The importance of particular interactions may vary depending on an individual's circumstances and not their racial group membership.

[2] and here - http://taxa.epi.umn.edu/~mbmiller/journals/pppl/200504/2/295-2.html

Ruston and Jensen (2005) make what I believe to be ambiguous references—for example, speaking of biological inequality without defining this term. I also believe they inadvertently create “straw men.” These straw men take the form of false dichotomies, such as between the culture-only model and the hereditarian model (as though there is nothing in between), and imaginary oppositions, such as between people who believe in the influence of genetics and people who engage in “denial of any genetic component in human variation.” There are probably no
such people, at least among serious scientists. What scientist, for example, believes that height or weight is entirely environmental?


It seems very clear that those arguing against the 20-80% genetic hypothesis find the "culture-only" category an improper strawman. --JereKrischel 02:49, 5 December 2006 (UTC)

In contrast -- Flynn believes that the black adolescent subculture that devalues education is largely responsible for widening the IQ gap. “It is more probable than not that the black/white IQ gap results from environmental factors,” he declared.[4]
Unfortunately, these quotes (A) demonstrate an ignorance of genetics on the part of the writers and (B) represent a minority view. About (A) - On [1] - this is true by irrelevant. Each individuals phenotype is a product of the interaction of their own genotype and environment (plus chance, etc). BGH is only ever true at the population level, and does not apply to individuals. (The same is true of WGH.) On [2] - Jensen for example obviously knows that BGH is a continuous variable, as he proposes formula to calculate it. However, as Snyderman and Rothman, the authors of the APA report, later Reynolds, and doubtless more have written, opinions tend to divide as to whether you think genetics contributes to the gap into two camps. Thus on (B), while we can certainly mention this "straw man" objection, it would be inappropriate to favor this minority opinion over the firmly established majority opinion that there are two camps. --W. D. Hamilton 04:36, 5 December 2006 (UTC)
For quote [1], it is making a clear statement that the straw-man for "culture-only" is a shell game - for example, if there is a genetic predisposition that is not activated unless a certain environment is present, is the "culture-only" true? Or is the relationship between genetics and environment more complex insofar as determining causality (another point not quite emphasized enough - even if we find correlations between genetic group differences and IQ, it does not prove causality).
Second, your quote does not support your assertion. Flynn is not making any statement as to the absolute value of environmental and genetic components - you're cherry picking a quote from him. Please provide a citation showing that the "culture-only is an invalid strawman" is a minority view - it seems to come up consistently in the literature. I have doubts as to any evidence showing that there is a "firmly established majority opinion". Perhaps you can cite at least several people who explicitly argue that they believe in a "culture-only" hypothesis...does Flynn ever use that term to describe his position? --JereKrischel 04:55, 5 December 2006 (UTC)
Like I said - doesn't understand (behavior) genetics. Pinker has written a paper about this issue. It is a common misunderstanding/obfuscation.
The consensus statements and the survey each present the same picture of opinion about BGH. The S&R survey is the clearest and most unambiguous presentation of this point. --W. D. Hamilton 05:04, 5 December 2006 (UTC)
BGH is not the question that was asked. The survey question was "The source of black-white difference in IQ". Heritability may be high even though the cause may not be genetic, even when looking at groups, not individuals. --JereKrischel 05:10, 5 December 2006 (UTC)
That doesn't make any sense to me. The options in the survey are about genetics v. environment. Labeling this as being a BGH question was my own input, but it is correct and should be unambiguous. Heritability is defined as the percentage of phenotype variation that is due to genetic variation. --W. D. Hamilton 05:14, 5 December 2006 (UTC)
It seems that the concept of heritability is not universally understood that way - see the definition I've cited below. I wish I had the entire survey article to see in context, and the original questions asked, but haven't been able to find them online. From the report, "a product of both genetic and environmental variation" does not seem to establish a lower or upper bound of 20% as defined in the article. Do you know if this survey is repeated on a regular basis to capture changes in opinion, or was this just a one time thing? --JereKrischel 18:04, 5 December 2006 (UTC)

6

Accusations of bias - - pinker quote seems out of place - removed. Similarly, as per earlier edits, Pioneer Fund section is referenced, so factual dispute not clear, replacing with POV-section

  • The Pinker quote is a spot on example of the accusation that conclusions are biased by fear of human nature. He wrote a book on the subject, and has more recently written several articles saying that fear of R&I is a problem and that people need to find a way to deal with the topic honestly.
Moved to separate subsection, "Complaints of persecution". Seems like a better separation there. --JereKrischel 02:39, 5 December 2006 (UTC)
No more time, but this is not correct. Fear of an adamantite human nature is the key cause. Persecution is merely on of the effect, along with denial and biased interpretation. --W. D. Hamilton 04:38, 5 December 2006 (UTC)
The topic being discussed are, "Accusations of bias". Pinker seems to be complaining about persecution - although only from the pro-hereditarian POV as a defense against accusations of bias (by implying that accusers are violent, censor others and improperly insult people). It seems that there is probably good evidence that both sides have been maltreated - certainly Pinker incorrectly conflates the bad behavior of some towards others with something inherent in their position. --JereKrischel 04:46, 5 December 2006 (UTC)
I suspect you're misreading Pinker, but I'll have to find the time to reread the original source to comment further. Obviously you should have read it as well. --W. D. Hamilton 04:52, 5 December 2006 (UTC)

The original "accusation of bias" (what was originally in that section) was that social scientists are biased against genetic/evolutionary/innate explanations of human behavior (as described by Pinker). The complementary charge against Jensen et al. was that of "racism" and/or that their research had zero or negative social utility. Thus, we are back to the original problem, which is that the PF/SPLC debate is specifically about racism, and so should be covered as such. Some kind of rearrangement of material between these two sections (&/or creation of new sections) seems to be the solution. --W. D. Hamilton 16:45, 5 December 2006 (UTC)

Interesting, I had no idea. I thought the section was specifically about accusations of bias directed against social scientists who promoted the racialist position - Pinker's statement seemed incongruous to the section heading. Maybe better section titles will help. --JereKrischel 17:56, 5 December 2006 (UTC)

7 - expert opinion section

"culture only" = "environment only" = "<20% BGH". there is no doubt about this from a reading of the literature. to show how this conclusion is obvious, consider a BGH of 10%. that would mean that less than 2 IQ points of the 1.1 SD BW gap among adults is due to genetics. a gap of double that size is what the latest papers suggest may exist between males and females -- a gap that went undetected for decades. such a gap is for all practical purposes entirely environmental/cultural, mostly because it would be highly difficult to establish such a small genetic contribution using even the most powerful methods. --W. D. Hamilton 04:52, 5 December 2006 (UTC)

I think you're conflating some arguments here. Heritability does not mean genetic causality - and I believe the argument being made is that there is 20-80% genetic causality from the "partly-genetic" hypothesis. Perhaps you could find the actual question asked, to see if there was some explicit mention of % genetic and % environment intended for the multiple choice answers? Although it seems you have a definite understanding of what you consider "environment only", I'm not sure if you can apply your understanding to every person who answered the survey. --JereKrischel 05:07, 5 December 2006 (UTC)
If something with around 20% genetic causality went undetected between males and females, and has now been detected, doesn't that mean that we can now also detect similar gaps of that small magnitude when talking about races (ignoring the fact that you can distinctly identify men and women, and can't do the same with "race")? If we found a 18% genetic causality for the B-W gap, would that still be considered a "win" for the "partly-genetic" crowd, if they had previously stated they thought it was at least 20%, or higher, or would that count as a "win" for "culture only"? Have direct predictions been made in the literature, or do they specifically note fuzzy ranges to limit their exposure to failure? --JereKrischel 05:20, 5 December 2006 (UTC)
You misunderstand my comparison to m-f diff. The comparison was of absolute magnitudes of the IQ gap, not genetic contributions. The m-f IQ (g) gap may be around 4 points, if it exists at all. In soc sci a effect size of less than d=.2 is considered "small". Because the BW gap is ~1 sd, the conversion from BGH to d is simple. If BGH = .18, that's a genetic gap of only 3 points. Without actually doing the math, my intitution of the statitics tells me that you would be hard pressed to detect such a genetic gap even with DNA techniques. It would certainly be a victor for "culture only" -- see the discussion about skin color correlation with IQ in the review article set for an example where such a conclusion is debate. --W. D. Hamilton 05:39, 5 December 2006 (UTC)
where the apa report says It is sometimes suggested that the Black/White differential in psychometric intelligence is partly due to genetic differences (Jensen, 1972). There is not much direct evidence on this point, but what little there is fails to support the genetic hypothesis. they are talking about data that is interpreted to mean that BGH is ~10(+/-10)%. in fact, it doesn't mean that, but that's beside the point -- they think it does and interpret that to mean that there is evidence against the partly-"genetic hypothesis". (FWIW -- if the data means anything, which i probably doesn't, it actually implies BGH is at least 50%.) --W. D. Hamilton 06:06, 5 December 2006 (UTC)
Again, I think you're incorrectly conflating BGH with genetic causality - is your understanding of BGH from the literature is that it means that there is a genetic cause? --JereKrischel 05:49, 5 December 2006 (UTC)
Yeah. Jensen 1998 gives the definition in terms of genetic and phenotypic variation. --W. D. Hamilton 05:59, 5 December 2006 (UTC)
It seems there is some controversy as to whether or not BGH is universally considered a measure of genetic causality...[5]...perhaps we should more clearly state something regarding that issue in the article, so that it is understood that even though someone may agree that the BGH number for something is "x", it does not mean they also agree that BGH represents causality. If you disagree with the basic premise that Vp = Vg+Ve, heritability measurements end up meaning something completely different. Of particular interest is the detail between "structuring" and "triggering" causes...maybe we can find some better way of making sure we properly distinguish BGH/heritability for the layman. --JereKrischel 06:18, 5 December 2006 (UTC)

Found this:

Heritability is a ratio between the variance of the genetic and the environmental contribution to a phenotypic manifestation in a population under a given set of conditions. [6]

It seems that what it is stating is that it is a measure of genetic causality given a specific environment (Lewtonin's complaint about locality)...this seems like an important caveat that isn't very clear when simply stating that it indicates genetic causality (implying that this would be true in any environmental condition). Is there some way we can make that more clear in the article? --JereKrischel 06:31, 5 December 2006 (UTC)

(para 1) These kinds of details are far beyond the scope of this article and the scope of what any researcher can honestly claim to know about the nature of the BW gap. They can't even nail down the Black WGH to everyone's satisfaction.
(para 2) I suspect anyone who is seriously thinking about it will realize this, but it may fit in somewhere. It's no different than saying "average height is the average of the height of a given population in a given environment". Heritability simply implies nothing about what is counterfactually possible if either genes or environment change. --W. D. Hamilton 14:55, 5 December 2006 (UTC)


But are we agreed that for sake of discussions, if not what is written in the article, that "culture only" = "environment only" = "<20% BGH"? --W. D. Hamilton 16:49, 5 December 2006 (UTC)

I'm fine using that for the sake of discussion, but specifically for the expert opinion survey I think it needs to be established that is what respondents understood or explicitly stated as ambiguous, and it definitely needs to be prominently mentioned for the layman - although the terminology may be conventionally understood in the industry, or from one POV, it does not seem to be universal. --JereKrischel 17:53, 5 December 2006 (UTC)
Tucked away somewhere I have a copy of the survey paper. It's covered again in their 1988 book. In one section, they say that the survey results show that Jensen's view on BGH is the most common one. So at least the authors of the survey interpret their results the way I'm trying to describe it here. Thus, I think it's safe to say that their readers did as well (I've seen no published objection to this survey along the lines that the questions are misleading -- other objections are made about their coverage of media portrayal). And thus, finally, I think it's safe to say that "culture only" = "environment only" = "<20% BGH" is a common understanding of the terms (at least circa 1988). --W. D. Hamilton 18:23, 5 December 2006 (UTC)

NPOV problem - brain size section

There are several hotly debated studies which have concluded that there are race-based differences in brain size. These studies are criticized as having found no racially-based difference, with critics arguing instead the the variation correlates much better in terms of geography.[57]

The main correlation with brain size is height/size; because of this average black/white brain is certainly larger than the average Asian brain (not proportionally, but in Absolute terms). Witelson’s, Kigar’s and Thomas’ (1999) examination of Albert Einstein’s brain illustrates that something more complicated than a brain’s size relates to it’s owner’s intelligence. They compared Einstein’s brain with an average specimen from a sample 35 intact, control brains. Einstein’s brain has about the same dimensions and the same weight as the comparison brain. However, in areas specific to Einstein’s unique skills, his brain was quite different. This leads one to conclude that it is overall brain structure and not brain size that determines one’s intellectual strength.

Herkovitz’s 1930 data suggest that there is no consistent black/white difference with respect to stature or crania. North American Blacks were superior to American Whites in Brain weight (See Tobias, 1970, p. 6:1355 g vs. 1301 g). Cranial size and number of excess neurons of North American Blacks compare favorably to Caucasoids including the English and French. -- The data presented by many of the people you subscribe to, obviously ‘unquestioned’, is entirely spurious and predictive of nothing.


The cited Jensen paper (I993) presents data for blacks and whites, for both reaction and movement time, for three different "elementary cognitive tasks." The results are not, despite contention, "consistent." Blacks are reported to have faster movement times on only two of the three tasks; and they have faster reaction times than whites on one task, "choice reaction time." Simple reaction time merely requires the subject to respond as quickly as possible to a given stimulus each time it occurs. Choice reaction time requires him/her to react differently to various stimuli as they are presented in an unpredictable order. Thus it is said to be more cognitively complex, and to require more processing, than simple reaction time. When Jensen first used reaction time in 1975 as a measure of racial differences in intelligence, he claimed that blacks and whites did not differ in simple reaction time, but that whites, with their higher intelligence, were faster in choice reaction time. He repeated this ludicrous claim incessantly, while refusing to make the raw data of his study available for inspection. Then, in a subsequent I984 paper, he was unable to repeat his earlier finding in a new study described as "inexplicably inconsistent" with his I975 results. Now, in the still newer I993 study, Jensen reports as "an apparent anomaly" that (once again!) blacks are slightly faster in choice reaction time than whites.

I know of no study that set out with the specific aim to find "race-based" differences in brain size. Rather, a variety of sources have happened to collect data which allow one to calculate the average brain size of individuals when divided into races. Thus, the current presentation has several problems: (1) it blurs the proper distinction between data and interpretation, in a controversial topic it is best to maximize this distinction; (2) it gives undue weight to the opinion that biological differences between races are an illusion, to the extent that it completely obscures the fact that so long as race is taken for granted, the existence of race differences in brain size is not an issue (as per Neisser 1997).

It is argued by some scientists that the combination of these correlations, of race to brain size, of brain size to reaction time, and of reaction time to IQ establishes a correlation between race and IQ. This conclusion is strongly objected to by other scientists, both on the basis of the initial correlation asserted between race and brain size, and the progressively weak correlation as the studies are combined.

Without citations, this has to be removed.

The rearrangement of material -- so that the correlation between brain size, reaction time, and IQ is made only half way thru the section -- leaves the reader without an understanding of the relevance of the data for too long. --W. D. Hamilton 05:09, 8 December 2006 (UTC)

I'll put the citations in for you - most of them are straight out of Lieberman. I'm not quite sure if it can be considered undue weight to consider race an illusion (in fact, I suppose we could argue the opposite). Insofar as a variety of sources that "happened" to collect data, you can hardly assert that the interpretation of the data is simply "data" without interpretation.
If you'd like to try a different arrangement of the section, I'm open, but the original layout seemed the product of mismatched edits. I'm not sure if I understand your "so long as race is taken for granted" phrase - isn't it better stated, "so long as a specific view of race is taken for granted"? After all, AFAIK, no large brain size studies have been performed indexed by genetic composition ala Cavali-Sforza, right? Not to mention, even if you did just do it based on say, self-identification, there are issues regarding whether or not the one-drop rule applies, and in which direction. Could you point me to a specific cite in Neisser 1997 that supports your contention? Thanks! --JereKrischel 06:55, 8 December 2006 (UTC)
I'm not quite sure if it can be considered undue weight to consider race an illusion -- "so long as race is taken for granted" phrase - isn't it better stated, "so long as a specific view of race is taken for granted" -- No. There is one common way that the bulk of the U.S. populace understands race. (It is described in brief in the intro section on race.) Although this understanding breaks down at the margins, the fact that there are "exceptions" only proves that there is a "rule". As for the genetics -- Tang (2005) found that for the vast majority (>99%) of Americans self-identified "race" is concordant with genetically-identified group (cluster) membership. Rosenberg (2002, 2005) found that this pattern holds on a global scale. Thus, when you ask no large brain size studies have been performed indexed by genetic composition ala Cavali-Sforza -- the answer is that such studies could only refine what is already observed, having genetic controls won't make the effects go away.
I believe the previous arrangement was quite straightforward. It consisted of a progression of (1) the correlations within populations, (2) the existence of gaps between groups, and (3) the cline/cluster debate (Leiberman) that those who don't believe in races thus obviously don't describe the variation in terms of race. Leiberman's point is quite straightforward as it applies to this article -- if he were talking about skin color, perhaps it would be clearer -- he would argue that races don't differ in skin color because skin color varies within races as well as between races and thus a latitudinal gradient is the preferable description. To the extent that you're interested in describing the geographically distribution of a single trait, there's no reason to argue with Lieberman. It's only when you want to talk about comparing two traits -- say IQ and brain size -- and where you have an immigrant population (like the U.S.) where it is becomes useful if not necessary to make use of racial groups.
Here's the text from the sub article quoting Neisser:
Ulric Neisser, who chaired the American Psychological Association's Task Force Report on Intelligence, examined Lynn and Rushton's data and writes, "although those studies exhibit many internal inconsistencies (and the within-groups variabilities are always much larger than the between-groups differences), there is indeed a small overall trend in the direction they describe."[1]
--W. D. Hamilton 08:17, 8 December 2006 (UTC)
I noticed you missed the next sentence - "Even taken at face value, however, such a trend hardly constitutes evidence for a genetic interpretation. It is already known that body size is strongly affected by environmental variables (most obviously by nutrition);". That seems like an importantly terribly caveat not to include, one that clearly refutes the idea of observed differences being a factor of group genetic differences. --JereKrischel 08:24, 8 December 2006 (UTC)
(1) This isn't the explanations section, so any talk about causes should be withheld. (2) I didn't go to the source, I just copied this from the sub article. (3) I believe this is what the bit about Boas is getting at albeit without the needed citations and/or care that Boas hasn't been preempted by new data.
A point I failed to mention above regarding Leiberman, is that his basic argument (race is an illusion) is not something that should be injected at random points in this article. It should be discussed all at once and then set aside. The foremost reason why this should be done is that is how the debate actually looks in the literature. For example, the APA report completely sets aside questions about the validity of racial categorization. --W. D. Hamilton 08:36, 8 December 2006 (UTC)
Actually, there are even more interesting things to say about all this, but they're beyond the scope of the discussion we need to focus on here. --W. D. Hamilton 08:47, 8 December 2006 (UTC)
I think that if there are disputes about whether or not there are significant differences, it should be mentioned even if we're not in the "explanations" section. Perhaps you can help rewrite the section so that it doesn't get into causes, but also doesn't present the data as being uncontested, nor be leading to a certain conclusion. There is disagreement as to whether or not a) there are significant brain size differences between races, b) whether or not reaction time tests are culturally loaded, and c) whether or not traversing the small correlations across factors yields anything more than statistical noise. Anything that clearly indicates those three points of disagreement would be fine with me.
Also, I think that there is ample example in the literature of challenging the utility of race when challenging data which asserts those groupings. The APA report hardly seems to set aside questions about the validity of racial categories - it states, "These groups (we avoid the term "race") are defined and self-defined by social conventions based on ethnic origin as well as on observable physical characteristics such as skin color. None of them are internally homogeneous." It seems to clearly acknowledge a primary critique of "racial" categorizations (that they are not internally homogenous). --JereKrischel 09:13, 8 December 2006 (UTC)
I think that if there are disputes about whether or not there are significant differences -- The disputes are about how the recognized differences should be described (in terms of clines or clusters, that is latitude or race). The explanations section distinction was aimed at discussion of causes, which is a separate issue. On the 2nd point, I'm not saying there aren't disputes, I'm saying that they should be discussed at once and not a continual counterpoint. Otherwise, every sentence that describes race differences could be appended with the counter claim that some people don't recognize the existence of racial groups. The section as it was dealt with it fairly well, but we've managed to continually change it over and over again. When time permits I'll try to work on it again. It would be better if we could figure out what the underly disagreement between our visions of this section are. --W. D. Hamilton 20:14, 8 December 2006 (UTC)

here is a sampling of past versions --W. D. Hamilton 18:53, 9 December 2006 (UTC)

JK, i'll try to find time to read your new version. --W. D. Hamilton 00:47, 16 December 2006 (UTC)