Talk:Race and intelligence/Archive 44

From Wikipedia, the free encyclopedia

Contents

[edit] Flynn IQ Data are Patent Nonsense

In Asian Asians: Achievement Beyond IQ, James Flynn estimates that when the mean IQ of white Americans is set to 100, the Asian American IQ is 98 or 99. Flynn cites numerous studies Flynn-Effect-adjusted supporting his estimate, almost all of which show Chinese Americans as scoring 100 on nonverbal IQ tests and 96-98 on verbal IQ tests. The 104 figure is probably taken from the famous Stevenson study, in which Stevenson compared children in Japan to children in Minneapolis and found identical IQ scores. The point is that 104 is not Flynn's estimate; 98-99 is. Consequently, someone should change that. Bulldog123 04:53, 31 January 2007 (UTC)

[edit] Mediation?

Are the parties in this dispute willing to submit the dispute to mediation? Based on this talk page, I don't think things will have cooled down enough for us to consider unprotection any time soon. --Ryan Delaney talk 05:12, 26 January 2007 (UTC)

I'd be happy to engage the help of a mediator. I think that as such a controversial subject, we must address some of the fundamental issues with this page before we can find NPOV together. My particular criticisms are in three major areas -> 1) the debate is presented in an invalid strawman POV manner, using the terms and phrases as expressed by the racialist literature. 2) there is excessive point-by-point argument and rebuttal (both in the text and the refs) to a level of detail more appropriate for an article in a scientific journal, rather than an encyclopedia. 3) POV "evidence" is presented as undeniable fact, rather than assertion, in both images and text, often leaving out critical information regarding the data upon which the assertion is based, or even the date of the information.
I'm afraid that WDH is in an unfortunate position here though - he's worked on this series of articles for almost 4 years (as Rikurzhen), and has a lot invested in the current structure and level of detail. I think he's done a great job of building a series of articles worthy of inclusion in a science journal, going into incredible depth with citations (although I may disagree with some of his interpretations), and it must be incredibly difficult to have such a large body of work criticized. I would hope that as an alternative to abandoning the work he has already done, that he takes these articles and hosts them on his own wiki, gnxpwiki.com, where any future article should surely link to. --JereKrischel 06:00, 26 January 2007 (UTC)
Mediation is unlikely to be productive. Third opinions from editors who are more knowledgable about this subject but have not weighed in yet might help, but these editors tend to hold strong views that are likely to prejudice their opinions. I would suggest that Slrubinstein, who has yet to reply to our summary statements above, is the most likely to set aside his own views to render an informed opinion. Ultramarine is equally or more knowledgeable, but I do not trust his neutrality. The issues that JK lists are certainly important topics to resolve, but I doubt that the outcome of mediation would differ from what's been demonstrated on this talk page. Thus, I think it would be a waste of time. In response to JKs suggestion that I apply my time elsewhere - if I were interested in writing my own opinions about this subject, I would submit them for publication as a review paper. I'm interested in seeing a comprehensive neutral presentation of this topic. I believe that is possible, but that it will require great rhetorical skill from the editors involved and a lot of effort put into reading/library time. I believe I've done my part on the latter, even if I lack skills in the former. --W. D. Hamilton 08:48, 26 January 2007 (UTC)
I think, WDH, the very fact that you consider it necessary to engage in "a lot of effort put into reading/library time" belies the original research nature of your contributions. You are writing here exactly what should be submitted to a journal for publication. You've done a wonderful job on that, but I simply don't think it is encyclopedic. The idea of being "comprehensive" in the way you have expressed it by your edits, and "neutral" seem to be mutually exclusive if you are only "comprehensive" on one POV. You've spent four years building this edifice, and I can sympathize with your emotional attachment to it. In the end I agree that mediation will probably not be productive for you.
That being said, would you be willing to abide by some sort of arbitration, rather than mediation? Or would any decision contrary to your position automatically seem to you as uninformed or inherently biased? --JereKrischel 10:32, 26 January 2007 (UTC)


I appreciate WDH's expression of confidence and posted my response to his and JK's summaries above. The question now is, is my response constructive? I also want to pause over two questions I have. First, I happen to believe that intelligence is highly heritable, but I was taught that heritability cannot be used to make comparisons between different populations. If one can make comparisons between different groups, what is the principle used if not heritability? Second, while I think EDH's claim that anyone who believes group differences have a genetic component must also accept the existence of biological races, I am not convinced. Is it possible that race can serve as a proxy for a genetic population, without actualy being the proper unit of analysis? Also, is it possible - for reasons of statistics, not genetics - that the act of identifying subjects according to race ultimately skews the results? In any event, I wonder whether both JK and EDH would agreee to this statement: the possibility that all differences in IQ scores between races is due to environmental differences does not mean that human intelligence is highly heritable? Slrubenstein | Talk 11:51, 26 January 2007 (UTC)

Actually, Slrubenstein, I think there is a rather simple answer to your question: Is it possible that race can serve as a proxy for a genetic population, without actualy being the proper unit of analysis? the answer would be an emphatic yes if you believe in the clinal definition of human variation, as advanced by the likes of Cavalli-Sforza, Beals and Lieberman. I strongly believe that human variation exists along so many dimensions (you could also say "factors") that trying to reduce it to a single Black-White-Yellow-Red (oversimplification is on purpose here, while I don't mean to offend anyone) set of categories is hopelessly simplistic. Racialists usually use Principal Component analysis to be able to reduce the clinal view of human variation (which is now by far the most accepted by anthropologists) to be able to say that race exists and is a useful categorization. "Race" just happens to the most visible resultant of human clinal variation, not necessarily the most significant, and therein lies an vital distinction. A basic question to be asked is: why would genes that determine skin tone also have a significant influence on intelligence? What, if any evolutionary purpose would it serve? Not sure if this helped answer your question.--Ramdrake 13:51, 26 January 2007 (UTC)

It helps - but I want to hear what WDH (who I assume will agree) and JK (who may disagree or agree with qualification) think. This still does not handle the question of how you measure the genetic component of inter-group differences, since heritability is inappropriate for that. Be that as it may, while I appreciate your answer ultimately we will end up back with the same basic questions: what verifiable and relevant sources take this view (I know this will be easy to answer), what verifiable and relevant sources contest this view, and what place should any debate over this view have in the article as a whole i.e. should this argument dominate the structure of the article, or is this one of many smaller debates that need to be organized according to some other themes? Slrubenstein | Talk 14:01, 26 January 2007 (UTC)

I believe Ramdrake has explained it well. You are correct that heritability is inappropriate to use between groups, or to assert genetic causality. [1] - In fact IQ is a great example of a trait that is highly heritable but not genetically determined. Recall that what makes toe number genetically determined is that having five toes is coded in and caused by the genes so as to develop in any normal environment. By contrast, IQ is enormously affected by normal environmental variation, and in ways that are not well understood. As Herrnstein and Murray concede, children from very low socio-economic status backgrounds who are adopted into high socio-economic status backgrounds have IQs dramatically higher than their parents.
However, it should be noted that pro-racialists have staged a strategic retreat over the past hundred years. It was assumed, at some points in the 20th century, that biological determinism was paramount, and that things like intelligence were significantly affected by genetics no matter what the environment. As the evidence for more and more environmental components builds, the pro-racialist/hereditarians have simply begun to claim that any genetic contribution, no matter how small, proves their point.
It may be sufficient to plot the axis as MacKenzie did - "Genetic", "Environmental" and "Jointly Genetic/Environmental". This does not necessarily address the "race is a valid proxy" issue, and pro-racialists and anti-racialists may insist that they belong in the "jointly genetic/environmental" category, rather than at one of the extremes. However, MacKenzie does prominently cite Jensen in the "Genetic" category, whereas WDH seems to see Jensen as a "jointly genetic/environmental" character. Jensen states, All the major facts would seem to be comprehended quite well by the hypothesis that something between one-half and three-fourths of the average IQ difference between American Negroes and whites is attributable to genetic factors, and the remainder to environmental factors and their interaction with the genetic differences. (Jensen, 1973a, p. 363) With this kind of assertion, he is placed into the "Genetic" category.
So if we do make a 3-way split, I think it is important for us to find some sort of guidelines for placing arguments made by different parties in their proper place. WDH would place Jensen in the "partly-genetic" category, and mislead the reader into believing that all jointly genetic/environmental evidence backed up his POV of 50-75% of any IQ difference was due to genetics. --JereKrischel 19:20, 26 January 2007 (UTC)
Also, reading over this heritability article, it seems that there is even another view - that there are genetic differences between groups, but that the B-W gap as it currently stands is a product of extreme environmental disadvantage for Blacks, which if removed would show that genetically, blacks were superior to whites. In other words, if there is a 15-point IQ deficit for Blacks, one could claim that 17-points of that deficit were due to environmental factors, and accept the fact that there is a +2-point genetic bonus that was offsetting the deficit. At issue here is relative magnitude of the causes, as well as the assumed direction of the cause - why do we always assume that a 15 point deficit must have deficit points from both environment and genetics? --JereKrischel 20:01, 26 January 2007 (UTC)
As the evidence for more and more environmental components builds, the pro-racialist/hereditarians have simply begun to claim that any genetic contribution, no matter how small, proves their point.
And what is their point, exactly? That all people are not equal, that it is possible, or even reasonable to assume that social inequality is a result of the genetic inferiority of the disadvantaged, rather than racism, classism, social or economic inertia, cultures of poverty or any of the other social and environmental factors that we know exist. Rather, it is a result of unidentified, not terribly well understood genetic codes. This is something that needs to be said.
As far as mediation, I think we're making progress with this new outline. I think that we can use it to make this article work. I'm trying to stay cool through this whole process, but I do see progress. Let's keep talking and call for help if we hit a wall at some point. futurebird 21:16, 26 January 2007 (UTC)
As the evidence for more and more environmental components builds, the pro-racialist/hereditarians have simply begun to claim that any genetic contribution, no matter how small, proves their point. -- futurebird, this reflects JK's personal opinion and not that of any source that i've seen or he has offered. --WD RIK NEW 03:38, 27 January 2007 (UTC)
I'm surprised you'd make that assertion, given your understanding of the history of biological determinism and eugenics. Have a look at Inheriting Shame: The Story of Eugenics and Racism in America by Steven Selden for more information. --JereKrischel 09:28, 27 January 2007 (UTC)

SLR's question is the correct one: Be that as it may, while I appreciate your answer ultimately we will end up back with the same basic questions: what verifiable and relevant sources take this view (I know this will be easy to answer), what verifiable and relevant sources contest this view, and what place should any debate over this view have in the article as a whole i.e. should this argument dominate the structure of the article, or is this one of many smaller debates that need to be organized according to some other themes? The first half of Ramdrake's response was correct (as in it mapped to an actual POV) until his questions (A basic question to be asked is: why would genes that determine skin tone also have a significant influence on intelligence? What, if any evolutionary purpose would it serve? Not sure if this helped answer your question.) B-W skin color differences appear to be caused by about 5 loci of major effect, one or two of which are now known. BW IQ differences would certainly be caused by many loci of small effect (i.e., not the same genes). The point of PCA is to look for correlation structure in a data set. Race is seen clearly only in the correlation structure of a large number of variables (phenotype or genotypes) not in any single trait (whether this a strength or weakness of race is a central matter of dispute), which could be clinal or have some other geographic distribution due to demographic history. (Of course note that the immigrant and admixed populations of the Americas and other regions have broken the clinal pattern of Old-world variation. In the U.S., race becomes more apparent as clines do not exist there.) What JK wrote doesn't appear to correspond to any significant POV I recognize, and so I will ignore it for now to concentrate of SLR's question. In the section above, SLR appears to have understood my criticism of JKs 3 point distinction. JK responded with a table showing 4 points, but the problem is that the fourth point is an empty set (no races - race differences caused in part by genetics). WP is in the business of describing actual views. It's my opinion that any discussion of the nature/existence/utility of race that is aimed at race itself rather than as a way to argue that observed phenotypic differences couldn't/probably-don't have a genetic basis is a separate issue, and should be discussed separately. Right now that POV is described in the initial section called race. I'm not arguing for it's sufficiency, just for its separateness -- in the past i've tried to expand it but that was reduced by other editors. The article on race is the appropriate venue for lengthy discussion of that general topic, and it should be summarized here. Thus, to the extent that arguments about the nature of race are made in relationship to the question of what causes group differences, they all appears to be aimed at defeating suggestions of a genetic contribution (a point JK at times has conceded). In response to SLR's other question: how do you determine what's called "between group heritability" - most of the data is indirect (you can't study BW twins!) but there are molecular genomic techniques (e.g. MALD) which could answer the question directly, but have not been applied. If I've missed something in this long, single response, please let me know where to fill in details. --WD RIK NEW 20:15, 26 January 2007 (UTC)

And you're absolutely right. However, most of the studies of the racialist researchers are based on self-defined racial identification, which raises the pertinent question, and considering most of those studies were done in the States, of whether self-defined racial identification is really based on the full range of loci that may be construed to identify a race under PCA, or may be more simply based on a most visible criterion, such as actual skin color. Take one eample, say Whitney Houston would certainly be identified as Black through her skin color, but a large number of her facila features might be considered as White. I wasn't totally serious when I threw the question you noted, but I wanted to make a point that there can be a large enough difference between genetic-racial identification and visual-racial identification that what might be true in one case might be wrong in the other. Hope this makes more sense now.--Ramdrake 20:30, 26 January 2007 (UTC)
Ramdrake, this research has been done. Here is the paper.[2] Self-identified race corresponds with genetic cluster assignment with nearly perfect correspondence. Of course, this is all part of the argument about race, which we don't need to have ourselves. --WD RIK NEW 20:36, 26 January 2007 (UTC)
WD/RIK - this research does not claim what you believe it to claim. Being able to find genetic markers that cluster with self-identified race does not build a case that all genes involved in intelligence are clustered in the same way. I believe this is a clear example of the fourth quadrant I illustrated - that there are racial differences by genetics, but that IQ is not mediated by them. --JereKrischel 22:33, 26 January 2007 (UTC)
(1) You've missed the point I was making. I was answering Ramdrake's question about whether the clusters identified by PCA can be assumed to correspond with racial labels. The cause of any phenotypic difference is an open question under the POV represented by the paper (I know this from other papers by this group). This is the same POV presented by Jensen (1998), before he offers evidence that there is likely a genetic contribution to group differences in IQ. (2) This paper has nothing to do with intelligence, just like most discussions of race do not have anything to do with intelligence. It's not in the fourth quandrant (race exists but group differences are caused by environment) because it doesn't answer any questions about the cause of group differences in intelligence. My criticisms about those instances of you're #3 really being specific cases of #2 remain. I look forward to suggestions from everyone about resolving this issue and the others mentioned by JK at the top of this thread. --WD RIK NEW 01:03, 27 January 2007 (UTC)
It is not the same POV as presented as Jensen. It certainly is the fourth quadrant, because it is fully compatible with the idea that there are racial genetic differences, but as per the body of work regarding IQ, the genetic contribution to IQ is not nearly as critical as the environmental contribution (see Erik Turkheimer 2003). --JereKrischel 09:45, 27 January 2007 (UTC)

This debate format is certainly not productive. Editors of this article will have to learn to play together. This is not optional. You guys can argue for months on the talk page and all you're going to accomplish is proving to me that the article needs to stay protected in its current form. --Ryan Delaney talk 10:34, 27 January 2007 (UTC)

Well, my bias in this specific question is obvious, but contrary to Ryan Delany's comment above, I think this debae has been very constructive. I think JK and WDH/WRN have, in the process of answering my various questions, both clairified their own positions considerably, have focused on key claims, and are helping narow the descussion down to basic points of contention. i think this is a useful exercise and already wonder if some material introduced by them in this section (and in the "long discussion" section above, where I just left a query for WD RIK NEW (aka WDH)) might helpfully be incorporated into the article. Slrubenstein | Talk 13:26, 27 January 2007 (UTC)
That said, I now have two more questions, specificaly for WRN but incidentally for JK and Ramdrake. (1) WRK writes above, "Race is seen clearly only in the correlation structure of a large number of variables (phenotype or genotypes) not in any single trait (whether this a strength or weakness of race is a central matter of dispute), which could be clinal or have some other geographic distribution due to demographic history." If I understand this correcly, it means that the conclusions one reaches about the correlation of race and intelligence depend on choices one can and must make prior to the statistical analysis. I wonder whether these choices are themselves the source or point of contention. If I am right I think the article needs a clear explanation of what the possible choices ae, how different choices can lead to different outcomes, and which choices specifically are points of contention, for whom and why. if I am wrong, I think the article needs to explain why (i.e. why there is no choice in the design or execution of the statistical analysis, and the way people like Rushton or Jensen or others have analyzed the statistics is the only proper, or clearly best, way to do the statistical analysis. (2) Above WRN writes "In response to SLR's other question: how do you determine what's called "between group heritability" - most of the data is indirect (you can't study BW twins!) but there are molecular genomic techniques (e.g. MALD) which could answer the question directly, but have not been applied." If this is so, I think the articl needs to be painfully explicit about this, and about any debates over how this indirect data is used. Slrubenstein | Talk 13:26, 27 January 2007 (UTC)
Both JK and WRN seem to agree that my questions are valid ones, and both have provided their answers. It is my hope that this process brings greater clarity and precision to the conflict between them, and helps us focus on very specific issues that have to be addressed most carefully, clearly, fully, and with special care to follow our WP:V and WP:NPOV policies. It is my hope that if JK and WRN do this, the result might be an article that both are satisfied with. That would be an article that includes views they personally disagree with, but acknowledge are held by enough people and are adequately relevant to merit inclusion in the article, and include the views that each most stronly feels need to be represented in the article. Slrubenstein | Talk 13:26, 27 January 2007 (UTC)
I'll try to give my viewpoint on Slrubenstein's observation, which I think he may be right is central to the debate the conclusions one reaches about the correlation of race and intelligence depend on choices one can and must make prior to the statistical analysis. Let me turn the clock back about 6 months, when we were having (another) hot debate on whether brain size (wherein skull size is considered a good but not perfect proxy) was racially-dependant. The article came out and affirmed this was the case, based on Rushton's interpretation of, among other things, Beals' studies. Now, if you go back directly to Beals' studies (which made comparisons of over 10,000 skulls from nearly every possible geography in the world), Beals' conclusions were that skull size was latitude and climate-dependant (latitude being very strongly correlated with climate, either affirmation can be considered valid). His hypothesis, on which he built an very interesting mathematical formula (with good predictive value) was that the colder the climate got, the rounder the skulls of populations in those areas would become, and rounder skulls had the side effect of yielding larger braincases, therefore larger brains. That's a nice explanation that has nothing to do with race, except the coincidental fact that European Whites lived north of African Blacks. Therefore, the conclusion was drawn that Beals' research lended support to the afirmation that "there is a difference in brain size according to race". That is one way to see it, but there is no certainty that this is the most appropriate way to see it, and even less proof of it. Just like there can well be a genetic component to intelligence, but nothing proves that if it indeed exists, it varies along the same lines as traditional racial distinctions. If it exists, it may vary according to very different princples and may happen to just partially intersect (or not at all) traditional racial boundaries. There are a lot of studies out there, and even more unknowns. Possibly my biggest concern is that the "intelligence varies with race" pseudo-axiom is presented in this article as the mainstream opinion, whereas in fact if you look at all the major fields who have tried to tackle the question (or have had it just thrown in their lap, for that matter), this is is so very highly disputed. In a nutshell, that's about it. Hope it makes sense.--Ramdrake 14:18, 27 January 2007 (UTC)
Question: do JK and Ramdrake feel that this view is adequately represented in the article? Question: would WRN agree that there is a way to incorporate this view into the article that is fully compliant with NPOV and NOR? I ask because I want to move beyond inflammatory language like "pseudo-axiomatic" and see whether we can make real progress in improving the article. I mean no offense to anyone, I am just trying to be pragmatic and constructive. Slrubenstein | Talk 16:56, 27 January 2007 (UTC)
Actually, we just got off topic. SLR, a random selection of traits will be sufficient [3]. These questions about race are beyond the scope of this article. There is no reason to have the race debate in this article when it is better had in the race article. The problem #1 identified by JK is that JK suggests first that race is a major issue in the causation question (arguably true, but overstated), and then that it identifies an independent third position from the two I think are supported by the major review papers. It goes without saying that the question of race is a background question for this topic, but in that content it is not an attempt to directly answer the causation question. -- Step back for a moment. Why do intelligence researchers continue to study race differences? For one thing, because someone started it 100 years ago and they found some substantial differences. More broadly -- why does the U.S. census continue to classify the race of U.S. citizens? Why do social science researchers in a myriad of fields take race as a variable in their studies? The suggesting implicit in some of these comments is that we need to derive answers to these questions a priori before writing this article. We cannot. We can describe the debate in the terms it exists and in its many facets. We can't answer questions that aren't answered for us already. -- Now a note on what Jensen (1998) has said on this subject. If the goal is to argue that race is a bad category, then you're talking at a different level than the concern of this article. If you use that argument to say that observations of race differences are suspect, Jensen argues you are making a confusion. Race is argued to be a bad category because of a non-correspondence of social and biological factors. But if you think the observed phenoypic group differences are due to culture, then race (or race/ethnicity) is exactly the variable that you want to measure because that's the social variable. Arguing that it isn't biological and thus you shouldn't study is inconsistent in Jensen's view. What you can argue, as Gould does in at least one setting, is that what we know about the biological basis of race mostly precludes a genetic explanation for phenotypic differences. --WD RIK NEW 17:01, 27 January 2007 (UTC)
Well, with all due respect, WRN, debates over race are not off-topic as long as (1) race is half the title of this article and (2) debates over the nature of race have entered into debates about the relationship between race and intelligence. Also, with all due respect and with regard to your closing comments about Jensen and Gould, what is important is not what you or I "can" argue but what verifiable sources have argued. I address this point not to WRN exclusively but to JK and Ramdrake too. It doesn't matter whether the points any of us editors make make perfect sense - editors' views do not go into articles. The question is, have these views been presented in reputabl;e, verifiable, sources? And if the answer is yes, then they should go into the article whether we agree with them or not. Slrubenstein | Talk 17:20, 27 January 2007 (UTC)
Ah, I mean off the topic #1 identified by JK above. I point out what Jensen and Gould say because I assume they have better knowledge than us about what arguments have been made and so their comments inform what the debate looks like. --WD RIK NEW 17:38, 27 January 2007 (UTC)
To all editors, I apologize if my language is at times inflammatory. I will try to mind it more carefully from this point on. To answer pointedly your question, Slrubenstein, no I don't feel that the view I'm trying to represent is properly represented in the article, and secondly, one review paper which shows that these views do exist in the scientific community is the following one: How "Caucasoids" Got Such Big Crania and Why They Shrank: From Morton to Rushton. The main review covers about twelve pages, and is followed by comments by nearly a dozen experts (including Rushton). It's not by far the only one, but that's one of the better primers on this particular position that I've seen.--Ramdrake 17:37, 27 January 2007 (UTC)
If we're going to be on this topic, then let me try to cut to a point that will inform about #1. Is Lieberman arguing about (1) race in itself or (2) race as it relates the the causes of phenotypic differences in intelligence (other than tangentially -- of course he argues against Rushton)? I think the answer is (1) and not (2). --WD RIK NEW 17:42, 27 January 2007 (UTC)
I believe Lieberman speaks about race as it has been claimed to influence intelligence, as per the literature of the likes of Morton, Rushton, Lynn, and co. In this respect, it is quite pertinent, IMHO, to the subject of the article. I think his argument against Rushton's theories is more than tangential, or else, Rushton wouldn't be named right in the title (or subtitle) of the article itself.--Ramdrake 20:07, 27 January 2007 (UTC)
I didn't explain my question well enough. My reading of Lieberman is that he's talking about race (and Rushton's use of race to describe brain size differences). He's not making claims about the causes of group differences in intelligence. Obviously it's related to this article--I never meant to imply it wasn't--but in what fashion? Lewontin's 1972 paper is related too, but only in relation to the nature of race itself. I think there are two ways it could be related and that it's one and not the other. The relevance is to the claim by JK that there is a third (or orthogonal) answer to the question of what causes group differences not the genetic versus environment split. --WD RIK NEW 21:57, 27 January 2007 (UTC)

[edit] The critical issue of how much attention we should give to "race"

I wanted to pull out a new section based on something WD/RIK said - There is no reason to have the race debate in this article when it is better had in the race article. I think this may be the very basis of our disagreement. I believe that if we proceed as WD/RIK suggests, and leave the questions of "race" as givens, rather than issues, we do a grave disservice to the reader. I understand that in a meta-analysis paper for a scientific journal, typically you'll state your givens, and then ignore any questions regarding those givens for the remainder of your thesis. This seems to be what WD/RIK is suggesting, and I believe he has some valid reasons for suggesting it - it would greatly simplify going point-by-point through various research papers if the question of "race" wasn't brought up with every reference. This would make the article cleaner. It would not make the article very NPOV, and I assert that it would be very misleading.

I've learned a great deal over the past few months following the bouncing ball to the references WD/RIK has put out here, and nearly every time I've ever dug into the details, the "race" issue comes up - the definition of it, or the usage of it, or the validity of it. No set of studies really ends up using the same definition, although certain folk, like Rushton and Lynn, are quick to conflate things together and ignore the many facets of "race" and how it is used. When we uncritically accept those claims, and tie them together with other studies with similar claims but different definitions of "race", I believe we are engaging in poor encyclopedic work.

I believe WD/RIK's statement regarding avoiding the race debate gives us his true reason for his disagreement with me, and his desire to simply paint the issue as "partly-genetic" and "environment only". He sees the race debate as a distraction to the clean presentation of study, counter-study, counter-counter-study, etc. He is happy to accept that all definitions of race are generally close enough to ignore, and as a geneticist, sees the obvious presentation as "partly-genetic" and "environment only".

I strongly disagree. And although I believe I've stated earlier that one of my biggest concerns was that this entire article seemed like a meta-analysis research paper, upon further contemplation, I think that was simply a knee-jerk reaction on my part. I believe I need to accept WD/RIK's detailed, point by point contributions as valid and not particularly OR, but I only think I can accept that if we are crystal clear on the assumptions of "race" being used in every specific study being cited. If we don't do that, even though the debate may be occurring in the race article, the data and studies being presented here are lacking important context.

I think at this point we need the help of an anthropologist to help us classify the various uses of "race" into manageable chunks (if at all possible). It may be require some index regarding to time (B-W in 1950's America ignoring European admixture in African Americans, or Hispanic pre-"ethnicity" census, or some other indicator), but I'd like to ask WD/RIK if he has any suggestions on how we could categorize citations by their usage of race. Rushton/Jensen and other great works of aggregation (Lynn) would probably be in a category "Non-uniform usage of race in source studies". Individual study citations (those actually doing real research, not just meta-analysis) would probably have a category "Uniform usage of race - self identification based on X groups", or something to that effect. --JereKrischel 07:41, 28 January 2007 (UTC)

[edit] Proposal 1

I think /Proposal 1 from JereKrischel has merit. The outline that he has created provides the framework for an entirely refactored page. This could be what this article needs to get to a NPOV balance. Now, if we can just get along with one another on the talk page.... We need to remember, this page is about the debate on race and inteligence, not a propaganda piece, or an excercise in political correctness. It's meant to cast NPOV light on the whole subject.

  • 1 Definitions of "race"
  • 2 Definitions of "intelligence"
    • 2.1 The Psychometric Approach
    • 2.2 Multiple Forms of intelligence
    • 2.3 Cultural Variation
    • 2.4 Developmental Progressions
  • 3 History of research
    • 3.1 18th century
    • 3.2 19th century
    • 3.3 20th century
  • 4 Mean scores of different ethnic groups (U.S.)
    • 4.1 Test Bias
    • 4.2 Characteristics of Tests
  • 5 Current debates
    • 5.1 Genetic hypothesis
    • 5.2 Validity of "race"
    • 5.3 Socioeconomic factors
    • 5.4 Culture factors

Mytwocents 05:42, 26 January 2007 (UTC)

Thank you, Mytwocents. I took the sections primarily from the APA report WDH cites, as well as Britannica's article, "race" and intelligence. I would be very interested to get input from WDH (or other hereditarian editors) as to other changes to the layout he would suggest, as well as perhaps an idea of where existing parts of the article could be split into. --JereKrischel 06:03, 26 January 2007 (UTC)

I suggest that JK build the article he/she proposes in a temporary location, where it can be evaluated and discussed. --Kevin Murray 06:32, 26 January 2007 (UTC)

This is probably a bad idea, because having "competing versions" is not acceptable on Wikipedia. --Ryan Delaney talk 07:09, 26 January 2007 (UTC)
It't not meant to be a "competing" version so much as an upcoming version, where everyone can contribute within the framework of the outline. Once it's roughed out, it can replace the existing version and the sub-sections filled out to make it a complete article. As of now I think the framework the JK has layed out has promise, but others may differ. JK and WDH may want to do most of the work on the respective sections, since they have been most involved recently. Mytwocents 07:32, 26 January 2007 (UTC)
This outline looks much better than what we have now. Can we make a subpage to start filling it out with things in the current article that work? futurebird 13:36, 26 January 2007 (UTC)
One problem: race is mentioned at the top, and then ethnic groups later on. In most of the sociological and anthropological literture race and ethnicity are two different ideas. people who identify as one can also identify as the other, and people who identify as one can change their identification to the other - but they are still conceptually distinct, and in real life they are not the same. If the mean results are indeed from ethnic groups, then there is simply no reason to introduce race at all. If the mean results are from races, we should say so. If we have a mix of mean results from races and mean results from ethnic groups we should discuss that - but we should not treat them as comparable data. Slrubenstein | Talk 13:45, 26 January 2007 (UTC)
I think in any discussion of "race" section, we need to make very clear the idea of the common conflation of "race" and "ethnicity" that is used in most of the studies cited, and how it differs from genetic clusters/clines. Nearly all of the work by Rushton, Lynn, and other Pioneer Fund folk use ethnicity as a proxy for "race", lumping ethnicities into larger groups, without actually having measured any sort of genetic markers. Since the data we do have is by self-identified ethnic group in the U.S., it seems important to specifically label it as "ethnic" data, but we can explain that some people use these ethnicities as "race". If we were to eliminate any data that was not based on self-reported ethnicity, we probably wouldn't even have a single bit of data - WDH, do you know of any large scale IQ testing that actually measured participant's DNA? --JereKrischel 19:37, 26 January 2007 (UTC)