Repost from the old site.
Note: this post is both lengthy (64 pages) and highly complex. I have tried to make it as understandable as possible, as this blog is geared towards the general reader, but some stuff is just inherently complex. If you think you can’t handle it, you might just want to bail out now; life is difficult enough as it is without me donating a frustrating reading and thinking experience to you.
If you want to sally forth, anyway, go ahead. Meantime, here is a good introduction by James Flynn that is fairly brief and easy reading. In particular, he deals with Jason Malloy’s notion that the Flynn Effect is worthless and not a rise in intelligence since it lacks a g-factor load. If you enjoy that, come back here and spear some of the red meat:
In recent days a particularly forceful assault has been leveled against my post, The Skyrocketing Black IQ, by Jason Malloy on the Gene Expression (GNXP) blog. Please see both of the linked articles above to get a background on the highly complex issues involved in this debate.
I first saw Malloy in the comments on the Dienekes blog.
Malloy hides his de facto racist message with various pretensions and obfuscations, perhaps unconsciously.
An artist from Madison, Wisconsin, he fashions himself an autodidact scientist and mathematician, and in my opinion, he and his ilk are among the most dangerous de facto racists on the Internet, mostly because they seem so reasonable and non-racist about their message. We need to take on him and his ilk with everything we’ve got.
The Malloy line is that environment does absolutely nothing and genes are everything. His political views? He says he is a liberal Democrat. I don’t understand this at all, but I’ve seen stranger things. To me, the politics he says he has and the politics he promotes in his writings are contradictory, but whatever.
He is neither a scientist nor a mathematician, has no degrees in either field, and has no peer-reviewed publications, yet used to go around to science blogs lecturing actual scientists on how to do science.
Real researchers have a skeptical attitude. Molloy does not.
After making comments on my post on Dienekes’ superb blog, Malloy then reiterated his rejoinders in a new post on the ultra-racist, super-arrogant, disturbing, childish and downright embarrassing GNXP blog.
Malloy insists he is not a racist; he is just dispassionate scientist without a science degree who is looking for the cold, hard truth. After some emails with him, I do not think he is consciously racist, but he is pushing a de facto racist message whether he intends or not.
A lot of these guys fashion themselves as “scientific martyrs” – a modern day Galileo. If only people would realize the unpleasant truth: Blacks are stupid, permanently so, and apparently nothing short of gene therapy will change this sorry state. I’m sure just about every Black person around is dying to believe such wonderful things about themselves.
There is no scientific consensus for such a painful conclusion, so there is no reason for any non-racist or Black-friendly person to take that line. The words of Malloy and the many like him amount to psychological warfare against Blacks. At the end of the day, it is more than that: it is a declaration of de facto racial warfare on every Black person on Earth hiding behind the lab coat of science.
Malloy has some charming things to say about US Blacks in that widely-linked piece. He ridicules their mental skills by calling them “story-telling ability IQ”. Their ancestors in Africa have a high “mosquito-dodging IQ“.
He disagrees with my suggestion that the Whites of the Greatest Generation have lower IQ scores than Blacks, remarking that 1945 Whites could not possibly be “even less intelligent” than our considerably dense Black citizens.
Malloy also claims that modern science proves Blacks are much more inferior as workers than Whites, that this is a serious drag on any employer of Blacks, and that the reason for this is their incurable genetic stupidity. The implication is that Blacks simply should not be employed in any sane workplace.
A White friend of Blacks can make sometimes-painful impressionistic observations about Black culture, without being nasty and racist-sounding about it. See here for an example. Malloy and those like him forgo that route and just slam away at Blacks, then are stunned when people note the racism, intended or not, that shines through loud and clear.
GNXP is a Social Darwinist blog where conceited, know-it-all, rude, immature young upper class “intellectual elitist” manchildren make asses out of themselves, and in so doing, idiotically try to poison their future careers. The racism and general elitism, especially the leaderships of South Indian high caste elitists Newamul Khan (Razib) and Paul Wickre (Godless) of GNXP has been cataloged on the Net in many places.
What’s ticklish is that if US Blacks are idiots, what does that make Razib’s Desis, who are apparently even dumber than the Blacks he scorns? Don’t ask. Like so many insecure narcissists, Razib attacks those above him (in this case Blacks and Hispanics) while projecting his inferiority complex on them. It’s Psychology 101, and it would be funny if it was not so pathetic.
GNXP is a hangout where petulant, conceited, condescending, basically immature cranks can pitch their atavistic and nihilistic misanthropy couched as “science” while they cheer themselves on.
Malloy’s debating and authorial style is sledgehammer-like. I have yet to see him concede a point, and he typically refers to opponents as ignorant. Not a good way to win friends and influence people, and it won’t fly in any peer-reviewed journal.
Malloy’s counterattack has been widely linked around the blogosphere, usually in a laudatory way, and was even discussed in a widely-read NY Times article. Unfortunately, instead of being the tour de force it appears to be, this carefully-penned attack is a de facto racist screed.
The author takes as face value that US Whites have higher IQ’s than US Blacks, that this difference is largely genetic and therefore unalterable by any environmental means (except presumably gene therapy) for at least the next century, and that the Flynn Effect (see below) is not an intelligence gain at all and has no bearing on the Black-White (B-W) IQ gap.
Furthermore and most importantly, Malloy seems to take utter delight in all of the above positions.
Except for the italicized first one, none of these positions are supported by scientific consensus, to put it mildly. Even the first position is widely disputed, at least in the popular press most folks read. Even the US Federation of American Scientists disagrees, though I think they are wrong on that. Bottom line: this debate is very much an unsettled issue, for better or worse.
Anyone who has gleefully taken such an extreme position and argues forcefully against all contrary evidence should have their motives questioned. The truth is that anyone staking out such an intellectual space is arguing from a de facto racist point of view, and unfortunately, Malloy’s widely hailed post is a tour de force of de facto scientific racism.
The real danger in this stuff (there are many more like Malloy out there) is the fact that it is very well researched and written and sounds quite convincing and reasonable. A maniac in a sheet or giving a Seig Heil can be scornfully dismissed. The de facto racism of a mild-mannered, reasonable-sounding liberal is much more frightening in its allure and potential to intoxicate.
The reason this is a racist position is that there is as yet no scientific consensus whatsoever about the B-W IQ gap. All we know is that it exists. All else is pretty much up for grabs right now (for exhaustive overviews of the state of the enormously complex debate, see here and here).
Many libertarians and extreme rightwingers are now taking a hardline, “Blacks are stupid” line on the B-W gap. That few Western Blacks or Mulattos, Hispanics, Amerindians, Polynesians, Black Africans, Southeast Asians or Arabs, not to mention Leftists, Liberals, Centrists or even mainstream conservatives take this position is telling.
This hard line is both unwarranted and unproductive.
On the other hand, it is important to not go to the other extreme. Pointing out the verifiable fact that IQ scores are low in Africa or there is a Black-White gap in the US is not racism at all, yet Western society seems to treat it as if it is. The treatment of James Watson’s basically reasonable remarks (other than the remarks about Black employees) is a sad testament.
And those of us opposing scientific racism need to come up with better arguments than, “No such thing as race,” “No such thing as intelligence”, “No such thing as IQ”, “IQ does not measure intelligence”, and on and on.
For instance, IQ is positively correlated with glucose metabolism, frontal grey matter and cerebral blood flow in the brain. A brain that works more, faster or better is likely to use more glucose and blood and to have more tissue than one that does not. This would hardly be the case if IQ had nothing to do with intelligence.
My post cited a phenomenon of rising IQ’s in the West for 80 years called the Flynn Effect (FE – named after James Flynn, the man who discovered it)1 and claimed that Blacks today have the same IQ score as Whites in 1957. 2007 Blacks also have IQ scores four points higher than Whites in 1945. If 1957 Whites were not stupid and did not have low IQ’s, neither are Blacks today stupid, nor do they have low IQ’s.
Here are Malloy’s complex charges:
He first asserts that there is something called g, or general intelligence factor (conceived by Charles Spearman)2, and that this is pure, or real, intelligence. Everything not g is not real intelligence. Instead, it’s just garbage, test-taking practice, “bias”, cultural bias, statistical noise, nothing at all, this or that, or whatever.
Then he asserts that FE rises are not on g, twisting and misrepresenting (Apparently deliberately!) several statements by Flynn. Malloy says they are not on g because the rises are not something called factor-invariant. On the other hand, Malloy notes the fact that the B-W gap is factor invariant.
In fact, Malloy is conflating two different ways of measuring g-factor here, one elaborated by Arthur Jensen, and another way of measuring g, called factor invariance, used by a team out of the Netherlands in 2004 (N2004)3.
The Spearman-Jensen g -factor (SJGF) does indeed correlate with the FE as shown by Flynn in a counterargument to Jensen who said his g-factor did not correlate with FE (the highest FE rises have been on the most SJGF-loaded tests of them all – see Raven’s Progressive Matrices4).
However, there is no correlation between the FE and the N2004 factor-invariance method of measuring the g-factor (but see here 3 for quotes from the study that imply that the authors felt there was some correlation).
Also, note that the same authors issued a paper three years prior in which they called into question Jensen’s prize argument, that the B-W gap was due to g and that it was genetic and irremediable – i.e., Blacks are doomed to be stupid forever.5
It’s confusing, and overviews of several studies indicate that whether or not the FE is correlated with g is still very much up in the air, in part due to varying ways in measuring g. A cautious conclusion is that the FE shows a fluid-g gain but not a crystallized g gain.
Fluid intelligence is a rough measure of how fast and efficiently one’s brain works, and tends to decrease with age. Crystallized intelligence is more what you know, rather than how fast or well your brain works. It is also the sum total of life experience that we call “wisdom”. As one might expect, crystallized intelligence tends to increase with age, and even aged people can score very high.
The FE is not about knowing more stuff (young people today often seem strikingly ignorant to me, at the same time that I am shocked by how well their brains seem to work), nor is it about wisdom (see my post here that shows that Black pathology in the US increased wildly in the past 50 years while Black IQ’s went through the roof due to the FE6).
Malloy concludes that FE gains are not real intelligence gains since they are not g gains (but see above), and hence are worthless. He also states that 2007 Blacks do not have the same IQ scores as 1957 Whites, nor do their scores surpass 1945 Whites. Flynn disagrees.
There is indeed a paradox between the FE and the B-W gap. The B-W gap shows the stereotypical g-factor pattern – that is, the g-factor will measure the degree to which superior test-takers will tend to have a score lifted across all subtests, or across the board.
We could say that they seem smarter in an “across the board” way. It is true that the Blacks of today are not smarter “across the board” than the Whites of 50 years ago, but neither is the converse true, as Malloy implies.
Flynn and others offer some intriguing theories on the differences between g and the FE, and Flynn suggests that the g construct is not terribly important. Flynn suggests that the FE shows gains in intelligence within groups over time, like a movie. If you stop the movie at any time and take a snapshot, you see g. Start the movie again, and you see the FE.
Therefore, it seems wrong for Malloy to bash the FE on the basis that it is not on g, since the FE and g seem to be two entirely different things.
Put another way, we are comparing apples and oranges. It is true that apples do not correlate well with some hypothesized “orange factor”, but they need not and should not, and anyway, the fact that they do not does not imply that an apple is not fruit. It’s just that there is no reason to be comparing them.7.
Furthermore, Malloy and Jensen have problems when trying to say that the more a test taps g, the more genetic it is. B-W differences are highest on the subtests that are greatest in g, and the more a test taps g, the more genetic it is. How then to explain the massive FE, which is mostly on the most heavily g-loaded tests of all?
Malloy is also incorrect to imply that non-g FE gains are “hollow”, worthless, or not “real intelligence”. For instance, Malloy repeatedly states that IQ differences between ethnic groups (he often compares Blacks and Whites) are g-based, hence real, genetic and irremediable.
However, comparisons of Whites with some 3rd World groups show that the difference is on non-g. Hence, it follows that these 3rd World groups have the same intelligence as Whites.8.
A hallmark of the B-W gap genetic theorists rests on the scores of US Blacks, who strangely score 85, while African and Caribbean Blacks score 67 and 71 respectively. The hereditarian notion is that White admixture is what raises the US Black score, since African Blacks are pure Black and Caribbean Blacks are mostly Black. There are serious problems with this approach.9
Malloy is fond of several other claims.
Foremost, they believe that the B-W gap is wholly genetic, and hence cannot be remedied by any environmental factors or interventions. Blacks are hopelessly stupid forever, and the implication is that all money spent trying to give them a brain is down the drain.
They are doomed to wail, “If I Only Had a Brain“, in the wilderness forever, like the Strawman in the Wizard of Oz. The implication is that, as a hopeless race, Blacks must be cast off onto the proverbial ice floe as in the Eskimo custom, set free to fail and die as evolution deems fit.
In particular, Malloy attacks adoption studies, saying that adopting Blacks into White homes failed to raise their IQ’s. However, the evidence from adoption studies is far more complex than that, and many have shown gains for Blacks in these situations.10
Furthermore, even the model that holds that IQ scores within families are heavily genetic and owe almost nothing to shared home environment has been challenged, with radically different loadings for poor and high-income households on these factors.11
An anomalous study in Germany (Eyferth 1961) based on the children of Black and White soldiers and White German women found no differences at all in IQ among the White and Mulatto children.
Note that many pushing a hardline “Blacks are forever stupid” line today are extreme rightwingers or libertarians. They are almost all Whites, and, in particular and creepily, Northern European Whites. There are a few high-caste South Indians and a Chinese here and there.
Jensen’s first article in 1969 stated that Blacks were dumb, that it was genetic and permanent and that Head Start should therefore be abandoned.
There is a reason for this political stance and there is a reason that almost everyone saying this far to the Right politically. There is an agenda being pushed here. If we can prove scientifically that Blacks are a hopelessly moronic race, and that their stupidity is embedded in their genome with no way of fixing the matter, we can cut off Blacks of any and all social programs designed to help them, educate them or make them smarter.
More savagely, the implication is that they are a useless evolutionary dead end that needs to be “phased out” in the frightening words of Richard Lynn. All money spent to help this dead-end hopeless race of idiots, either in the US or in Africa, is down the rathole. We should not give them medical care, income, rent, welfare, food stamps, disability, WIC or any other assistance or help in any way.
Africa, a continent reeling in disease and poverty, needs to be thrown out of the nest like a baby bird and told it is on their own, cutting off all food, medical and developmental aid. The result will be massive death and suffering in Africa. In the US, similar policies will be less severe but still frightening and tragic.
It is important to see the cruel and deadly racist motives behind scientific racism, even when promoted by consciously non-racist people. Even if their arguments are correct, the prescriptions most of them are pushing, if adopted, will mean chaos and widespread death and suffering for countless Blacks. It amounts to a declaration of war on Black people everywhere.
Scientific racism also claims that evidence shows that educating Blacks is useless, and most are ineducable. Many point to an set in stone “achievement gap” to show that Blacks are permanent dolts and need to be cut off. However, there is impressive evidence that the achievement gap is closing and that spending money on Black education does improve Black scores,
From Malloy’s piece:
An additional popular argument is that the Flynn Effect, the observed rise in IQ scores over time, is evidence that African-Americans or African countries will eventually reach parity with white norms. This typically includes the premise that white intelligence in the recent past was even lower than modern black intelligence. A typical example:
US Blacks, with an average IQ today of 85, have the same IQ as US Whites with an IQ of 100 in 1957. If 1957 US Whites were not stupid, then neither are US Blacks today. It’s time to shut up about the “low Black IQ”, since by any reasonable standard, it is not really low at all.
These arguments are wrong for the simple fact that the Flynn Effect is not a gain in real g factor intelligence, while the differences between nations and ethnic groups are differences in g factor intelligence. These findings led a 2004 team to state:
It appears therefore that the nature of the Flynn effect is qualitatively different from the nature of B-W [Black-White] differences in the United States… [so] implications of the Flynn effect for B-W differences appear small…
James Flynn, namesake of the secular increase, reiterates (DOC) these points:
Factor analysis is a way of measuring this tendency of some people to do better or worse than average across the board; and it yields something called g (a sort of super-correlation coefficient), which psychologists call the general intelligence factor…When you analyze IQ gains over time, you often find that they do not constitute enhancement of these latent traits — they do not seem to be general intelligence gains, or quantitative factor gains, or verbal factor gains (Wicherts et al, in press).
In the language of factor analysis, this means that IQ gains over time tend to display ‘measurement artifacts or cultural bias’. For a second time, we are driven to the conclusion that massive IQ gains are not intelligence gains or, indeed, any kind of significant cognitive gains (pp 27-28).
Flynn believes the secular increase represents important changes in specific narrow aspects of developed cognitive style, but not a rise in g intelligence. It is therefore incorrect that 1945 US whites were less intelligent than 2007 US blacks. The Flynn Effect has little apparent bearing on racial intelligence gaps.
A brief perusal of the Flynn links that Malloy provided above showed clearly that Malloy grotesquely took Flynn’s words out of context, possibly deliberately.
For one, Malloy implies that Flynn concedes that FE gains are not gains in any kind of real intelligence. For years Flynn did think that the gains were a practice effect or test artifact, but in recent years, he has changed his tune, and now really does believe that we are smarter in some ways. Even racist blogger Steve Sailer agrees.
Further, even scientific racists Steve Sailer, Richard Lynn and his co-author Tartu Vanhanen believe strongly in the FE, Lynn so much so that Sailer calls it the Lynn-Flynn Effect.
Malloy has been staking out this extreme scientific racist stance for years now – see here from 2003. Ole Eichhorn, a colleague of his at GNXP, does not share his views and believes strongly in the FE. Malloy at times parrots scientific racist hardliners Jensen and Charles Murray. Murray is a libertarian and Jensen is apparently an extreme rightwinger.
In a recent paper with fellow scientific racist Rushton, Jensen laid out his position: that an acceptance that Blacks are genetically less intelligent than Whites should logically lead to an end to anti-discrimination lawsuits. The burden of proof in such cases would fall on the plaintiff in proving that the employer was motivated by bias and not by Blacks’ scientifically proven genetic intelligence deficits.
This is the brave new world that the scientific racists have in store for us: where Blacks are forever resigned to much lower socioeconomic, health and educational status, with no way for them to equalize things even a bit. It’s this, and not “search for truth”, “science” or Galileo-martyr complexes, that motivates scientific racism, and it is for this reason that we must fight it with all we have.
I sent Flynn an email linking both my post and Malloy’s counterattack. He emailed me back13. Flynn made three rather confusing points, quoting from a book of his now on sale at Amazon:
- The FE gains are on g but at the same time they are not on g.
- The FE gains are real intelligence gains in spite of the fact that they are not factor-invariant.
- Flynn agreed with me that Blacks today have the same IQ’s of Whites of 1957, but he made a confusing statement that seemed to imply that Blacks may continue to gain on Whites, and possibly even move to close the gap.
Here is the statement for you to play around with: Blacks do match the whites of 50 years ago but that does not settle the debate. It can be argued that they are the beneficiaries of the effects of modernity over that time and this could be either more or less than the environmental disadvantage they suffer at any given time.
According to this mail, Flynn disagrees with Malloy’s assertions above and also disagrees with all views that Malloy has attributed to Flynn above. However, in a recent email, Malloy agreed that he and Flynn disagree about the meaning of the FE.
There are a lot of confusing terms in here – g factor, factor invariance, etc. See the footnotes and click some of the links to try to understand it better, but this is stuff is not easy going.
1. The FE in the US shows a 18 point gain over the 70 years from 1947-2002. In my opinion, Blacks have scored higher, reaching 23.5 points.
There are many possible causes of the FE in modern life. According to Flynn in his new book, they may be: Increased demands from more professional, technical, and managerial jobs; increased leisure time; changing cognitive demands of personal interactions; or changing attitudes toward intellectual activity. Nutrition, better schooling, video games, television, computers, cell phones and the increased complexity of modern life are all thought to play a role.
2. The g-factor is best thought of as a correlation coefficient that shows the extent to which superior test takers, if they score better on one test, tend to score better on other tests across the board. It indicates a brain that works “better all across the board” than others with lower g.
A group higher in g than another group will tend to score better than the lower group in an across the board kind of way. Like IQ, g also correlates with quite a few correlates of intelligence such as increased grey matter in the brain.
Jensen’s original extension of Spearman’s g involved the concept that that g-factor is what he called “Level 2 conceptual learning” – the ability to solve problems = conceptual learning = synthesizing ability. Non-g factor was conceived by Jensen as a “Level 1″ type of learning, and is roughly the retention of input and rote memorization of simple facts and skills = associative learning = memorizing ability.
Most IQ subtests would appear to measure Level 2 conceptual learning quite well. Actually, by Jensen’s own reasoning, he felt that the B-W gap was only 60% in g.
3. Raven’s, the most g-loaded test ever made, shows the largest FE of any test. The test was in fact specifically designed to measure g-factor intelligence only. The test has more predictive validity than any other test used in occupational groups and especially predicts mobility in jobs. It measures ability to “make sense out of the buzzing confusion of life”.
Scientific racist Philippe Rushton shows that Raven’s has no cultural bias whatsoever.
4. This extremely complex paper notes: gains cannot be explained solely by increases at the level of the latent variables (common factors), which IQ tests purport to measure.
In other words, according to this study, some of the:
…inter-generational difference in IQ is attributable to bias or other artifacts, and not real gains in general intelligence or higher-order ability factors…A cautionary conclusion would be that part of the gains (excluding the sub-tests Similarities and Comprehension) could be explained by genuine increases in intelligence.
Whereas implications of the FE for B–W differences appear small, the implications for intelligence testing, in general, are large.
As we can see, even Malloy, Jensen and Rushton’s favorite study indicates that part of FE gains could be on their version of g.
5. Dolan and Hamaker 2001 reanalyzed the data from several earlier studies and concluded that Spearman’s hypothesis is not an “empirically established fact” (i.e., that Black-White IQ differences may be due to differences in common factors other than g) due to insufficient power in the data to choose between alternative models.
The paper said:
This leaves the validity of Spearman’s hypothesis, considered a central justification for the genetic explanation, an unresolved question…
However, they did confirm that the Black-White IQ gap is not due to measurement artifacts, and is instead due to some measured factor that varies both within and between groups. These same people stated that the g factor did not explain the B-W gap as it was not explanatorily accurate.
The paper reanalyzed the data from several previous studies (Jensen and Reynolds 1982; Naglieri and Jensen 1987) that used the statistical method invented by Jensen (the method of correlated vectors – g factor) with a more recent and improved method (multigroup confirmatory factor analysis).
On the basis of the present, as well as other results (Dolan 2000), we are convinced that the Spearman correlation cannot be used to demonstrate the importance of g in b-w differences with any confidence…
It is possible that the analysis of all available data sets (perhaps using an appropriate meta-analytic procedure) will demonstrate that a model incorporating the weak version of Spearman’s hypothesis provides the best description of the data. However, until this work is undertaken, we cannot accept Spearman’s hypothesis as an “empirically established fact 
6. There is quite a bit of existing evidence suggests that FE gains across subtests may indeed be positively correlated with g loading. See Colom 2001 for good correlations between FE and g, and Juan-Espinosa 2000 for much stronger correlations between g loadings and FE gains.
Jensen’s The g Factor, pp. 320–21, reviews a number of studies of the relation between FE subtest gains and g loadings, all of which show weak positive correlations. However, Rushton 1999 finds that a measure of g developed on the Wechsler Intelligence Scale for Children (WISC) has loadings that are negatively correlated with FE subtest gains in several countries.
Yet this was challenged in Flynn’s new book, Extending Intelligence: Enhancement and New Constructs, where he makes the argument that FE gains are greatest on tests of fluid g rather than crystallized g. Flynn finds a positive (though statistically insignificant) correlation between a measure of fluid g he develops and FE gains in the data used by Rushton.
Two tests in Estonia did not find a correlation between g loadings and FE gains on those tests, but these are achievement tests with a strong crystallized bias. See Must 2001 and Must 2003.
7. For an introduction to Flynn’s new book in which he explores the various paradoxes of the FE, see Amazon’s page on the book. For a much more complex take on how FE gains can be on g but not on g, see this paper.
From the paper:
Numerous studies indicate that measures of intelligence reliably produce a latent general factor, referred to as g, which is associated with real world competence and is highly heritable (see Jensen 1998).
However, analysis of measurement invariance between normative cohorts in gains on five standardized measures of intelligence indicated that IQ gains cannot be attributed to change in a general latent factor but primarily reflect systematic variation in specific aspects of cognition leading to bias between cohorts on a number of subtests of intelligence measures (Wicherts et al., in press).
Something in the experience of successive generations is bringing about changes in aspects of cognitive functioning that are associated with g within but not between cohorts.
See also Flynn’s superb but hard to follow lecture (based on his new book) at the Psychometrics Centre in Britain where he made fascinating points and intriguing allegories while offering fascinating examples. From the lecture:
(1) The factor analysis paradox: Factor analysis shows a first principal component called g or general intelligence that seems to bind performance on the various WISC subtests together. However, IQ gains over time show score gains on the WISC subtests occurring independently of one another.
How can intelligence be both one and many?…
(1) The WISC subtests measure a variety of cognitive skills that are functionally independent and responsive to changes in social priorities over time. The inter-correlations that engender g are binding only when comparing individuals within a static social context…someone who tended to be superior on any one would tend to be above average on all.
Let us assume that the 100 meters, the hurdles, and the high jump all had large and similar g loadings as they almost certainly would. A sprinter needs upper body strength as well as speed, a hurdler needs speed and spring, a high jumper needs spring and timing.
I have no doubt that a good athlete would best the average athlete handily on all three at a given place and time. However, over time, social priorities change. People become obsessed with the 100 meters as the most spectacular spectator event (the world’s fastest human). Young people find success in this event a secondary sex characteristic of great allure.
Over 30 years, performance escalates by a full SD in the 100 meters, by half an SD in the hurdles, and not at all in the high jump. Although average performance has risen “eccentrically” on various events, the following is still true: superior performers still do better than average on all 10 events and are about the same degree above average on various events as they were 30 years before.
Arithmetic, Information, Vocabulary, and Similarities all load heavily on g(IQ) and on a shared verbal factor. Despite this, Americans gained 24 points on Similarities between 1947 and 2002 (1.6 SD’s), 4 points on Vocabulary, and only 2 points on Arithmetic and Information…
To sum up. Factor analysis and g(IQ) describes a static situation where individual differences are compared and social change is frozen. The degree to which superior people are above average on the various subtests sets their respective g loadings.
IQ gains over time describe a dynamic situation in which social priorities shift in a multitude of ways. No better math teaching, more leisure but with the extra leisure devoted to visual rather then verbal pursuits, the spread of the scientific ethos, and a host of other things all occurring together.
The average on Similarities rises but the average on Arithmetic and Vocabulary does not. How odd it would be if social trends mimicked factor loadings in determining what real world cognitive skills progress and which mark time! If they did so, IQ gains would appear factor invariant, but that would be purely accidental (Wicherts et al 2004).
Although radically different trends alter average performances on various WISC subtests between Time 1 and Time 2, note that this leaves a certain stability untouched. Superior performers are much the same degree above average on each and every subtest at both Time 2 and Time 1. Therefore, much the same g will emerge.
Our first paradox is resolved. At any particular time, factor analysis will extract g(IQ) — and intelligence appears unitary. Over time, real-world cognitive skills assert their functional autonomy and swim freely of g — and intelligence appears multiple. If you want to see g, stop the film and extract a snap shot; you will not see it while the film is running. Society does not do factor analysis.
It is a juggernaut that flattens factor loadings and imposes its own priorities. Similarities requires you to solve problems on the spot without a previously learned method. The Performance subtests… require … arranging pictures to tell a story.
General information, math and vocabulary show only small gains on most tests, though the Dutch scored high gains. There have also been IQ identical rises in NAEP tests. Children are learning to read at an earlier age, but they are no better at reading adult literature than before. Basic math skills have shown impressive gains, but mathematical reasoning has been flat.
That means people are no better at geometry and algebra than they were in the past. Ability to solve problems without a previously learned method will be good for managerial, professional, and technical jobs to fill — jobs that often require decisions without the guidance of set rules. We are better at solving novel problems verbally, visually and abstractly.
Previous generations were more rule-governed and less likely to “think outside the box”. Previously, people engaged in pre-scientific, more concrete operational thinking. Now, they are post-scientific and more abstract.
What do dogs and rabbits have in common? They are not retarded, they use syllogistic logic, but they will appear retarded on Similarities. There is more lateral thinking now, and the human mind has been liberated – the liberation of reason from the concrete. There has been the spread of the language and categories of science and on the spot problem solving has been enhanced.
So, Flynn shows that IQ gains are real and that they are the result of environmental progress. The artifact, test taking or bias options cannot be supported by evidence. Test sophistication shows a particular pattern that is not present in the Flynn Effect. Flynn also resolutely shoots down cultural bias.
A series of studies were undertaken by a Russian researcher in first decade of the 20th Century amongst Siberian peasants. He was studying their cognitive styles. Looking at their test answers today, we can see how they were not retarded at all, but would have scored retarded on some of our modern IQ tests.
From the lecture:
Q: All bears are white where there is always snow; in Novaya Zemlya there is always snow; what color are the bears there?
A: I have seen only black bears and I do not talk of what I have not seen.
Q: But what do my words imply?
A: If a person has not been there he can not say anything on the basis of words. If a man was 60 or 80 and had seen a white bear there and told me about it, he could be believed.
This indicates that the Siberian peasant uses syllogistic or concrete logic that was useful for him at the time. As you can see, the questioner tries to get the peasant to think scientifically, but the man stubbornly refuses to, possibly because he is not able to. Not because he is stupid, but because he has never learned how to think this way.
If you tried to explain to the peasant the more abstract and rational way of looking at the problem, he might have argued with you, but more likely he would give you a look that says, “Why do you think that is important?” There was little use in his world for abstract thinking.
Again, in the lecture, Flynn shows that if you asked a boy in 1920 what dogs and rabbits have in common, he would probably have given you an “odd” answer like, “You use a dog to hunt a rabbit.” Nowadays, if someone gave an answer like that, people would crinkle their eyes, look puzzled and say that that answer was “Idiotic!”.
This means it shows contempt for scientific norms, not that it is incorrect – one may indeed use a dog to hunt a rabbit – but modern science does not deem that relationship the most important one between the two animals.
Even a youngster in the ghetto with poor grades nowadays may be able to give the answer we deem “correct”: “Both dogs and rabbits are mammals”. The boy in 1920 is not slow or retarded, he is just using concrete and pre-scientific thinking. The boy in the ghetto, despite his gangsta language and clothing, may at least be able to think like a scientist.
Flynn also shows how Rosenau and Fagan (1997) compare the 1918 debate on women’s suffrage with recent debates on women’s rights and make an excellent case that the latter shows less contempt for logic and relevance. This shows that we are becoming less illogical and irrelevant in our thinking over time.
That our ancestors were illogical or irrelevant in some aspects of their thinking does not mean they were retarded or slow. Illogicality and irrelevance were useful to them in some ways as argumentation styles and in making sense of the world in pleasing and comforting ways; they were culturally sanctioned and incurred little to no penalties.
The Darrow Trial in the 1920′s is another example. They were not retarded or stupid, just unscientific.
8. For instance, B-W IQ differences in South Africa are not factor invariant, so according to Malloy, these differences are not real at all. Malloy’s position would have to be that there are no differences between White and Black IQ’s in South Africa, despite a White score of 100 and a Black score of 67. Is this a position that he is willing to take?
Furthermore, in his book, The g Factor , Jensen himself says that White – South Indian IQ differences are not based on g. Therefore, according to both Jensen and Malloy, Whites and South Asians score the same on IQ tests, despite a White score of 100 and a South Indian score of 81. Is Malloy willing to concede this?
9. Clearly, US B-W differences compared to US White – African differences cannot be explained by the European admixture in the blood of US Blacks. US Blacks are 17.5% White on average. Adding 17.5% White (IQ 100) to the African IQ of 67 gives us 67 + 6 = 73. Instead of the expected 73, we get 85. Clearly, something other than genes is raising US Black IQ.
In addition, numerous studies have been done to try to determine whether or not White admixture in Blacks is correlated with IQ. These studies have found nothing – there is little or no correlation between the degree of White blood in a Black and their IQ score.
Since the US B-W gap is said to be based on g and purely biological, yet there is no difference in Black IQ based on White ancestry level, this takes apart Jensen’s notion that both the B-W difference and g are biological.
10. For instance, Tizard 1974 showed that the Black orphans raised in an orphanage were smarter than the White orphans. Moore 1974 and Willerman 1986 found that Blacks adoptees raised by White mothers had higher IQ’s than adoptees raised by Black mothers.
Scarr and Weinberg (1976, 1983) in a study of Blacks adopted into upper-income White homes in Minnesota, showed that Black children benefited by adoption into high-income white families. Black children had IQ’s of 94 (5 pt gain) and mixed-race (mulatto) adoptees had IQ’s of 99.
The effect wore off by late adolescence and adulthood, but this just shows that home environment only benefits Black children, and the effect wears off by late adolescence to adulthood. But late adolescent to adult Blacks may need a new sort of stimulating environment to maintain their gains from childhood.
Most adoption studies, as noted by the Minnesota trans-racial study above, show gains only in childhood, with gains wearing off by adulthood. However, one study did find lasting gains into adulthood. This effect lasted into adulthood.
11. For instance, from Turkheimer 2003: The models suggest that in impoverished families, 60% of the variance in IQ is accounted for by the shared environment, and the contribution of genes is close to zero; in affluent families, the result is almost exactly the reverse.
12. Huang and Hauser 2000 show that, controlling for social background, Blacks made significant gains in closing the Black-White achievement gap from 1974-1998. Since then, they have begun to lose ground as rightwing policies to cut back on education funds for Blacks have taken hold, harming Blacks. Boozer and Cacciola 2001 show that reducing class size caused an impressive increase in achievement test gains.
13. It is normally a violation of netiquette to print private emails without someone’s permission, but the folks at GNXP demanded that I post Flynn’s letter in its entirely. I do not reprint the whole letter, but I do reprint the parts that are important.
I bold the parts of his letter that are at least a bit confusing to me. A characteristic of the narcissistic young men on GNXP is that they never admit that there is anything they don’t understand, and they never concede a point. Being a grownup and all, I do both.
You raise three questions and my current book (What is intelligence?) illuminates all three. You can get it from amazon.com for $14.96 or probably from a Border’s bookstore:
(1) IQ gains are significant despite not being factor invariant – see chapter 2;
(2) Black gains on white are gains in g even though they are not g gains – sounds odd but true – see pp 60-63;
(3) Blacks do match the whites of 50 years ago but that does not settle the debate. It can be argued that they are the beneficiaries of the effects of modernity over that time and this could be either more or less than the environmental disadvantage they suffer at any given time.
- Blair, C., Gamson, D. Thorne, S., Baker, D. “Rising mean IQ: Cognitive demand of mathematics education for young children, population exposure to formal schooling, and the neurobiology of the prefrontal cortex”, Intelligence 33 (2005) 93–106.
Boozer, Michael A. and Cacciola, Stephen E. “Inside the ‘Black Box’ of Project STAR: Estimation of Peer Effects Using Experimental Data,” Discussion Paper 832 (Economic Growth Center, Yale University, 2001).
Colom, R., Juan-Espinosa, M., and García, L.F. “The Secular Increase in Test Scores Is a ‘Jensen effect,’” Personality and Individual Differences 30 (2001): 553–58.
Klaus Eyferth “Leistungen verschiedener Gruppen von Besatzungskindern in Hamburg-Wechsler Intelligenztest fur Kinder (HAWIK),” Archiv fur die gesamte Psychologie 113 (1961): 222–41.
Flynn, James R. “The History of the American Mind in the 20th Century: A Scenario to Explain IQ Gains over Time and a Case for the Irrelevance of g,” in Extending Intelligence: Enhancement and New Constructs, edited by P. C. Kyllonon, R. D. Roberts, and L. Stankov (Hillsdale, N.J.: Erlbaum, forthcoming).
Huang, Min-Hsiung and Hauser, Robert M. (2000) Convergent Trends in Black-White Test-Score Differentials in the U.S.: A Correction of Richard Lynn
Jensen, Arthur The g Factor: The Science of Mental Abilities, pp. 320–21. (Westport, CN: Praeger, 1998)
Juan-Espinosa, Manuel and others “Individual Differences in Large-Spaces Orientation: g and Beyond?” Personality and Individual Differences 29 (2000): 85–98.
Moore, Elsie G. J. “Family Socialization and the IQ Test Performance of Traditionally and Transracially Adopted Black Children,” Developmental Psychology 22 (1986): 317–26.
Must, O., Must, A.. and Raudik, V. “The Flynn Effect for Gains in Literacy Found in Estonia Is Not a Jensen Effect,” Personality and Individual Differences 33 (2001).
Must, Must, and Raudik, V. “The Secular Rise in IQs: In Estonia the Flynn Effect Is Not a Jensen Effect,” Intelligence 31 (2003): 461–71.
“Myth: Social intervention cannot raise IQ;”, in “Intelligence Quotient”, chapter in The Encyclopedia of Adoption.
Rushton, J. Philippe “Secular Gains in IQ Not Related to the g Factor and Inbreeding Depression—unlike Black-White Differences: A Reply to Flynn,” Personality and Individual Differences 26 (1999): 381–89.
Scarr, S. and Weinberg, R. A. “IQ Test Performance of Black Children Adopted by White Families,” American Psychologist 31 (1976): 726–39.
Scarr, S. and Weinberg, R. A. “The Minnesota Adoption Studies: Genetic Differences and Malleability,” Child Development 54 (1983): 260–67 .
Tizard, Barbara “IQ and Race,” Nature 247, no. 5439 (February 1, 1974).
Turkheimer, E., Haley, A., Waldron, M., D’Onofrio, B. and Gottesman, I. I. “Socioeconomic status modifies heritability of IQ in young children”, Psychological Science 14 (6) (2003), 623–628.
Wicherts, J. M., Dolan, C. V., Hessen, D. J., Oosterveld, P., Baal, G. C. M., van Boomsma, D. I., & Span, M. M. “Are intelligence tests measurement invariant over time? Investigating the nature of the Flynn effect.” Intelligence 32 (2004), 509–537 .
Willerman, Lee and others, “Intellectual Development of Children from Interracial Matings: Performance in Infancy and at 4 Years,” Behavioral Genetics 4 (1974): 84–88.