Talk:Race and intelligence/Archive 16

From Wikipedia, the free encyclopedia

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 1 superior/inferior 1.1 source quotations 2 Stereotype Threat 3 Footnotes 4 The Use of Racial, Ethnic, and Ancestral Categories in Human Genetics Research 5 The initial image implies a degree of overlap that does not exist 6 Genetic structure of the Hispanic population 7 NPOV dispute 7.1 "first assumption" 7.2 bamshad in jama 7.3 nectar's question 8 Some thoughts on Tang et al. and the NPOV dispute 9 first assumption : lewontin was wrong 10 List vs sentence form? 11 "what is race" and more on "Lewontin's fallacy" 12 Wiki is not a place to come for serious information 13 Proxy and nongenetic factors 14 Chart 15 Interdisciplinary field 16 Canadian news on Rushton, race, IQ and genetics

superior/inferior

Skipping around in the summary: For instance, the disparity in average IQ among racial groups is sometimes mistaken for the idea that all members of one race are more intelligent than all members of another, or that ranking group averages from "high" to "low" implies an overall ranking from "superior" to "inferior".

While the first part of the second sentence is true, let's not mince words on this: the race and intelligence findings cited throughout this article explicitly argue that some groups have inferior intelligence than others. That's one of the main points (or perhaps goals) or this research. If one believes "intelligence" is a measurable trait, then some groups will have inferior levels or amounts of it. Inferior means "lower in rank" (first definition in OED). Isn't that what this research is all about-- testing hypotheses of intellectional superiority/inferiority of various "races"?

I am going to remove or that ranking group averages from "high" to "low" implies an overall ranking from "superior" to "inferior" unless there are objections. Jokestress 22:48, 28 August 2005 (UTC)

Probably wasn't clear. Touched it up. --Rikurzhen 00:29, August 29, 2005 (UTC)

The revisions: For instance, the disparity in average IQ among racial groups is sometimes mistaken for the idea that all members of one race are more intelligent than all members of another, or that ranking group IQ averages from "high" to "low" implies a fundamental ranking of races from "good" to "bad" or "superior" to "inferior". This touch-up doesn't address the issue, which is that this article presents evidence claiming that certain "race" groups have superior intelligence than other groups. If intelligence is a metric, then some group has the lowest intelligence, right? This article is making a case that Blacks have the lowest group intelligence among Americans. If you are going to keep superior/inferior, you need to admit that this article DOES claim a fundamental ranking of group intelligence by race. Good/bad is subjective and accurate as used in the sentence above, but intellectually superior/inferior is exactly the case being made. Right? It seems disingenuous to say that some people erroneously assume this from the evidence presented, when that is exactly what the evidence presented is supposed to do-- it suggests Black intellectual inferiority. Right? Jokestress 03:59, 6 September 2005 (UTC)

The point being made is that ranking races quantitatively by intelligence does not imply (1) a moral value ranking (good/bad) or (2) an overall ranking of races (superior/inferior). The "overall" point (#2) was more explicit in the previous edit, and was aimed at the idea that you can't rank races "overall" on the basis of just "intelligence". Can you think of a better way to write that? --Rikurzhen 04:30, September 6, 2005 (UTC)

Well, ranking "races" quantitatively by group intellectual superiority certainly goes a long way toward implying overall superiority. I'd do the last phrase like this:

[...] ranking group IQ averages from "high" to "low" implies a moral ranking of races from "good" to "bad" or an overall ranking of "superior" to "inferior".

I'd also tack on a concluding sentence of some sort that doesn't mince words, saying:

The conclusions that some racial groups are intellectually inferior, and the hypothesis that a genetic component may be involved, have led to heated academic debates that have spilled over into the public sphere. Jokestress 06:48, 6 September 2005 (UTC)

that sounds good.

ranking "races" quantitatively by group intellectual superiority certainly goes a long way toward implying overall superiority -- Murray's contention was that was a cultural rather than logical implication --Rikurzhen 06:58, September 6, 2005 (UTC)

That's an interesting change in wording. The change from "the group has a lower IQ" to "the group is intellectually inferior" shifts from describing the trait as lower, to describing the group as inferior (in the trait). While both sentences are equally correct, "inferior" certainly has other connotations, including the value statement of 'worth less than' ("of little or less importance, value, or merit")(M-W). In my opinion the new version uses vague wording that will certainly be misinterpreted as a value statement, which is not a conclusion of the scientific research. Is there any reason not to use the apparently more precise "the conclusion that some groups have lower intelligence...?"--Nectar ^T 07:42, 6 September 2005 (UTC)

nope... change it. --Rikurzhen 07:48, September 6, 2005 (UTC)

Then we should probably avoid superior/inferior in the previous sentence for the same connotative reason (my original point). The denotation of rank order is covered in lower intelligence and the connotation is covered in bad/good. Jokestress 10:07, 6 September 2005 (UTC)

Inferior/superior is definitely the contrast we want in the preceeding sentence, so Jokestress seems to be arguging for parrallelism, which is very convincing. I haven't mulled it over enough to figure out whether we aren't moving from one definition of inferior to another when we do that contast. I'll try to find some quotations from Murray -- from which this disambugation came -- to see if it clears up his point. I'll paste them below for us to mull over. --Rikurzhen 17:09, September 6, 2005 (UTC)

source quotations

The concepts of “inferiority” and “superiority” are inappropriate to group comparisons. On most specific human attributes, it is possible to specify a continuum running from “low” to “high,” but the results cannot be combined into a score running from “bad” to “good.” What is the best score on a continuum measuring aggressiveness? What is the relative importance of verbal skills versus, say, compassion? Of spatial skills versus industriousness? The aggregate excellences and shortcomings of human groups do not lend themselves to simple comparisons. That is why the members of just about every group can so easily conclude that they are God’s chosen people. All of us use the weighting system that favors our group’s strengths.

If you think this is mushy nonjudgmentalism, try a thought experiment: Suppose that a pill exists that, if all women took it, would give them exactly the same mean and variance on every dimension of human functioning as men—including all the ways in which women now surpass men. How many women would want all women to take it? Or suppose that the pill, taken by all blacks, would give them exactly the same mean and variance on every dimension of human functioning as whites—including all the ways in which blacks now surpass whites. How many blacks would want all blacks to take it? To ask such questions is to answer them: hardly anybody. Few want to trade off the unique virtues of their own group for the advantages that another group may enjoy.

Sometimes these preferences for one’s own group are rational, sometimes not. I am proud of being Scots-Irish, for example, even though the Scots-Irish group means for violence, drunkenness, and general disagreeableness seem to have been far above those of other immigrant groups. But the Scots-Irish made great pioneers—that’s the part of my heritage that I choose to value. A Thai friend gave me an insight into this human characteristic many years ago when I remarked that Thais were completely undefensive about Westerners despite the economic backwardness of Thailand in those days. My friend explained why. America has wealth and technology that Thailand does not have, he acknowledged, just as the elephant is stronger than a human. “But,” he said with a shrug, “who wants to be an elephant?” None of us wants to be an elephant and, from the perspective of our own group, every other group has something of the elephant about it. All of us are right, too.

Stereotype Threat

A very important aspect that is missing in the article is Stereotype Threat (unfortunatly not yet discovered by wikipedia). Detailed information can be found here. from that site:

"Stereotype threat is “the threat of being viewed through the lens of a negative stereotype or the fear of doing something that would inadvertently confirm that stereotype” (C.M. Steele, 1999, p.46)."

"As described by Claude Steele (1997), members of stereotyped groups such as African Americans are especially wary of situations in which their behavior can confirm the negative reputation that their group lacks a valued ability. The extra pressure caused by the fear of reinforcing the negative stereotype interferes with performance, resulting in lower scores”

Steele, C. M. (1997). A threat in the air: How stereotypes shape intellectual identity and Performance. American Psychologist, 52, 613-629. User:Junwerqt1

I was sure we had something about this somewhere, but you point out correctly that there is nothing about it now, though it should. It belongs to Race and intelligence (Culture-only or partially-genetic explanation) and possibly its summary. And welcome to Wikipedia! Arbor 15:22, 30 August 2005 (UTC)

Update: Stereotype threat exists, but it redirects to Stereotype. And Here is an old edition of this article that mention it. By all means bring it back! Arbor 15:31, 30 August 2005 (UTC)

If anyone is interested in working on stereotype threat, be careful about what sources you use. Most new media and even some text books make a crucial mistake when reporting the results of Steele's experiments. Sterotype threat can make the IQ gap worse, but it is not a cause of the documented gap. --Rikurzhen 17:08, August 30, 2005 (UTC)

I disagree. As the existent evidence on the race-IQ-relationship is only correlational, it is impossible to make final statements about its causation. Therefore, unless the account is falsified, Stereotype Threat is a viable explanation for (some) race-related variance in the IQ (even though some of the work on ST is partly miscited). That stereotypes can affect performance in test situations is a fact (see study cited above and a ton of others). As low test performance may lead to deidentification with academics or the subject in question, the effect may perpetuate itself. Of course, its causal contribution to the IQ gap on the societal level is untested as well. --Junwerqt1 01:02, 31 August 2005 (UTC)

Right. There are lots of things that could be the cause. But it has been misreported ([1]) that there is evidence that stereotype is the cause. You seem to be familiar with this -- so carry on :) --Rikurzhen 01:31, August 31, 2005 (UTC)

Footnotes

I'm not sure what happened at a quick glance, but the footnotes did not match up with the article. To fix this, I removed the following:

1. ^  J. Philippe Rushton, "Race, Intelligence and the Brain: The Errors and Omissions of the 'Revised' Edition of S.J. Gould's The Measure of Man," Journal of Individual Differences 23, no. 1 (1997): 169–80.
2. ^  Phil Gasper, "A scientist of the people," SocialistWorker.org, June 7, 2002.
3. ^  Goosed-Up Graphics: A generalization of the Lie Factor, graphs from Full House: The Spread of Excellence from Plato to Darwin, by Stephen Jay Gould (Three Rivers, MI: Three Rivers Press, 1997): 109, fig. 16.
4. ^  Tucker 2002, Lombardo 2002, Kenny 2002
5. ^  "Racial Scientist Rushton Takes Over Pioneer Fund," Bethune Institute for Anti-Fascist Studies, January 2003.
6. ^  Linda S. Gottfredson, "The General Intelligence Factor," Scientific American.

Did these possibly get moved to one of the related articles (and hopefully copied there correctly)? In any case, they didn't seem to correspond to any of the items currently in the text. --Michael Snow 22:31, 29 August 2005 (UTC)

The Use of Racial, Ethnic, and Ancestral Categories in Human Genetics Research

There is a new review paper in Am. J. Hum. Genet. about race and genetics research. Of particular note is the copyright: This article is in the public domain, and no copyright is claimed. Here's the full text from the publisher's web site. If that's not publically available, I'll post it in my user space. --Rikurzhen 08:03, August 30, 2005 (UTC)

User:Rikurzhen/NHGRI-race-review-2005 --Rikurzhen 18:53, August 30, 2005 (UTC)

The initial image implies a degree of overlap that does not exist

While it correctly shows the degree of overlap in distributions, it gives a false picture of the degree of overlap in the US population as a whole. The peaks of each curve should not be individually normalized.

If they aren't given equal heights, then the graph is essentially un-readable as Whites make up >2/3 of the U.S. population and Asians ~2%. --Rikurzhen 17:33, September 1, 2005 (UTC)

There are questions about the accuracy of the Asian data, anyway. Small sample size for data presented. Probably should be treated like Ashkenazi Jews in other one-- mentioned, not graphed. Maybe give an N= for each one in the legend? Jokestress 17:49, 1 September 2005 (UTC)

The Black and White curves have been confirmed for N>100k people. The number of studies with Asian and Hispanic data is smaller, but the lack of precision is like a 2 or 3 point range -- compared to the extreme precision of the White and Black means. Nonetheless, the rank order is certain, so there's no reason to doubt the graphic as a visual aid to understanding how distributions can have different means yet be overlapping. We'd have to do some serious reading to figure out how many people were sampled to get the Asian and Hispanic means -- but the SAT gives the same answers and N there is large. (Number of Ashkenazi IQ numbers is considerably less, like ~50? in The Bell Curve -- and estimates of averages have >5 point spread.) Removing the Asian curve would be unhelpful because so much is made about the higher IQ of Asians -- e.g., that the tests aren't racist, etc. --Rikurzhen 18:07, September 1, 2005 (UTC)

Another reason to mention the Asian/Hispanic imprecision is that the White/Black curves are from the same normalization, but the Asian/Hispanic numbers come from the literature average, which includes a wide variety of tests with different normalizations. If I had to guess, I would say we could be under-reporting the Asian/Hispanic means -- they could be shifted 2 points to the right -- and also note these are two groups for which English could be a hamper to administering some kinds of tests. --Rikurzhen 18:13, September 1, 2005 (UTC)

A basic failure in the image is lack of specific numerical values. A description of percentile, median, and mean would probably be useful in this section. As noted above, language barriers may be an issue, but education should also be taken into account. When scanning this document one reaches a further table of education level (high school completion) correlated with IQ. Apart from the obvious correlation = causation fallacy, the mere fact that a mathematical segment of most IQ tests will ask a sequential numbers theory question based on prior knowledge should be established at this area. I also fail to see the age based accuracy discussed adequately, if at all, within this article.

Genetic structure of the Hispanic population

In case this is useful:

Molecular anthropologist Mark D. Shriver's research at Penn State University found "Hispanics in certain New Mexico and Colorado locales averaged 58 percent white ancestry, 39 percent New World Indian, and three percent African. In contrast to the "bimodal distribution" of blacks and whites in America, Mexican-Americans clustered around their average admixture level of 58 percent European."[2]

--Nectar ^T 06:11, 10 September 2005 (UTC)

check it: Image:Admixture triangle plot.png --Rikurzhen 07:16, September 10, 2005 (UTC)

How does this reconcile with the claims regarding "hybrid vigor" among "mixed-race" kids' IQ scores? Jokestress 09:57, 10 September 2005 (UTC)

I think "hybrid vigor" (aka "heterosis") only works for recent (as in the last one or two generations) admixture. --Rikurzhen 10:04, September 10, 2005 (UTC)

NPOV dispute

Is the NPOV dispute still applicable to this article? In my opinion, this article is a model of neutrality by stating opinions and thoroughly citing them. If anything, it gives too much credence to the "it does not exist" arguments - for example, by going over the claim that "the field is invalid due to the inability of 'Race' and 'Intelligence' to defined" (when in fact Tang et al. 2005 puts to rest claims that self-reported racial categorizations are invalid, and most psychometricians accept the g-based model) early in the introduction. Dd2 19:17, 10 September 2005 (UTC)

If we are ready to get back to the first debated assumption, I am happy to discuss that again. The second debated assumption is fine as long as the opening sentence makes it clear we are talking about inteeligence tests versus intelligence. There's nothing disputed in the statement that Blacks as a group score one standard deviation lower than Whites as a group on intelligence tests, so that is NPOV now. However, the first assumption is not the debate. The debate is about the biological reality of race. Just because Tang shows that genetic markers can distinguish established phenotypic groupings does not make "race" valid. Tang doesn't really put anything to rest except that one can use genetics to reify a folk taxonomy. You could probably use the same technology to predict hair color with similar accuracy, but doesn't make the three hair colors (blonde, brunette, redhead) any less of a folk taxonomy.

So, the issue is the scientific classifciation versus folk taxonomy debate of race. If we are to teach the controversy, this issue should be acknowledged here and expanded on race. Once we get the summary ironed out, I want to go through the rest of the article to discuss some NPOV issues. Jokestress 21:54, 10 September 2005 (UTC)

Sometimes I think Popper did more harm than good. What meaning of the word "valid" are you getting at? Did you read my "red head" example -- I think it makes it clear what sense of "valid" is needed to make group comparisons. --Rikurzhen 22:01, September 10, 2005 (UTC)

And I guess because repetition never hurts: to answer the question "is race a valid category" requires one to specificy at least 3 parameters: (i) what definition of race? (e.g. essentialist, populationist, etc.), (ii) what scale (e.g. global, U.S.), and (iii) what domain (e.g., anthropology, race and intelligence). Or as Neil Risch said in the quotation a few sections above, it's mostly a matter of "terminology". --Rikurzhen 23:19, September 10, 2005 (UTC)

So for a little more of me responding to me... there are combinations of those 3 parameters that make the claim "is race a valid category" false. However, those combinations do not impinge upon the science of "race and intelligence" as it is actually practiced, and thus there is no contemporary scholarly controversy on that dimension. As an example, one might argue from the essentialist definition of race that Hispanic is not a race and therefore that data is not about race and intelligence. This argument is not made because the emphasis of the research is on the group differences, not on an a priori definition of race, and whether those groups happen to be called races (or ethnicities or populations) doesn't matter so much as that they are operationally defined groups. To contemplate the hereditarian hypothesis (whether it's true or false), you need to further know that those groups are genetically non-identical -- which is also known. --Rikurzhen 02:40, September 11, 2005 (UTC)

I too think that this article is exceptionally NPOV. If anything I feel that the minority view of race and g-based intelligence as being invalid concepts is given too much weight. Race is held as being a valid concept by the majority of biologists and geneticists. I know this has been gone over ad infinitum already on these talk pages, but "race" is certainly valid for many areas of scientific study. Sickle-cell Anemia, etc., etc. So I really have trouble with what I view to be an excessive debate regarding the validity of the concept of race. Anyway, just my 2 cents, but I feel that the attempts to "NPOV" this are watering down what to me seems to be a very neutral and well-done article. The votes against deletion from the community also seem to suggest that there is not too much NPOV dispute, and suggested cleanups and qualms were properly dealt with. Discuss the fact that race and intelligence are debated concepts, but I certainly don't see any POV problem with the article as it stands now. --Defaulticus Rex

I will be adding from the copious amount of contemporary scientific debate taking issue with the POV in this article (yes, I saw the redhead analogy). As this is a hobby, it will take time. To wit:

The use of the concept of race in pharmacogenomics, forensics, and human molecular genetics continues apace, despite the imprecision of the category and the growing number of voices suggesting caution, and even a "sunset clause" for its continued deployment. The new technologies that can generate SNP patterns and profiles for any population have created an ever growing risk that racial categories will be mistakenly re-inscribed as "genetic." Troy Duster, "Enhanced: Race and Reification in Science." Science, Vol 307, Issue 5712, 1050-1051 , 18 February 2005

The fact of the matter is that scientists taking the long view of the history of science see that this new technology is being put to use in ways previous technologies were used to reify race. This article does not reflect much, if any, of the contemporary scholarly commentary from this POV. This thing as it stands is the gnxp party line, which is a rather specific POV. A huge chunk of the bibliography with which I am familiar is not represented here. I have learned a ton from this exchange, but I do not believe this article warrants removal of the NPOV tag yet, by any stretch. More soon! Jokestress 04:20, 11 September 2005 (UTC)

Note that Duster's argument amounts to do not use "race" rather than race is not valid. The point being drawn along this line is about the word "race" -- about terminology -- not about what exists in the world. Tang et al showed that blind to anything but sequence data, samples of the U.S. population can be split into groups that closely correspond with self-identified "racial labels". Whether we call those groups "races", a point about which there are good arguments in both directions, is truly irrelevant to the current article. (Unless we are supposed to head their concern and literally censor the use of the word race in this article, seemingly a bad idea.) None of this impinges on the fact that groups identified by ancestry have different average IQ scores, and so there's little point in bringing it up in this article. Truly... if there was something important about this topic that wasn't being represented I would feel honor bound to bring it up, but there's really only smoke and no fire to the relationship between the debate over "race" and the research described in this article. --Rikurzhen 04:46, September 11, 2005 (UTC)

I've got an idea. Rather than trying to comb the literature for the strongest possible "race does not exist" statement to juxtapose against the strongest possible "yes it does" statement*, let's try to figure out what can be said unambiguously and balanced against all prominent POVs about race that will properly equip the naive reader to understand the subtlty of the contemporary science of human population structure. I'll try to work on this. (*because they'll be addressing the question with different values for the 3 parameters I mentioned, and thus not really be refuting one another) --Rikurzhen 05:04, September 11, 2005 (UTC)

I have nothing to add by my vote: The NPOV tag is a travesty. This article is the very model of what Wikipedia strives for: accuracy, honesty, and verifiability in reporting viewpoints. (As opposed to arbitrating the Truth.) This statement does not encompass the introduction, which currently claims that the usage of race as a valid categorization (in Soc.Sci.) should be disputed. That's such an extreme minority POV that its inclusion in the introduction violates WP:NPOV, and clearly so. Similarly, the label interdisciplinary in the first line is original, to put it politely. (By that logic, Evolution is interdisciplinary as well, since theologians, computer scientists, philosophers, history-of-science people, and lots other scholars have their things to say about it, some of it even sound.) Let me just re-plug my strong preference for Nectar's introduction from mid-July, possibly with the current bullet point about the reification of intelligence merged into it. (Also, I am not saying the article is finished, or even good, or complete. I am just saying that it doesn't violate WP:NPOV except for what I pointed out.) Arbor 07:45, 12 September 2005 (UTC)

Evolution is an interdisciplinary field, as is this. One group I monitor, HBES, is an excellent example of how interdiscplinary this is. [3] As Ove mentions, science as fetish object is a POV issue with this article. Trying to POV fork anything that isn't a certain kind of scientific POV is a major problem here. The intelligence aspect of this article is much more developed than the race part. I can assure you we are not going back to some version from two months ago. It's going to take a few more months to iron out all the POV problems in this article. Let's get the intro dealt with first and go from there. Jokestress 10:18, 12 September 2005 (UTC)

"first assumption"

trying to write in the form of an "assumption" is difficult. the background information > basic concepts > racial distinctions section gives a good summary of race:

Racial distinctions are most often made on the basis of skin color, facial features, ancestry, and national origin. Some scientists argue that common racial classifications are not meaningful, often on the basis of research indicating that more genetic variation exists within such races than between them. To define terms, racial labels most commonly used in the United States relate to genetic ancestry (Tang et al., 2005).

trying to compress that to one line could be challenging. --Rikurzhen 08:09, September 11, 2005 (UTC)

BTW, are we giving the criticism of 'race' in this summary the fairest wording? It seems like "meaningful" here is being used as a blurry proxy for the more scientific term, "useful." (They're referring to "meaningfulness" in scientific settings, not some other domain, right?)--Nectar ^T

Here's the original Lewontin...

‘‘It is clear that our perception of relatively large differences between human races and subgroups, as compared to the variation within these groups, is indeed a biased perception and that, based on randomly chosen genetic differences, human races and populations are remarkably similar to each other, with the largest part by far of human variation being accounted for by the differences between individuals. Human racial classification is of no social value and is positively destructive of social and human relations. Since such racial classification is now seen to be of virtually no genetic or taxonomic significance either, no justification can be offered for its continuance’’. R.C. Lewontin (1972). Lewontin RC. The apportionment of human diversity. In: Dobzhansky T, Hecht MK, Steere WC, editors. Evolutionary Biology 6. New York: Appleton-Century-Crofts. 1972. p 381–398.

it seems stronger than "not useful" --Rikurzhen 09:13, September 11, 2005 (UTC)

bamshad in jama

the bamshad article in jama gives a great example of the terminology dispute over race. bamshad suggests that race should be replaced with "geographic ancestry".

Race is frequently used by clinicians and biomedical researchers to make inferences about an individual’s ancestry and to predict whether an individual carries specific genetic risk factors that influence health. The extent to which race is useful for making such predictions depends on how well race corresponds with genetic inferences of ancestry, how frequently common diseases in different racial groups are influenced by the same vs different gene variants, and whether such variants have the same effects in different racial groups. New studies of human genetic variation show that while genetic ancestry is highly correlated with geographic ancestry, its correlation with race is modest. Therefore, while data on the correspondence of race, ancestry, and health-related traits are still limited, particularly in minority populations, geographic ancestry and explicit genetic information are alternatives to race that appear to be more accurate predictors of genetic risk factors that influence health. Making accurate ancestry inferences is crucial because common diseases and drug responses are sometimes influenced by gene variants that vary in frequency or differ altogether among racial groups. Thus, operationalizing alternatives to race for clinicians will be an important step toward providing more personalized health care.

...Referring to "geographic ancestry" instead of race is an emerging alternative that is both more accurate and less contentious.

in contrast "geographic ancestry" is probably the best re-definition of race.

so in the end it comes down to a choice between (1) restricting "race" to old definitions and thus finding "race" to be lacking or (2) giving "race" a new definition and thus finding "race" to be useful --Rikurzhen 06:25, September 11, 2005 (UTC)

nectar's question

sort of related: nectar asked "Do any intellectuals argue something besides for IQ should be the foundation of race and intelligence research?"

i updated intelligence (trait) with a description of the distinction between intelligence, IQ and g. so one could say that many intellectuals think g should be the foundation of race and intelligence research, rather than IQ, but of course IQ tests are the primary means of measuring g. --Rikurzhen 20:37, September 10, 2005 (UTC)

Some thoughts on Tang et al. and the NPOV dispute

A few thoughts on the use of science, the Tang et al. article, and the NPOV dispute (pardon the length):

1. It seems to me, being new to WP, that much of the discussion on the talk pages about race (and race and intelligence) regards science as somewhat of a "fetish object". Many statements try to disqualify opinions that are not duly "scientific" or state opinions as true because they are "based" on science. We should keep in mind that there is not only a distinction between scientific (objective?) work and opinion (although I believe that this is much less than is given credit for here on WP) but also between "good" science and "bad" science. Just because a paper is published in a peer-reviewed journal does not make it infallible, and one must wait for future work to fully understand the impact of any single research project. This ties in well with the fact (and it is a fact) that scientific method is itself built around controversy; no one article can or should "prove" anything in science, it can only give weight to a particle position. Consequently, thinking that Tang et al.'s article lays anything to rest is to misunderstand science.

2. Speaking of Tang et al.'s article, I would like to offer two comments: a) The first is to state that the researchers make the assumption that there is a great deal of homogeneity within SIRE (Self-Identified Race/Ethnicity) categories. They write: "An underlying assumption is the relative homogeneity within a single SIRE group. The validity of this assumption must be evaluated." This assumption, which hopefully seems at least somewhat questionable, reveals the amount of subjectivity that is a part of any methodology. b) While the results of the article purportedly support the usefulness of self-identified racial categories in genetic research, they do not necessarily support the validity of essentialist definitions of race (based on phenotypic characteristics like skin colour). It is precisely the difficulty, evident in this and other studies, in assigning all individuals to a “genetic cluster” linked to their phenotype that shows not the arbitrariness but the imprecision of racial categories. Genetic analysis, no more than folk taxonomy, can get beyond the problem of locating racial borders (and without clear-cut borders, there can be no essentialist categories).

3. I agree with Jokestress that the NPOV dispute is not over for this article. The problem lies with the underlying, a priori acceptance of racial categories that is implicit in focusing the article on the culture- only versus partially genetics argument. The controversial nature of race is somewhat masked through this diversion. The following quote from Paul Brodwin is relevant in this regard: “the scholarly (and left-liberal) opposition to “genetic essentialism” is not really a reaction to contemporary genetics, but rather to its reception. The essentializing occurs at the level of popular reconstructions of genetic science, and professional anti-essentialist interventions should be directed there as well.” (Brodwin, P. “Genetics, Identity, and the Anthropology of Essentialism” Anth. Quarterly 75(2), 323-31). --Ove 19:44, 11 September 2005 (UTC)

Ove, I just skimmed your very thoughtful comments, and at first blush I think I mostly agree with everything except your conclusion that there is a NPOV problem with the article. I think the literature clearly supports the claim that the essentialist definition of race does not describe the actual world and that a claim that "race is not valid" where race="essentialist race" is a true statement. However, it is a mistake to equate "self identified race" with "essentialist race", since the former assumes no particular definition of what races (fundamentally) are other than a social label.

The remarkable new finding of Tang et al is not that genetic cluster exist in humans that closely mirror geographical ancestry (this is to be expected given the low level of gene flow in the past and we previously shown by Rosenberg and others), but rather that for the major population groups of the U.S. (white, black, east asian, hispanic) there is such a precise concordance between self-described ancestry and genetically inferred ancestry.

The mistake that I think is being made on this talk page is to assume that the debate over "what is race? / is this definiton of race valid?" impinges on the interpretation of data about "race and intelligence". Now that the space of reasonable opinions about human population structure and "race" has been narrowed down by the collection of data, all of the remaining POVs are compatible with any of the major causal theories about the cause of the IQ gap. --Rikurzhen 20:07, September 11, 2005 (UTC)

first assumption : lewontin was wrong

A first assumption that just comes to mind is this -- Lewontin was wrong. Of course we wouldn't literally write that, but rather something informative about the current understanding of human population structure, self-identified race, and genetics. That seems to be the sticking point that I wasn't cathing on to. --Rikurzhen 20:14, September 11, 2005 (UTC)

Yes, I think that is close. One major issue is (as you note above) conflicting usage of the term. Since lay readers are going to be assuming the lay definition for the most part, it is important to teach the controversy by showing the assumptions behind the definitions. Your suggestion below has the "weasel word" feel of the culture-only vs partially genetic false dichotomy. The assumptions need to be laid out in simple declarative sentences. My first crack at debated assumptions started with the words "race" and "intelligence." No qualifiers, since lay readers need a very stripped-down explanation for this summary. "Self-identified race" is not the debate, as I have said earlier. It is "race" as a term and a taxonomy. In order for lay readers to understand the scientific hypotheses, they need to understand how the terms are being used. I think various editors have done a fantastic job of making this clear regarding the intelligence vs intelligence test issues in this summary, so I am optimistic we can do the same with "race." My suggestion is below. Jokestress 00:23, 12 September 2005 (UTC)

What race is doesn't (usually) matter for social science research. It can be approached from the POV of logical positivism, which is what many researchers acutally do. Thus, they use "self-identified race" rather than "reserach-defined race". Only in the the context of a causal hypothesis, do we need to actually consider what it means to be White, Black, or East Asian. It is an uncontroversial fact that those labels imply both culture and genetics.

The "is race a folk/scientific taxonomy/" question doesn't seem to capture the current understanding or the Lewontin-style objection to race. But what makes it completely unhelpful is that it doesn't matter at all for the interpretation of "race and intelligence". A folk taxonomy can be highly useful and even suggest a genetic explanation, as per the redhead/not-redhead pain example. --Rikurzhen 04:13, September 12, 2005 (UTC)

It's about the RE in SIRE, not the SI. As Ove noted, SIRE has its uses, but that SIRE has its uses is not the debated assumption here (as I have said before, I don't think anyone is debating that). The debated assumption is about the scientific usefulness of "race" when interpreting data on intelligence. Social science really isn't the issue here. The issue is the hereditarian hypothesis that lower IQ test scores represent a genetic component that correlates with race. To make that hypothesis falsifiable, "race" must be genetically defined, and there is a debate about whether "race" is valid or useful in that context. Taking the hair color analogy, if some scientist found that blondes have the lowest group intelligence of the three self-identified hair groups, and then hypothesized that this lower intelligence was linked to their phenotypic and genotypic makeup, you would need to define hair color genetically. That's not about the social reality of having three hair color groups or the social science of intelligence, it's about the genetic definition of hair colors at the fundamental level. I don't know if that's exactly clear, but SIRE isn't the issue, it's the categories from which people are allowed to identify, and whether those rise to the level of scientific versus social categories. Jokestress 07:00, 12 September 2005 (UTC)

these seem to be the key points in what you wrote:

• The debated assumption is about the scientific usefulness of "race" when interpreting data on intelligence. -- good
• To make that hypothesis falsifiable, "race" must be genetically defined, and there is a debate about whether "race" is valid or useful in that context. -- good
• whether those rise to the level of scientific versus social categories -- not as good

also note that this is only about the causal hypothesis, not the simple act of asking the question of whether races differ in IQ -- on this we seem to agree

it is fairly obvious that it is a prerequiste to hypothesizing a genetic explanation for group differences that these groups are not genetically identical. this is a necessary condition. but it is also a sufficient condition (in most situations). So the assumption could be something as simple as: racial/ethnic groups are not genetically identical.

this is known to be true w/o a doubt. for example: "Racial and ethnic categories are proxies for a wide range of factors, potentially genetic and nongenetic" (Mountain and Risch, 2004)

the part about rising to the level of a scientific cateogory versus being a social category doesn't seem to capture any of the relevant details of this (AFAIK) --Rikurzhen 07:42, September 12, 2005 (UTC)

[To summarize:] The "scientific versus folk taxonony" question is not informative about what is actually at sake in proposing a hereditarian hypothesis of race differences. --Rikurzhen 21:51, September 12, 2005 (UTC)

Write suggestions here:

• human populations have internal structure stemming from geographic and genetic ancestry, which is at least modestly concordant with the social labels of self-identified race/ethnicity (Rikurzhen)

• race is a scientific classification, rather than a folk taxonomy. (Jokestress)

• [In sentence form, to replace the list: Rikurzhen] This research is grounded in several controversial assumptions. Intelligence is measurable (see psychometrics) and is dominated by a unitary general cognitive ability. The social categories of race and ethnicity are at least modestly concordant with human population structure based on geographic and genetic ancestry.

The problem with sentence form is that it loses the dependence on the first part (This research is grounded in several controversial assumptions) and recasts the two dependent clauses as declarative sentences, thus reinforcing the POV is question. The bullet points help casual readers understand the assumptions behind 1) "race" and 2) "intelligence" required for this research to commence. That's why I recommend the bullet points for clarity.

[Sorry I missed this paragraph] -- not a big issue for me. --Rikurzhen 03:19, September 13, 2005 (UTC)

Your rewrite of assumption 2 is interesting. If we lose are at least modestly concordant and come up with a more accurate and less qualified phrase in its place, we will be getting close. All of this is leading up to the grandaddy of POV issues: the suggestion (or more generously, hypothesis) that certain groups are genetically predisposed to be less intelligent. Jokestress 02:40, 13 September 2005 (UTC)

"At least modestly" is a minimum assesment of the concordance between racial groups and geographic ancestry groups. For example, Tang et al found >99% concordance. But "at least modestly" concordant is sufficient for race or ethnicity to be "proxies for a wide range of factors, potentially genetic and nongenetic" (Mountain and Risch, 2004). At this point, there is no doubt that SIRE groups are at least modestly concordant with genetic/geographic groups. There is little (no?) debate on this point, which is fortunate for our work on this article. (Of course there are things about which there is disagreement, but these are not the kind of things that the "race and intelligence" article needs to worry about, but rather are good topics for the "race" article.)

.... mmm, one caveat ... this is about the US population. worldwide, race is only modestly concordant with genes/ancestry because of clinal variation in allele frequencies. --Rikurzhen 04:15, September 13, 2005 (UTC)

grandaddy of POV issues: the suggestion (or more generously, hypothesis) that certain groups are genetically predisposed to be less intelligent -- huh??? -- surely you don't doubt that this is a major hypothesis? --Rikurzhen 03:16, September 13, 2005 (UTC)

The article implies that this hypothesis (or elements supporting it) is "settled" or "put to rest," reflecting editors' beliefs, but not the ongoing debate in scientific circles. It is of course a perfectly valid scientific hypothesis, but the issue is the evidence used to support it and the assumptions required to falsify the hypothesis. This article as it stands has a lot of confirmation bias, as I have noted before. Editors have favored evidence supporting the hereditarian hypothesis and tried to dismiss or POV fork any criticism of the research they cite. Hereditarians produce knowledge produced that supports this hypothesis, but critics do not engage in this research in the first place because they feel much of the race and intelligence research is fundamentally flawed in its methodology. Jokestress 04:59, 13 September 2005 (UTC)

re: settled -- i'm incredulous -- the causal hypotheses section lists two major hypotheses and the views about each are described in detail. you seem to have previously recognized that this is the major unanswered question.

re: confirmation bias -- I'm afraid that claim doesn't stand up to the burden of proof that WP demands. That's essentially a criticism of the research. We can't make up our own criticisms: WP:NOR.

re: fundamentally flawed in its methodology -- again a criticsm of the research, but you won't find that coming from the scholars who are in the position to make that evaluation. for example, the APA report does not reflect this position. i see no literature to support this claim that isn't already described in the article (e.g. Gould). frankly, i'm curious what would make you come to this conclusion. --Rikurzhen 05:42, September 13, 2005 (UTC)

Let's take this thread to the section I'm going to start below. --Rikurzhen 05:52, September 13, 2005 (UTC)

Alternatively, we could crib off of Mountain and Risch. How 'bout this: "The social categories of race and ethnicity are proxies for a range of factors, genetic and nongenetic." This acheives the same end I think. --Rikurzhen 03:22, September 13, 2005 (UTC)

Hmm, that's interesting (and citable in terms of NOR), but The issue seems to me to be that both race and intelligence are social constructs being reified through science. The intelligence one as it stands does and excellent job of explaining the issue, but maybe that's because more people seem close to agreement on what "intelligence" is (at least in terms of psychometrics). I read an interesting article the other day about how Native American tribes are being confronted with DNA evidence from people claiming to be part of a tribe [4]. They always used the "fraction" idea (half, quarter, etc.) that was common use for other types of American "miscegenation" (quadroon, octaroon). Is race a metric? Jokestress 04:59, 13 September 2005 (UTC)

social constructs being reified through science -- i'm not sure about the use of reified there, but i get the idea. but how is it inappropriate to use science to analyze a concept that sprung from a prescientific/social context? for example, there's a robust folk psychology that is the subject of scientific investigation.

Is race a metric? -- i.e. quantitative/fractional. ancestry certainly is, so that question will depend on whether you are picking a definition of race and then comparing it to reality or whether you have an example of what people mean by race and then try to find a definition for that. the former (e.g., the 19th century essentialist defintion) implies that race is not factional (how can an essential quality be had fractionally?). the latter implies that whatever race is (e.g. geographic ancestry) it must be fractional.

these are all interesting questions for race (which I helped write and I hope it covers that idea adequately), but a bit off topic for race and intelligence.

I think we should go with the "proxy for genetic and nongenetic factors" idea in the meantime. --Rikurzhen 05:30, September 13, 2005 (UTC)

List vs sentence form?

Rikhurzen, what was factually accurate about this change? [5]. I think the caveats should be inside each bullet pointed sentence and it should be in paragraph form. zen master T 20:56, 12 September 2005 (UTC)

I agree that moving to sentence form is probably better for that material, but at the moment we're still trying to work out what exactly we're trying to warn people about (see above). The change from "controversial" to "disputed" was enough to make the claims false, where they were merely straining credibility before. --Rikurzhen 21:10, September 12, 2005 (UTC)

Assumptions by definition do not having consensus so are therefore disputed, right? I actually didn't realize you and Jokestress were discussing this, sorry. As long as it gets fixed soon I will be ok. zen master T 21:16, 12 September 2005 (UTC)

NP. One would think that a controversial assumption was disputed, but presently there is no one doing the disputing. So what we have is the shadow of a prior dispute, where it's not yet clear that the controversy is gone (espeically to the public). This is due in part on the poor flow of discussion from experts to the public. --Rikurzhen 21:22, September 12, 2005 (UTC)

Or put another way, everyone is very careful about these ideas yet there is no obvious on-going dispute (i.e. debate). That's why we're still looking for what exactly is being assumped and to what extent that is controversial. --Rikurzhen 21:26, September 12, 2005 (UTC)

I think you aren't seeing the debating and the disputing, IQ tests and what they mean (if anything) are highly disputed. Racial catagorization is also highly disputed, especially by geneticists (who should be considered authorities on the subject). The USA was founded on the principle of treating everyone as an individual so I question intelligence researchers' true motivations. zen master T 21:35, 12 September 2005 (UTC)

Okay... check the discussion above. [To summarize: there are no contemporary parties in the race/genetics who are actually making contradictory statements. Mostly, they are just using definitions of the terms in order to emphasize different aspects of reality. Unforuntely, they're taking their time about actually listening to what each other is saying as to arrive at a consensus. Bamshad's essay in the latest issue of JAMA comes the closest I've ever seen to briding the terminological divide that is causing the appearance of dispute.] --Rikurzhen 21:40, September 12, 2005 (UTC)

The article currently does not reflect any consensus and is rather one sided, isn't it? Maybe I misunderstand you, but I meant be aware of the debating and disputing coming from critical sources... zen master T 21:53, 12 September 2005 (UTC)

Sure... I'm referring specifically to the debate within genetics and medicine wrt using "race" in biomedicine (e.g. the JAMA article). To the extent that that issue impinges on "race and intelligence" there is a general concensus about the science. On (other) issues that do not impinge on "race and intelligence", such as health/research policy, there is lots of the normal kind of scientific and ethical dispute. --Rikurzhen 22:15, September 12, 2005 (UTC)

Shouldn't the intro section of the article be more generic than just that specific issue though? More generically, significant dispute exists over the usefulness of IQ test results and racial categorizations, right? Genetic scientists dispute the findings of intelligence researchers and the genetic scientists are the ones that are considered the experts in genetics, right? zen master T 00:18, 13 September 2005 (UTC)

You're thinking of Lewontin and Gould? --Rikurzhen 00:26, September 13, 2005 (UTC)

The final paragraph of the intro discusses the public controversy in both broad and specific terms: The findings of this field are often thought to conflict with fundamental social philosophies, and have thus engendered a large controversy. Public debates in this area often contain misconceptions. Critics examine the fairness and validity of cognitive testing and racial categorization, as well as the reliability of the studies and the motives of the authors, on both sides. Critics often fear the misuse of the research, question its utility, feel that comparing the intelligence of racial groups is itself unethical, or fear sociopolitical ramifications, whether justified or unjustified. For instance, the disparity in average IQ among racial groups is sometimes mistaken for the idea that all members of one race are more intelligent than all members of another, or that ranking group IQ averages from "high" to "low" implies a moral ranking of races from "good" to "bad" or an overall ranking of "superior" to "inferior". The conclusion that some racial groups have lower intelligence, and the hypothesis that a genetic component may be involved, have led to heated academic debates that have spilled over into the public sphere. --Rikurzhen 00:30, September 13, 2005 (UTC)

No, I am thinking in general terms of all the critical sources that have disputed intelligence research in various ways, one of the most noticable is the criticism from genetic scientists. The article currently does not include or misrepresents valid complaints that critics have, but with every editor's help I am convinced we can fix it. Specifically, caveats should be included where they are needed within each sentence, rather than confusing the presentation of a complex and controversial subject with a list. zen master T 00:36, 13 September 2005 (UTC)

Okay ZM. You know -- I am a genetic scientist. If you have specific citations in mind, you could list them here. --Rikurzhen 00:54, September 13, 2005 (UTC)

Okay, what do you think of genome scientists coming out against intelligence research? I believe they specifically stated there isn't a clear cut genetic basis for appearance based racial categorizations, that was cited on this talk page previously. Separately, what about the need to apply caveats where they make the most sense? The goal with a neutral encyclopedia is to present both/all sides of a controversial issue, instead of allowing proponents to rewrite and lessen the other side(s)' criticisms, right? zen master T 01:12, 13 September 2005 (UTC)

If you don't mean Richard Lewontin or Stephen Jay Gould or maybe Leon Kamin (circa 1970s), then I'm not sure I know whom you're referring to. --Rikurzhen 01:45, September 13, 2005 (UTC)

Or maybe Steven Rose? --Rikurzhen 01:48, September 13, 2005 (UTC)

ah... do you mean this? [notice i posted it; good interview] --Rikurzhen 01:57, September 13, 2005 (UTC)

from a PLoS Genetics interview with Neil Risch[6]

Gitschier: Let's talk about the former, the genetic basis of race. As you know, I went to a session for the press at the ASHG [American Society for Human Genetics] meeting in Toronto, and the first words out of the mouth of the first speaker were “Genome variation research does not support the existence of human races.”

Risch: What is your definition of races? If you define it a certain way, maybe that's a valid statement. There is obviously still disagreement.

Gitschier: But how can there still be disagreement?

Risch: Scientists always disagree! A lot of the problem is terminology. I'm not even sure what race means, people use it in many different ways.

The take home message is that the interpretion of questions about race is highly contingent on the precise meaning and context of the question. The normal condition of "Scientists always disagree" is blown up to massive proportions by all of the political entanglements of how this data is interpreted.

I was actually refering to older citations. But given the disagreements and the entanglements it wouldn't be fair to allow a one sided presentation of this subject, right? zen master T 02:43, 13 September 2005 (UTC)

That could be a straight-forward interpretation of WP:NPOV, but of course that is not all WP:NPOV has to say. WP:NOR is equally important, so I'm cautious that in translating what has been have written about this subject into the encyclopedia no errors or misunderstandings are introduced. Recognizing the various meanings of the word "race" is an example of a potential source of misunderstanding the literature, which we've recently been discussing. --Rikurzhen 03:01, September 13, 2005 (UTC)

"what is race" and more on "Lewontin's fallacy"

the question "what is race" can be answer in two ways:

1. it can be answered as an analytical philosopher would answer the question -- by taking a definition (probably the essentialist definition), and then examining that definition against reason and evidence
2. it can be answered as a scientist would be inclined to answer the question -- by taking the SIRE labels and examining what they actually delineate and using that data to formulate a definition of race

both are legitimate ways to answer the question, but it a mistake to confuse an answer to one for an answer to the other.

• the most common answer to #1 is -- "race" so defined is a bad category; or if you like "race" does not exist
• the most common answer to #2 is -- "race" is a label for ancestry at a continental/geographic level.

for the question of whether to continue using the term "race" in research, question #1 is important (but maybe #2 also). for the question of explaining race to a naive public, question #1 is important. for the question of whether a racial difference in phenotype is due to genetic difference, question #2 is important.

for the article on race both questions are important. for this article, question #2 is important and question #1 is not. of course, we should have a short warning/pointer to race. we already have a few of those scattered around.

While most contemporary researchers writing specifically about race are generally careful not to confuse #1 and #2, this was not the case in the 1970s. Moreover, in the 1970s the empirical answer to #2 was not 100% clear as we didn't have large scale genome sequencing. In that context Lewontin made a logical mistake that lead him to the conclusion that the answer to #2 was that race was not corcordant with genetics. That mistake was widely propagagted, but it is unequivocally a mistake: we now recognize the error in his logic and have empirical data telling us it must have been a mistake anyway. Yet in 2001 that mistake was still being repeated by the editors of Nature and in 2005 Sternberg continues to repeat that mistake. (Though note you won't find people arguging de novo that Lewontin was right.) Nonetheless, we quoted Sternberg and give Lewontin's (not by name) argument w/o even mentioning that it is a fallacy. I don't see how we can do much more to present all POVs w/o simply "giving equal time" (which is forbidden by WP:NPOV).

The point of mentioning Lewontin is that while Lewontin's claim is still being repeated 30+ years later, this isn't because there is competing data out there that that bolsters his claim. That is, don't confuse the persistence of "Lewontin's fallacy" with the persistence of a research program which supports it.

So to summarize, (1) there are lots of negative things to say about race, but those aren't the kind of things that make much of a difference when interpreting causal hypotheses about the IQ gap in the U.S. and (2) the persistance of Lewontin's claim about genetic distance (actually FST) and racial groups should not be mistaken for a living debate about the genetic differentiation of racial groups -- rather it's a product of people outside genetics not noticing that Lewontin's 1972 (!) conclusion has not held up.

--Rikurzhen 06:41, September 13, 2005 (UTC)

Most of the scientists who criticize race and intelligence research specialize in analyzing the data of others for flaws (like the Cyril Burt criticism some time back). I need to put up a couple of entries for prominent anti-adaptationist scientists who don't even have WP entries yet. In the meantime, I am liking your revision of the first debated assumption more each time I read it, though "factors" may be too general for a lay reader-- can we think about alternatives that are more specific? Jokestress 10:40, 13 September 2005 (UTC)

Rikurzhen, although I agree that there is a semantic problem within the use of the term "race", I see a few problems with the way you present the debate.

1. The term "race" is indelibly intertwined with certain historical, social, and material connotations that cannot be removed simply by saying some “scientists” understand it in a novel way (eg. As lineages as opposed to essentialist taxonomic categories), and since this article will mainly be read by laypeople, you will never overcome this problem with explanation about what “race” really means (to you, to scientists, or whomever).

2. Many scientists, including geneticists, would refrain from using “race” in the way you describe in point 2 (perhaps for exactly the same reason I just mentioned) and would even go so far as to not accept the ancestral or geographic importance of the concept of “race” (for some examples, please see this issue of nature, which I have linked previously). (and I would like to remind you that one of the main criticisms of research such as Tang et al. is that they make a priori use of existing social constructs such as SIRE categories – remember that they state as one of their assumptions that there is homogeneity within geographic (read racial) categories – to think that they simply look at some objective evidence and come to a novel conclusion is incorrect)

3. Consequently, this article cannot be divorced from a) debates about the validity of racial categories among scientists and b) the understanding of “race” by the majority of the readers. The first way in which race can be defined, as you describe, will always be an intrinsic part of this article and therefore should be addressed directly and not swept under the rug. --ove 17:08, 13 September 2005 (UTC)

Yes, as Ove said, my concern is teaching the controversy to the casual reader, who will be almost every reader of this article. As in the academic debate, the terminology issues must be explained and reconciled to avoid misunderstanding. The intro needs the most succinct expression of the issues about race that we can come up with, and then we need to get into those issues in summary style here, with links back to the main race article. I am going to spend a week or so adding entries for the major philosophy of science figures critical of this sort of research (I did a quick one of Philip Kitcher this morning), and do some more work on the people who signed the collective statement by Linda Gottfredson. I think I have about half of them done, focusing on the more influential psychologists who signed "Mainstream Science on Intelligence." Race has a lot of unresolved issues as it pertains to this article. Jokestress 17:41, 13 September 2005 (UTC)

Jokestress, be forewarned that slrubinstein and I more or less exausted the well of major arguments about race as we wrote the race article. From our literature review, almost all of the negative things to say are question #1 things or about worldwide genetic/geographic ancestry (i.e. Asia is a mess w/ lots of admixture). Neither of these things tell you much about this article. --Rikurzhen 17:54, 13 September 2005 (UTC)

[edit conflict]

ove:

swept under the rug -- not the aim of my argument; as i pointed out the race is invalid argument is made clearly in this article (in several points in the lead sections) and without mentioning the rebuttal that I cited. the aim of my argument was to explain to the editors why they see this argument so frequently, but why there is not much more to say about it than to simply state it. any attempt to expound on it will necessarily start copying material from the race article which will very much crowd out this article and seemingly violate the making assumptions but leave a pointer rule from WP:NPOV. the example associated with this rule is that you don't repeat in full a favored anti-evolution argument in every article about evolution, but you can if it is apporpriate, leave a short message saying that some people don't believe in evolution and see creationsim.

the point is that this article need to not try to expound of Lewontin's thesis any futher than to explain it's point and scope. to do so would not be fruitful. that's a burden that race has to carry. --Rikurzhen 17:48, 13 September 2005 (UTC)

jokestress (2x above):

"factor" -- sure, we can ponder that. "factor" is being used because it is a term of art for biology/statistics (wrt design of experiments).

also -- be careful about the age of any particular reference. it may turn out to be anachonristic. basically, anything written before 1996 is (99% likely) going to be out of date. --Rikurzhen 17:48, 13 September 2005 (UTC)

In support of my comment "swept under the run" I offer a few points of evidence.

1. In the sections of the article leading up to, say, "History" (which may very well be all that is read by the casual reader), contains numerous examples of unquestioned use of the category "race" - often used as "racial/ethnic group". Only two statements go beyond this acceptance: the initial statement about assumptions "the social categories of race and ethnicity are proxies for genetic and nongenetic factors" which, lets be honest, most readers will not understand, and the bit about "Some scientists argue that common racial classifications are not meaningful", which is immediately followed by "racial labels most commonly used in the United States relate to genetic ancestry".

2. Related to my first point is the fact that you are somewhat incorrect in your assessment of the current state of the literature. After all a few minutes search on EBSCO, I turned up a number of articles that do not only argue against the genetic position of race and intelligence but against the validity of "race" itself or the reality of "racial" difference in IQ scores. Please see: - Intelligence, Race, and Genetics. Sternberg, Robert J.; Grigorenko, Elena L.; Kidd, Kenneth K.; American Psychologist. Vol. 60(1) January 2005. pp. 46-59. - Genes, Race, and Psychology in the Genome Era : An Introduction. Anderson, Norman B.; Nickerson, Kim J.; American Psychologist. Vol. 60(1) January 2005. pp. 5-8. - THE CULTURAL MALLEABILITY OF INTELLIGENCE AND ITS IMPACT ON THE RACIAL/ETHNIC HIERARCHY. Suzuki, Lisa; Aronson, Joshua; Psychology, Public Policy, and Law. Vol. 11(2) June 2005. pp. 320-327. - There are no public-policy implicatons: A reply to Rushton and Jensen (2005). Sternberg, Robert J.; Psychology, Public Policy, and Law, Vol 11(2), Jun 2005. pp. 295-301.

3. (copy point three from above)Consequently, this article cannot be divorced from a) debates about the validity of racial categories among scientists and b) the understanding of “race” by the majority of the readers. The first way in which race can be defined, as you describe, will always be an intrinsic part of this article and therefore should be addressed directly and not swept under the rug.--ove 23:22, 13 September 2005 (UTC)

Ove, you will find in the history of this article that Sternberg (2005) was once quoted (the abstract in full) and that I was the editor who placed that text in the article. So please understand that I have no personal goal of sweeping things under any rugs. ;) At present, that quotation has moved into the public controversy article and we've never gotten around to making a robust summary of that article as per Wikipedia:Summary Style. However, on that point you'll find that argument made by Sternberg is essentially the same as the argument made by Lewontin in 1972 and that should we do anything more than describe that argument we would have to also detail how it is wrong. At that point, we're essentially re-writing the race article. This is discouraged and a way to handle it is given in the WP:NPOV policy. We should make the existence of controversy about the (for lack of a better word) validity of "race" as it is variously understood, a sentence summary of the prominent objection ("Some scientists argue that common racial classifications are not meaningful..." ... genetic difference within vs between) and leave it at that.

You're claim that you can't understand "race and intelligence" without fully understanding "race" is, IMHO, incorrect. Everyone (Americans at least) understands race well enough to talk about it most of the time, for example the demographic breakdown of the Hurricane Katrina disaster. So clearly it is because we are talking about the very sensitive topic of intelligence and the possibility of genetic causes that people are inclined to bring such hightened scrutiny. However, as WP editors we are bound to not be advocates for one position over another. If we were to fill the entire intro of the article with doubt-filled statements about race merely for the reason that we think people should be very careful about "race", then we would be favoring one opinion over another (then what good would it be to have a race and intelligence article separate from a race article?). Just as it would be wrong if we were to fill the article on evolution with warnings about the tentativeness of scientific theories. This would be undue advocacy, but moreover it would drown out the didactic mission of an encyclopedia. So instead, we must trust that the interested reading will heed a plain warning that there is more to race than common sense would entail and that they can learn about this from reading the race article. --Rikurzhen 01:37, 14 September 2005 (UTC)

Is there any study about IQ gap between people self proclaiming as stupid vs. vs. people self proclaiming as intelligent ? Ericd 23:32, 13 September 2005 (UTC)

While there are articles on people who are incompetent und unaware of it [7] and data showing that those who sign up to take IQ tests are not above average in IQ as a group, I have not seen much on self-identified intelligence and IQ. This would require a stronger social construct of rigid identities one could have (moron, imbecile, idiot used to be names for groups used in the Binet-Simon scale and based on IQ test scores). As you might imagine, no political identites ever coalesced around these names. On the other end of the spectrum are a lot of people who fetishize IQ scores and come up with all kinds of striations to distinguish themselves from others who revel in comparing IQ scores as a form of competitive ranking. Jokestress 00:46, 14 September 2005 (UTC)

Probably not. Intution tells me that most people don't self-proclaim to be stupid. There are however countless studies correlating IQ with behaviors and outcomes judged require intelligence (see IQ). --Rikurzhen 01:37, 14 September 2005 (UTC)

Rikurzhen, I was wondering if you have some sources on Lewontin's fallacy. The only thing I could find was the Edwards' article (which itself is only cited 8 times in other's work). Otherwise, it seems to me that some scholars (see that Nature supplement I cited previously)still use Lewontin's basic argument. I think it is it is a fairly strong use of language to designate something as fallacy without more backing. This is a similar complaint to my concern with the overuse of the Tang et al. article on WP (ie. that a single source is used to convey the impression of a dominant argument within a field).--ove 17:19, 21 September 2005 (UTC)

It doesn't count as an academic reference, but there is of course a chapter/section in Richard Dawkin's latest book The Ancestor's Tale: A Pilgrimage to the Dawn of Life which is a strong endorsement of Edward's argument in the public debate. (Or course, Dawkins is hardly a neutral commentator of viewpoints espoused by the Lewontin.) Mathematically, there isn't much to say about Edwards' paper: he points to a very trivial mistake in Lewontin's argument. It's something I would fail my undergraduate student of probability theory for (basically, assuming independence of highly correlated random variables). So the maths is beyond dispute, and hardly worth quoting. But the maths is of course not the central thing to disagree about. If you (or Lewontin) require races to be separable by a single locus (or a single trait in the phenotype, like "curly hair") then obviously races have no taxonomic value. If you don't (i.e., if you allow several loci, or several observables, like curly hair and skin colour and blood type XYZ and lactose tolerance and whatnot) then Lewontin is wrong, and laughably so. It's a question about the semantics of the term race, not a question about maths.

By the way, if you have a better suggestion for what we should call Lewontin's line of reasoning, please tell me. I am thinking of something like Lewontin's argument. I would very much like to expand the article at Lewontin's Fallacy, but like you I am somewhat put off by the title. However, we cannot invent neologisms at Wikipedia, so I am adverse to coining the phrase Lewontin's argument. Suggestions are welcome. Arbor 19:23, 21 September 2005 (UTC)

Arbor's description is quite accurate -- Lewontin's mathematically mistake was trivial but widely repeated, but the deeper problem was the way he thought of race as hinging on a single variable, when it is most certainly multidimensional. WRT the math, the distinction to be understood is that the magnitude of the differences between races is not especially revealing (because it is non-zero) about our ability to differentiate races. The new material I added to the race article adds (IMHO) much to the subtlty needed to fully understand race (although you don't need that much subtlty to understand this article.) --Rikurzhen 21:03, 21 September 2005 (UTC)

Isn't one of Lewontin's points semantic -- that most people (NOT physical anthropologists or evolutionary biologists) try to reduce definitions of race to one trait or a small handful of traits? Slrubenstein | Talk 22:14, 21 September 2005 (UTC)

It's been a long time since I've seen Lewontin's 1972 paper, but I'm not sure that's what he was saying. Could have been. Do you have a copy? --Rikurzhen 05:39, 22 September 2005 (UTC)

Err...what ability to differentiate races? I was not aware than anyone had ever done such a study. Guettarda 22:22, 21 September 2005 (UTC)

(still slow here in wikiland) Bamshad and Graves need to be discussed as responses to Edwards' singling out Lewontin. When this thing isn't so sluggish, I'll add them here and on the entry for Lewontin's fallacy. Jokestress 23:10, 21 September 2005 (UTC)

I assume you're talking about the race article -- we don't mention Edwards in this article? this is a nice review article too --Rikurzhen 06:39, 22 September 2005 (UTC)

Seriously, has anyone ever been able to distinguish race without first pre-defining race? Has anyone ever found a true transition zone between races (i.e., a zone in which "racially associated characters" changed more rapidly than they did on either side of the transition zone)? Distinguishing "race" after defining "race", (for example, by picking distinct populations) is circular. Looking for differences in "race" in America need to be controlled for differences in the distance between the source populations, otherwise you are conflating race with distance-related factors. If race is based on self-idenification, then it is a social construct. You could say that there are differences in IQ (and, if you so choose, in intelligence) based onself-identified social grouping which at their core have different geographical origins is valid - saying that there are differences in intelligence based on "race" as a biological entity are not, since (AFAIK) race has never been demonstrated to be biologically real. People can say "I can see it, so it must be real", but can they? Has anyone ever sampled a "transitional zone" and determined whether there is agreement between what people call one race and what people call another? I'd love to see the results of asking Americans to place people from Kazakhstan into "races". Guettarda 13:07, 22 September 2005 (UTC)

Read the race article. If it is not clear from that article, then we have some work to do there. You're questions are based on a certain definition of "race", which you'll find spelled out there. --Rikurzhen 14:39, 22 September 2005 (UTC)

Perhaps this will help -- one take home message from a talk I heard recently -- race describes ancestry but ancestry is more subtle than race. --Rikurzhen 21:23, 22 September 2005 (UTC)

Wiki is not a place to come for serious information

To be honest, I can not understand why there is an informational website that can be altered, and subjected to change by any individual on this planet? Who's bright idea was this???

I have come to this site on many occasions, with the sole intention of leaving the dumbest contribution I could possibly think up. Usually my posts will include words like "poo-poo, and pee-pee," but nobody seems to notice. This site is absolutely rediculous, and a concentrated effort should be made to close it down.

I have also heard that many racists will use this site as their own personal reference to others; I think this is because racists are free to change and munipulate details as they please. An overwhemling amount of false information can be found via this website, and something needs to be done to put an end to this. —Preceding unsigned comment added by 70.68.168.189 (talk • contribs) September 19, 2005

Most of your graffiti was probably undone by others within minutes. The same arguments that show that Wikipedia should be rubbish apply to the internet in general, to newspapers, and even to "traditional" encyclopedias. Despite the theory that Wikipedia should be rubbish, the reality is that it's actually good, or at least better than most alternatives. If you actually have an error to report on this article, such as an incorrect reference or unfair summary of some scientific research, I'd like to hear it. If you just want to rant about Wikipedia, read this article about Wikipedia itself, especially the Wikipedia#Further reading section.

I don't care whether this article is read by racists, rapists, paedophiles, terrorists or homophobes. It's either a well written, accurate, neutral article, or it isn't. I happen to think it is a good article. Racists can make their own websites, and publish their own newspapers or buy their own TV networks. You'd be better off calling to shut down many TV networks before attacking Wikipedia.

Your questions, and my responses, should probably be deleted from this thread sooner or later. They are irrelevant to this article. --Aaron McDaid 11:26, 21 September 2005 (UTC)

I find the standard for citations to be serious information. Any individual could use whatever tool is available for information gathering to find primary sources, but the listings here (provided you actually consider reading them) are excellent.

Proxy and nongenetic factors

OK, WP is still slow, but I want to get back to this despite the sluggish interface.

Assumption one as it stands: the social categories of race and ethnicity are proxies for genetic and nongenetic factors.

This is close, but I am not sure "proxy" is the best word for a lay reader. "Substitutes" or "equivalents" seem a little closer but not entirely satisfactory.

Further, I am not sure "non-genetic factors" are debated as assumptions. It seems the assumption (and its controversy) hinge on the argument that social categories of race and ethnicity are based on (or equivalent to) genetic categories. Thoughts? Jokestress 19:04, 28 September 2005 (UTC)

"are proxies for" --> "are indicators of" ?

"genetic and non-genetic" -- the "and" is key; not sure if it would be clear without mentioning both

I don't think we're going to come up with a better bullet point (give it a try); the race article is now >100k, so obviously we can't summarize all that in one sentence --Rikurzhen 22:46, 28 September 2005 (UTC)

How about this:

the social categories of race and ethnicity correspond with genetic categories.

What I am trying to do is get a parallel construction-- the crux of the debated assumption seems to be that social and genetic categories match up. Again, I am not sure we need to mention non-genetic factors, since this statement is not disputed to my knowledge: "the social categories of race and ethnicity are proxies for non-genetic factors." Jokestress 00:35, 30 September 2005 (UTC)

the social categories of race and ethnicity are concordant with genetic categories, such as biogeographic ancestry ?

That's not bad. "Concordant" is better. My next question would be, are there established categories of biogeographic ancestry? I have seen more than four mitochondrial categories used for humans, and they didn't match up exactly as White, Black, Asian, and Hispanic. Jokestress 01:07, 30 September 2005 (UTC)

I'm not sure that there is an established definition of a "genetic category", but I take it to mean clusters inferred from genotypic data, which could be single locus (like the mtDNA haplotypes you mention) or multilocus (e.g. a whole-genome SNP panel). "Biogeographic ancestry", where biogeographic means the geographic distribution of organisms, would be genetic ancestry labels like "Sub-Saharan African", "Western European", "Central American". So the assumption is that racial labels are concordant with these ancestry labels, which constitute a kind of genetic categorization. --Rikurzhen 01:20, 30 September 2005 (UTC)

An FAQ [8], including "What is Race?" and "What is BioGeographical Ancestry (BGA)?" --Rikurzhen 01:29, 30 September 2005 (UTC)

Yes, it seems BGA is generally formulated on a continental (or sometimes subcontinental) level. So are social categories of ethnicity concordant with BGA? Europe/Africa/Asia all make sense, but it seems that Hispanic is a blend of several BGAs and is based on language/culture rather than genetics. You could have a Black Hispanic and a White Hispanic. I suppose if a genetic test could distinguish a Shiite from a Sunni or a Hutu from a Tutsi, there would be concordance for ethnicity, but I am not aware of info about that. Jokestress 02:10, 30 September 2005 (UTC)

People who self-identify as "Hispanic" are generally distinguishable from people who identify as "White" by genetic markers (this was done in Tang 2005). With enough markers it would be possible to break the Hispanic group down into subgroups like Mexican, Cuban, Puerto Rican, etc. As long as there has been some mating isolation for some time, then it should be possible to use genetic markers to distinguish ethnic groups, with varying levels of success depending on the extent of admixture. I would imagine that English, Welsh, Scottish, and Irish are not easy ethnic categories to delineate among Whites in the U.S. --Rikurzhen 02:21, 30 September 2005 (UTC)

looking for useful wiki links i came across this page -- Maps of American ancestries -- perhaps you can distinguish english, welsh and scotts-irish. --Rikurzhen 05:34, 30 September 2005 (UTC)

Chart

Can someone with the data put a y- axis label on the chart? I'd also like to note in the footnote that the chart does not have the same N value for each group, since I don't think that's clear to a casual reader and (slightly) affects the margin of error on each. I don't think that needs to go in the caption. Jokestress 00:27, 30 September 2005 (UTC)

The values of the Y-axis won't make any sense. The area under each curve is 1, and the max of each curve is around 0.4 --Rikurzhen 00:33, 30 September 2005 (UTC)

The N for each group??? Are you thinking of the normalization data? Because the Asian and Hispanic curves don't come from that dataset. Also, Roth 2001 has N > 1e6 for the Black-White gap, so there's no reason to mention the N value from a particular study. --Rikurzhen 00:50, 30 September 2005 (UTC)

Interdisciplinary field

The first sentence currently describes race and intelligence research as an interdisciplinary field, and previously described it as an area of intelligence research. Interdisciplinary field seems less helpful in helping the reader categorize this area. Are we referring to other fields in the social sciences? As long as the area is primarily within intelligence research or psychometrics, it seems fine to leave it at that in the header. --Nectar ^T 06:10, 30 September 2005 (UTC)

For a while this said race and intelligence was an area of social science research, but since so much of this article discusses genetics, biology, and statistical models as well, it seemed more accurate to say interdisciplinary. See also my comment to Arbor earlier about how evolution is an interdisciplinary field in the same way as this one. Granted, most of the people doing the work are psychologists, but the field seems to be shifting rapidly as new technologies bring in people from other disciplines. Jokestress 07:16, 30 September 2005 (UTC)

Canadian news on Rushton, race, IQ and genetics

Canadian news on Rushton, race, IQ and genetics w/ nice summary of recent events -- could be helpful here or related article --Rikurzhen 01:16, 2 October 2005 (UTC)

We should also add the news out of Wake County, North Carolina last week that standardized test performance of Black and Latino students has dramatically improved in the ten years since the implementation of a comprehensive socioeconomic desegregation program. [9] Jokestress 02:17, 3 October 2005 (UTC)

I don't have access, but that sounds like good achievement gap (and related sections) material. --Rikurzhen 02:29, 3 October 2005 (UTC)

It would be interesting to see the data on the scores of the higher performing segments of students to see if there was an adverse effect or no effect there. (One official states briefly in the article there was no effect).

Working log in information, courtesy of bugmenot:

• janet.robinson@sogetthis.com
• password

--Nectar ^T 06:27, 3 October 2005 (UTC)

Randall Parker gives a critical review at his Parapundit blog, arguing that the test score gains for Black and Latino students in Wake County are actually slightly smaller than the test score gains for Black and Latino students that have occurred state-wide. He argues the state's tests have been dumbed down, which (if true) would pose a difficulty in measuring real gains.--Nectar ^T 05:57, 4 October 2005 (UTC)

That's unfortunate, but probably true. The easiest way to eliminate the B-W gap is the reduce the g-loading of the standardized test. Of course, then the test loses predictive validity, which is mostly due to g. As per Murrays commentary article, that fact that the school achievement gap (measured with a good achievement test) has narrowed so little is disheartening wrt the extent that the B-W gap is most due to g. --Rikurzhen 08:15, 4 October 2005 (UTC)