Smarter, Taller, Healthier ?

by John Q on April 14, 2007

John Holbo’s naming of the two-step of terrific triviality reminded me that this manoeuvre is one of the basic steps in the nature-nurture dance. I looked at
Pinker0211-1Steven Pinker’s agile performance
a while back.

Anyway, this reminds me of a vaguely related point I’ve been meaning to make for a while. Debate over the relative influence of environment and heredity on intelligence has been going on for at least a century without much change or resolution, or any obvious reduction in the level of vitriol. The only significant new information in the last few decades has been the discovery that average IQ scores have risen substantially over time (the so-called Flynn effect). There has been vigorous debate over whether this effect is real or spurious.

On the other hand, no-one seems particularly exercised about the relative effects of nature and nurture on height, even though the observed patterns seem to be much the same: a fairly high correlation between parents and children, significant class effects, a correlation with wages and a surprisingly strong increasing trend over time.

And much the same things can be said about health, except that the parent-child correlation is specific to particular conditions.

Height, health status and measured intelligence are all positively correlated so it seems as if we should be looking for the same kind of explanation in all cases. This will be left as an exercise for readers (that is, I haven’t got around to working on it myself).

Update The comments do a good job of making my point. There’s plenty of vitriol on the subject of intelligence, but not much new. On the other hand, there’s some interesting, and reasonably civilised, discussion of genetic and environmental determinants of height.

{ 116 comments }

1

David 04.14.07 at 4:14 am

If I recall correctly, the Steven Pinker dance is called “The Heritability Hustle”

2

aa 04.14.07 at 5:13 am

Sure. But basically the choice is between medieval science and modern science. So if you actually want to know anything about this, all you have to do is wait. An unfamiliar notion.

3

William Blurke 04.14.07 at 5:39 am

By what logic do differences in culture between the US and Sweden or any other country have to do with Steven Pinker or the “nature-nurture dance”? And isn’t it at least a little odd that this discussion appears right after a post titled “The Paranoid Tendency in American Life”?

4

John Quiggin 04.14.07 at 7:41 am

wb, the only connection I’ve made between the two topics is the appearance of the same rhetorical manoeuvre. As regards paranoia, there’s a lot of it about.

aa, wtf?

5

Wade 04.14.07 at 8:10 am

Some readers of this blog had other names for this trick that were already in use and which were either more apt or more memorable. I don’t see what Holbo’s term has to recommend it besides cute alliteration.

6

kaw 04.14.07 at 8:28 am

“Height, health status and measured intelligence are all positively correlated so it seems as if we should be looking for the same kind of explanation in all cases.”

You mean, like prenatal health care and/or nutrition?

7

SusanC 04.14.07 at 8:44 am

John,

You make some gtood points, but I have a few quibbles, especially towards the end.

Unless we are prepared to embark on large-scale genetic re-engineering, our only hope is to focus on those aspects of our condition that
are amenable to nurture.

That’s a pretty big unless. The technology of genetic engineering is advancing so rapidly right now that I wouldn’t rule out the possibility of this happening within our lifetimes. I could imagine it starting with gentic screening for autism and schizophrenia and slowly expanding to other characteristics.

The real threat today is neither the rational use of force in the manner of Clausewitz nor aggressive genes inherited from the Pleistocene past but the culturally generated
generated craziness of Osama bin Laden and Timothy McVeigh.

The media loves to hype terrorism, but the typical Westerner’s risk of being killed by the likes of Osama bin Laden and Timothy McVeigh is very, very small. I’d be more worried about the craziness of leaders of nation states – the number of people they can kill has rather more digits in it.

In any case, by calling it “culturally generated” craziness, aren’t you just bluntly asserting the very question that’s in dispute? The relative effect of genetic and environmental factors on terrorism is an interesting question.

Finally, to argue that war is economically irrational, it’s not sufficient to show that it causes a loss for the country as a whole. Did the people who made the decision to go to war make a profit personally? (e.g. did it enable them to stay in office longer?) I could imagine a theory of war based on public-choice economics.

8

John Quiggin 04.14.07 at 8:46 am

“You mean, like prenatal health care and/or nutrition?”

That’s an obvious one, but there are plenty of other causal pathways, for example from education (own and parents) to health status.

9

SusanC 04.14.07 at 9:00 am

On the other hand, no-one seems particularly exercised about the relative effects of nature and nurture on height.

People don’t care about height as much. For example, it won’t cause much upset if you publish data on male vs female height differences.

10

Chris 04.14.07 at 9:23 am

Susan makes a good point. The nature v. nurture debate only really rages in “controversial” areas like intelligence. That’s not to say there still aren’t debates about whether specific components of our concept of cause, how we recognize faces, or even how tall we are, but they’re friendly debates. And they usually involve, you know, data.

11

Hidari 04.14.07 at 10:22 am

to follow up Chris’s and SusanC’s point a notable skeptic (I can’t remember who it was, but it may have been Michael Shermer) once pointed out that no one writes books with titles like ‘Why are the Chinese so good at table tennis and why is nobody talking about it?’But people DO write books about why African-Americans are ‘good at sports’ and why ‘political correctness’ ‘covers that up’. The implication being, of course, that black people are genetically predisposed to being good at sports and the hidden implication being that black people are good at physical tasks and white people are, therefore, good at mental or cognitive tasks. Genetically speaking.

It’s the same point. The idea that rising standards of nutrition, greatly increased access to information (via, for example, the internet), improved education and generally increasing wealth is, surely, the reason as to why IQ test results are increasing. But that’s too dull for some people, and so pseudo-controversy is created by promoting ‘controversial’ (i.e. fashionably right wing) ideas.

Insofar as the review of Pinker’s book goes, can I point out one huge, gaping, gigantic hole in Pinker’s theory? Pinker rages again and again and again against Cartesianism. He calls it the Ghost in the Machine. He thinks it’s old fashioned, out of date, ridiculous.

But Pinker’s ideas derive directly from Chomsky’s. (Pinker started off as a linguist, and did some good stuff in that field before he suddenly reinvented himself as an expert on absolutely fucking everything).

But Chomsky’s ideas are really an attempt, as Chomsky has always admitted to create a sort of modern, secularised, materialist Cartesianism. Chomsky even wrote a book entitled Cartesian Linguistics: A Chapter in the History of Rationalist Thought.

To quote the Wikipedia: ‘Certain mechanical factors of language function, such as response to stimuli, are evident in both humans and animals; however, Chomsky cites from several 17th century Cartesian experiments which show that the creative aspect of language is specific only to human beings. This is, in essence, the Cartesian theory of language production.

Chomsky writes, “one fundamental contribution of what we have been calling “Cartesian linguistics” is the observation that human language, in its normal use, is free from the control of independently identifiable external stimuli or internal states and is not restricted to any practical communicative function, in contrast, for example, to the pseudo language of animals”. “In short, animal ‘language’ remains completely within the bounds of mechanical explanation as this was conceived by Descartes and Cordemoy” and the creative aspect of language is what separates humans and animals.’

In other words, Chomsky always explicitly based his opposition to Empiricism and Behaviourism on his Rationalist Cartesianism, and, in fact, so does Pinker. But he then backtracks and simply denies that that is what he is doing.

Pinker deals from the bottom of the deck, argumentatively speaking.

12

bad Jim 04.14.07 at 10:33 am

A year ago I visited Amsterdam, and I expected the average Dutch man to be an inch taller than I am, 185cm minimum. It might even have been the case.

13

novakant 04.14.07 at 10:38 am

OMG, let’s just hope Steve Sailer doesn’t show up

14

The Constructivist 04.14.07 at 11:10 am

He did when I did the other version of this dance–the race v. racialization shuffle–around this time last year. Maybe he’s googling his name less often now.

15

Chris 04.14.07 at 12:13 pm

hidari, re Cartesianism, that’s a good point. Specifically, Chomskyism, and nativism in general in cognitive science, tends to be a form of rationalism (the brain/mind is imposing structure — e.g., linguistic structure — on input from the world). Given Pinker’s obsession with Evolutionary Psychology in the Cosmides/Tooby/Buss mode, it seems odd for Pinker to be complaining about Cartesianism. Next thing ya know, he’s going to say that cognitive science is too Platonic, and we should instead all be nativists.

16

Richard 04.14.07 at 12:58 pm

People don’t care about height as much.

true, at least it’s not on the agenda. Nevertheless, it seems like a powerful social indicator; in my (anecdotal and wholly unscientific) experience tall people get an easier ride generally and are given more credence and authority in personal interactions. IQ and class ruffle feathers, but social power comes from many quarters.

17

VernInBama 04.14.07 at 1:41 pm

Perhaps high IQ is something that offered no positive survival value prior to the development of modern society (assuming it now has value.) If this is the case, it follows that today’s systems of health care, nutrition, and education all contribute to the proliferation of a trait that previously was selection neutral or was selected against.

My granddad used to talk about somebody being ‘too smart by half.’ It seems a reasonable explanation for the rise of neocons.

18

Chris Stiles 04.14.07 at 1:56 pm

Nevertheless, it seems like a powerful social indicator; in my (anecdotal and wholly unscientific) experience tall people get an easier ride generally and are given more credence and authority in personal interactions.

There are a reasonable number of academic studies that show this to be the case. With the twist that what matters is your height at 16-18 rather than your adult height – the explanation being that being tall helps you think of yourself as a leader and helps you gain self-esteem.

19

Tom T. 04.14.07 at 2:23 pm

#16 & #17: True for women as well?

20

Ignacio Prado 04.14.07 at 2:26 pm

I think the “Cartesian” label–used by Chomsky and those influenced by him to describe the idea that certain cognitive capacities are innate–has always been unhelpful. For Descartes, our innate ideas were of things like God, infinity, etc. They were supposed to be concepts that a finite being–anything that was the product of contingent causal forces—could never grasp on his or her own without the help of divine implantation (the Third Meditation is supposed to be a “proof” of this).

The idea that evolution–or any other natural causal process– could vindicate anything as lofty as a Counter-Enlightenment notion of “human nature” that was insusceptible to progressive social improvement is decidedly NOT ‘Cartesian,’ at least if the term is being used in the attempt to capture a genuinely historical parallel.

[It should be noted that Chomsky is more historically self-conscious than others in constructing the parallel with Descartes. He has always emphasized that it is the language faculty’s capacity to be “infinitely generative”—i.e., to produce an infinite number of well-formed sentences given a finite set of syntactic rules and a finite lexicon—that it is difficult to account for using a behaviorist, stimulus-response psychology. He has also had, on occasion, the courage to say that he doubted whether evolutionary accounts of the origins of language could do the explanatory work required to explain THAT capacity of the language faculty. Unfortunately, those admissions of skepticism have occasionally caused people to cry “creationist.”

His most recent (and evolving) thoughts on this are, as far as I know, part of this jointly authored essay:

http://www.sciencemag.org/cgi/content/abstract/298/5598/1569 ]

21

Richard 04.14.07 at 3:12 pm

reply to #19, re #16: as far as I can tell (anecdotally and unscientifically), regarding leadership, yes. Possibly not in other areas. The issue of height being important for a model seems to be more a function of long proportion than eye-height, though that, too, may be a fruitful object of study.

22

abb1 04.14.07 at 3:26 pm

I got the impression (unscientific also) that tall people generally don’t care much about ‘leadership’; on the contrary – it’s short people who crave ‘leadership’ to compensate for a feeling of inferiority.

23

William Blurke 04.14.07 at 3:51 pm

Chomsky’s arguments as to the separation of animals from humans is Talmudic in origin, and both connect directly to his politics and sense of man’s capacity for reason.
He begins from an assumption (a faith).

24

William Blurke 04.14.07 at 4:33 pm

“Liberals constantly invoke Sweden as a governmental model without paying much heed to the fact that Sweden’s government succeeds as much as it does because it governs Swedes.”

I mentioned that to a Swedish cognitive psychologist over dinner last night. She said: “Well… duh.”

I’ll quote myself, from elsewhere on the same subject:

Social democracy was a fact before it was an idea. People who imagine otherwise are the sort that thinks somehow that they aren’t representative of a group: that they are “individuals.” But if you know anything about the history of the esthetic or intellectual “avant gardes” you know -or you should by now- that none of them invented anything, but were merely the first to see what already existed. They observed they didn’t “create.” But then people began to put the cart before the horse and the result was catastrophe.

Intellectuals who are engaged with culture engage with Burke. Literary novelists are Burkeans, in the sense that the prize observation and description over “creation.” Political liberal intellectuals in this country refer to anti-burkean models of culture in the inventions of science fiction and fantasy, and being rationalists, they can’t imagine that their intellectual models are as much products of history as their own invention. Henry’s defense of this logic is as simplistic as anything by Goldberg. I always ask what it means that the humanities have developed the linguistic tropes of the hard sciences. What is the historical significance of the concept of the “nerd.” What is “geek culture” if not historically determined?
History is the history of actions and consequences, not of intended consequences. It interests me that the logic of individualism that the left and even the civilized bourgeois used to disdain is now considered the basis of every debate in this country: on both sides. It’s a recent development. Why? What is libertarianism as an ideology (and a joke) other than a product of social atomization?
Why is Kos called a Progressive? Because after the turn to the right of American politics over the past 30 years the now mainstream right-wing had the habit of calling liberals “leftists” and now liberals believe it. And people reason from this logic as if it was always so. Atrios looked up to Bill Clinton. How is it possible to maintain even the shadow of a foundation when everything, including the meaning of words is shifting? Rationalism outside of historical context is a joke.

Goldberg isn’t that smart, but the people attacking him make their careers out of assumptions that they are rational actors of the moderate left.

“Rational actors of the moderate left.” Think about that one for a minute. Think about what it means.
More fun here and here

25

Barry 04.14.07 at 5:01 pm

Mr. Burke, I think that you have some thread navigation difficulties.

26

William Blurke 04.14.07 at 5:40 pm

The post begins with a reference and I was responding to that.
But you have a point.

27

eudoxis 04.14.07 at 7:27 pm

The Flynn effect is slowing down or reversing. An increase in height and earlier age of onset of puberty are still ongoing. That’s the latest, anyway.

28

Schwartz 04.15.07 at 12:19 am

“Debate over the relative influence of environment and heredity on intelligence has been going on for at least a century without much change or resolution”

Only someone who knows nothing about the relevant research in this area would ever say that. That being said, it’s better not to talk about this research openly. The public can’t handle the truth.

“This will be left as an exercise for readers (that is, I haven’t got around to working on it myself).”

Wait, are you doing a parody of Goldberg?

29

David Kane 04.15.07 at 1:05 am

Debate over the relative influence of environment and heredity on intelligence has been going on for at least a century without much change or resolution, or any obvious reduction in the level of vitriol. The only significant new information in the last few decades has been the discovery that average IQ scores have risen substantially over time (the so-called Flynn effect).

Riiight. No new information in “decades.” Ever read Gene Expression or Wikipedia?

30

John Quiggin 04.15.07 at 1:38 am

Rather than dark hints about information the public can’t handle, perhaps you’d like to spell out the vital new discoveries that have been made that apparently settle the issue once and for all.

The Wikipedia article spends lots of time on twin studies and includes estimates of heritability of adult IQ ranging from “less than 50 per cent” to “as much as 80 per cent”. With the exception of the Flynn stuff, it’s not obviously different from what would have been written on the same topic in, say, 1970.

31

novakant 04.15.07 at 1:59 am

the trouble with all of this is that it’s all unscientific bullshit, real scientists are very humble with their claims about the brain, for the simple reason that we still don’t know a lot about it’s functioning:

pointing to Broca’s area and saying that’s were our capibilities or language are located is neither a sufficient explanantion of language, nor how it’s processed within the brain;

nobody has yet come up with a scientific explanation of consciousness or learning on the brain level;

we know that the brain is supremely flexible in rebuilding capabilities lost due to injuries but we don’t have a clue how the brain does that;

we can point at the red areas of a PET scan all day, but without a comprehensive explanation of the underlying processes that is as much an explanation as Descartes claim that the pineal gland is the “seat of the soul”

so we’re still talking about input, output and a black box, of which we know very little, and that’s why all these claims are unscientific bullshit

32

David Kane 04.15.07 at 2:19 am

“The Wikipedia article spends lots of time on twin studies”

Well, yeah. Twin studies are important. You are quick to denounce Burt as a fraud for his twin studies. Do you think that these twin studies are frauds to? Consider this claim and citation from Wikipedia.

The Minnesota Study of Twins Reared Apart, a multiyear study of 100 sets of reared-apart twins which was started in 1979, concluded that about 70% of the variance in IQ was found to be associated with genetic variation.[6]

6. Thomas J. Bouchard Jr.; David T. Lykken; Matthew McGue; Nancy L. Segal; Auke Tellegen (October 12, 1990). Minnesota Study of Twins Reared Apart. National Institutes of Health / Science, Oct 12, 1990 v250 n4978 p223(6). Retrieved on August 6, 2006.

Now, if you think that a number like 70% heritability is reasonable (and/or thought this in 1990), then perhaps this research and much like it is not “significant new information in the last few decades.” If so, great! We agree.

Or perhaps I am just confused. What do you think that the heritability of IQ is?

33

aa 04.15.07 at 2:33 am

John – as I said, an unfamiliar notion.

34

John Quiggin 04.15.07 at 3:04 am

I didn’t give any number for heritability of IQ, but 70 per cent for white Americans seems like a midpoint estimate now, and has been so ever since I have been paying attention to this literature, which is many decades. As far as I can see the range around this number hasn’t narrowed much over time. Obviously neither has the extent to which people get upset about it, nor the extent to which both sides of the nature-nurture debate are prone to premature claims of victory (see Galton, Watson, Skinner, Eysenck, Bell Curve etc etc).

What is new is the clarification (via the Flynn effect) that heritability doesn’t mean what most people would understand by it, since it doesn’t take account of changes in the environment over time, which turn out to be at least as important as cross-section environment differences (again, for white Americans).

aa – wtf^2?

35

theogon 04.15.07 at 3:06 am

John – as I said, an unfamiliar notion.

Uh, quite.

36

John Quiggin 04.15.07 at 3:21 am

Belatedly responding to susanc, I agree with much of what you said.

Only one response. I think the cultural component in the craziness of McVeigh and OBL is pretty obvious, and AFAIK, there’s no evidence of any familial predisposition to such things.

37

Matt Weiner 04.15.07 at 3:25 am

David Kane: “You are quick to denounce Burt as a fraud”

*snicker*

38

Darth Quixote 04.15.07 at 4:03 am

It is true that the correlation between height and IQ is about the same magnitude as the correlation between head size and IQ. However, unlike the latter, the correlation between height and IQ does not hold true within families. That is, the sibling with the higher IQ does not tend to be taller (whereas the sibling with the higher IQ does tend to have a larger head). The largest study bearing out the null within-family relation between height and IQ includes over 16,000 sibling pairs.

Some studies have looked for a genetic correlation between height and IQ. Two recent studies have gone opposite ways here.

The balance of the evidence strongly favors no functional relationship between height and IQ. (This poses a contrast to the case of head size and IQ. The correlation doubles when it is actual brain volume that is assessed through magnetic resonance imaging; head size is an attenuated proxy for brain size. Also, when whites and blacks at age 7 are matched for true-score IQ, the difference in average head size between whites and blacks completely disappears, whereas some gap in IQ remains when whites and blacks are matched in head size. This is what you would expect if sheer volume is merely one of many properties of the brain that causally determine IQ.) The most likely cause of the population correlation is cross-assortative mating for these two traits, both of which are highly valued in our society. A recent New York Times article about speed dating reported some very persuasive data showing that women are willing to trade off height against income in prospective dating partners. (A 5’1” guy has to make something like $500K before a girl will go out with him.) Going the other way, men value leg length in women. Thus, it must inevitably be the case that gametic disequilibrium among loci affecting height and IQ is created by the assortment of the enhancing alleles for both traits into the same families. (I mentioned the preference for leg length because it might explain the otherwise odd fact that the correlation between height and IQ is accounted for largely by sitting height.)

As for the parallel secular increase in both traits, it is reasonable to suspect common causes of both. However, I do not believe that any convincing evidence has been put forth to explain what these common causes might be. Given the high heritability of both traits and the absence of a common-environment variance component for IQ (at least for the middle 80% of the population), promising culprits are changes in the environment that have affected more or less the entire population (e.g., inoculations against once-common diseases).

I hope this sheds some light on the nature of the relationship between height and mental ability.

39

aa 04.15.07 at 4:04 am

On the BBC at the moment:

The cause of high blood pressure may lie within the brain, rather than with problems relating to the heart, kidneys or blood vessels, research suggests.

I’ll save you the trouble – wtf^3. But you’re not really trying.

40

Darth Quixote 04.15.07 at 4:11 am

Also, it must remembered that some part of the Flynn Effect cannot be accounted for by true gains in the latent ability factors measured by IQ tests. In the lingo of structural equation modeling, the same measurement model cannot be fit to different cohorts manifesting the Flynn Effect. Some kind of measurement artifact is causing the intercept of the regression of observed scores on ability factors to drift upward.

41

Alan Kellogg 04.15.07 at 4:27 am

aa,

Have you seen the report the BBC refers to? Have you had any chance to study it? What gives you the confidence to dismiss it? I’m serious, how do you know the researchers involved are wrong?

42

steve 04.15.07 at 4:34 am

John,

You may have accepted the 70% heritability figure decades ago, but I think that many social scientists still have trouble accepting it. (Or, perhaps more precisely, prefer not to think about it and its possible implications.)

Regarding the Flynn effect, as far as I understand the interpretation is controversial. In many of the cases studied the raw score increase coincides with a time period when average number of years of schooling increased considerably. It’s not surprising that conscripts (much of the data covering long periods is from military intelligence tests) who left school at 10 or 12 might not reach their full potential, e.g., compared to those that had a full K-12 education. In other words, I don’t think the Flynn effect suggests that intelligence is massively more malleable than previously suspected. It also appears that the effect may have saturated in the West, very possibly coinciding with the onset of universal K-12 education and elimination of nutritional deprivation. (Note: the Flynn effect is most pronounced for very abstract tests like Raven’s progressive matrices, and there it seems that exposure to similar types of reasoning might distort the correlation between the Raven’s score and overall intelligence. More general IQ tests don’t show as large a Flynn effect.)

Finally, once you accept heritability you at least accept the *possibility* of group differences in intelligence based on different distributions of genes in different populations. While the particular genes that influence intelligence are not yet known, it *is* known (due to recent advances in genomics) that there is significant clustering of gene distributions according to geographical (e.g., continental) origin. (i.e., alleles which are common in one group can be very uncommon in another.) Certainly, no one doubts the reality of group differences in the genetic component of height, which is about as heritable as intelligence in the twin studies.

43

John Quiggin 04.15.07 at 5:15 am

“In many of the cases studied the raw score increase coincides with a time period when average number of years of schooling increased considerably. It’s not surprising that conscripts (much of the data covering long periods is from military intelligence tests) who left school at 10 or 12 might not reach their full potential, e.g., compared to those that had a full K-12 education.”

Or, indeed, those who had a university/higher technical education. An obvious implication of all this is that the proportion of people capable of benefiting from post-secondary education is much higher than was widely supposed in the past.

44

Hidari 04.15.07 at 7:40 am

Is David Kane the only person in the world who doesn’t know psychology’s worst kept secret…that Bouchard’s research is…er…how can I put it….well put it this way….that many people suspect he is following in Burt’s footsteps in more ways than one?

Note to Chris (comment 15)…of course Chomsky is also a (secularised) Platonist……(c.f Plato’s problem). This isn’t disagreeing with you, just pointing out that Pinker et al simply ignore all the bits of Chomsky’s thought that they don’t like.

45

Mthson 04.15.07 at 9:16 am

Re: Flynn. There’s evidence the Flynn effect was predominately in the lower half of the bell curve. (http://en.wikipedia.org/wiki/Flynn_Effect#_ref-Colom2005_0)

Re: #44. The results of Burt’s studies of IQ in terms of IQ heritability are identical to the most common modern estimates. His conclusion was around 70%, same as the conclusion of the APA’s report on intelligence following the Bell Curve controversy.

46

conrad 04.15.07 at 9:44 am

Re the 70% heritability figure and the progress of research in the field over time, I came of age in the 70’s, and I recall quite well that the general consensus on the heritability of intelligence then was at about 40%, not 70%. It seems to have gone up dramatically since then, due to better studies. So whoever is claiming that this figure was 70% back then seems full of it to me. That’s a 75% increase in raw numbers, which seems highly significant to me. If someone thought back then that it was 70%, they’d have been dismissed as a loon, which may be just what this guy is.

47

Matt McIntosh 04.15.07 at 10:34 am

Re: #44,

Insinuations of scientific fraud are not to be bandied about lightly. Either make a case or stop polluting the discourse.

48

John Quiggin 04.15.07 at 11:26 am

Eysenck and Jensen were claiming 80 per cent at the time, so it wasn’t much of a consensus.

49

Hidari 04.15.07 at 11:32 am

‘Re: #44. The results of Burt’s studies of IQ in terms of IQ heritability are identical to the most common modern estimates. His conclusion was around 70%, same as the conclusion of the APA’s report on intelligence following the Bell Curve controversy.’

Ha!

May I recommend Michael Howe’s excellent ‘IQ in Question’ (Sage 1997) for a fair discussion of this and many other IQ related issues.

50

Tracy W 04.15.07 at 11:46 am

hidari (comment 11) – Pinker’s ideas on linguistics derive from Chomsky, okay, I’ll take your word from it. Why does that mean that there’s any sort of contradiction based on Chomsky’s views on Cartesianism (be that modern, secular, materialist)? A simple explanation may be that Pinker isn’t into Cartesianism, regardless of Chomsky’s views on the matter.

51

Teddy 04.15.07 at 12:26 pm

If it’s really true that high heritability was already accepted in the 70s, how can we explain the fact that a group of very distinguished scientists (which included Francis Crick, Otis Dudley Duncan, Jacques Monod and Paul Meehl) issued a warning in 1972 about “suppression, censure, punishment and defamation that are being applied against scientists who emphasize the role of heredity in human behavior”. They said: “it is virtually a heresy to express a hereditarian view, or to recommend a further study of the biological bases of behavior. A kind of orthodox environmentalism dominates the liberal academy, and strongly inhibits teachers, researchers and scholars from turning to biological explanations or efforts.”

52

Hidari 04.15.07 at 1:27 pm

‘Pinker’s ideas on linguistics derive from Chomsky, okay, I’ll take your word from it.’

Pinker’s ideas on linguistics ARE Chomsky’s ideas. The ONLY difference is that Chomsky has always been sceptical of evolutionary explanations for the Language Acquiring Device (or module or faculty or whatever you want to call it) and Pinker isn’t. *

The other point is that Chomskyan linguistics aren’t simply a scientific thesis that ‘happen to have’ a Cartesian ‘feel’ (i.e. when explained by Chomsky). They are, in their essence, Cartesian (i.e. Descartes-as-seen-by-Chomsky…whether Descartes, if he were alive today, would like or approve of Chomsky’s ideas is a whole other ball game): e.g. they presuppose an ‘inner’ and ‘outer’ world (inner being subjective, outer being objective), they presoppose Rationalism (NOT Lockean empiricism…indeed it is Lockean empiricism (which Pinker calls the ‘blank slate’) idea which both Pinker and Chomsky argue against most vehemently), and so forth.

You can’t just say ‘I approve of Chomsky’s theories except the Cartesian bit’: that really would be to ‘throw the baby out with the bathwater’.

*Giving this some thought this may not be strictly speaking, true. One of the more tedious aspects of Pinker’s over abundant academic output is his tendency to simply ‘add’ bits to his general theories without seeing whether the whole ‘infrastructure’ holds together. For example, he now expresses (qualified) admiration for George Lakoff, and quotes from ‘Metaphors We Live By’ etc. But Lakoff and Johnson were attempting to argue AGAINST Chomskyan linguistics. You can’t just chop bits of their theories off and then stick them onto Chomskyan theory and say ‘look at the wonderful synthesis I have created’ when all you really have is an incoherent mess.

All this and more will be explained in my forthcoming book, zooming to a book shop near you in September and bargain bins round the world in October.

53

Matt McIntosh 04.15.07 at 2:54 pm

Hidari, I’m still waiting for a retraction or an actual argument for your insinuation that Bouchard has committed some sort of scientific fraud (comment #44). If you don’t support it or retract it I don’t see why anyone is obligated to take you seriously, and I would be very disappointed in the Crooked Timber bloggers if they allowed such near-libelous antics on their site.

54

Mthson 04.15.07 at 4:20 pm

Re:Hidari #49. It sounds like your and Michael Howe’s opponent in this debate is the mainstream psychological community, represented in my comment by the APA’s 1995 report on intelligence. That’s fine, but Howe’s work is, by his admission, a minority/fringe opinion, and is not an attempt at a “fair” review, in the sense of taking into account both sides – or even the mainstream POV – on an issue.

55

zdenek v 04.15.07 at 5:20 pm

# 11 and # 52 ( Hidari on Pinker ). Some confusion here . Pinker’s rejection of Cartesianism involves rejection of Descartes’ Dualism and all the problems associated with this position. ( remember that dualism involves immaterial substance which cannot be fitted into naturalistic account of the mental phenomena ).

To the extent that Chomsky is sympathetic to some sort of Dualist view of mind Pinker parts company with Chomsky.

So to that extent it is perfectly possible and defensible to reject that part of Chomsky outlook which doesn’t sit comfortably with cognitive psych . which is strongly materialist ( mind is either identical to brain or supervenes on it ).

56

zdenek v 04.15.07 at 5:49 pm

Hidari writes :–

“The other point is that Chomskyan linguistics aren’t simply a scientific thesis that ‘happen to have’ a Cartesian ‘feel’ (i.e. when explained by Chomsky). They are, in their essence, Cartesian ”

This is fine as far as it goes but still somewhat muddled. You need a distinction ,when discussing Descartes’, Chomsky’s or Pinker’s views on this subject’ between *metaphysical* thesis these thinkers are embracing and a *psychological* thesis which provides an account of the mechanisms involved in cognition.

Now Pinker and Chomsky see eye to eye with Descartes with respect to the sort of mechanism involved in cognition but Pinker parts company with both as far as metaphysics go because he rejects Dualism.

Btw it is also Chomsky’s sympathies with Dualism which underwrites his rejection of evolutionary accounts of language.

57

ragadaisus 04.15.07 at 5:54 pm

Re 46: A figure of 70-80% for heritability was accepted and unremarkable into the early 1970s. Then, in response to Jensen, there was a stream of literature with outlandish statistical acrobatics designed to deny heritability. That faded, thank goodness, and we are back to 70-80% for adults (lower for children).

I would like to endorse #47 and #53: it is not appropriate on a list like this to make such wild accusations, especially while hiding it behind innuendo, and most especially when the backup citation is admittedly fringe.

58

Darth Quixote 04.15.07 at 6:07 pm

I second Matt’s call for an explanation from Hidari as to why the casual comment about Bouchard is not a scurrilous bit of libel. And, moreover, for something more substantial than the invocation of an “authority.” This deserves a description of documentary evidence or internal analysis of the data that Bouchard and his (numerous) colleagues have reported.

59

zdenek v 04.15.07 at 6:08 pm

Chris at # 15 :–
“Given Pinker’s obsession with Evolutionary Psychology in the Cosmides/Tooby/Buss mode, it seems odd for Pinker to be complaining about Cartesianism. Next thing ya know, he’s going to say that cognitive science is too Platonic, and we should instead all be nativists.”

Cartesianism has a metaphysical component ( dualism ) and a psychological component ( description of the mechanism which we call mind ).

As far as Pinker’s rejection of Cartesianism goes that is now easy to understand : he is rejecting for obvious reasons ( you cannot acommodate immaterial mind within scientific picture ) dualism but not the other component of Cartesianism in so far as he sees mental capacities as consisting of prewired modules that are largely innate.

60

aa 04.15.07 at 7:01 pm

Alan K – Nothing was “dismissed” by my comment.
You may be confused by the “wtf^3”, which is simply me channeling John as he dismisses the remark (or its relevance).

I’m sorry to disappoint you, but I’m afraid I agree with you on this particular point.

61

tps12 04.15.07 at 10:40 pm

I love the mention of the height thing. I realized in a recent argument online that the two sides of the nature/nurture debate on intelligence boils down to “intelligence is like height, it’s almost completely determined by genetics” vs. “intelligence is like height, it’s to a great extent determined by environment.” The vitriol is so high because we argue from different premises.

62

David Kane 04.15.07 at 10:44 pm

John writes:

I didn’t give any number for heritability of IQ, but 70 per cent for white Americans seems like a midpoint estimate now, and has been so ever since I have been paying attention to this literature, which is many decades.

I apologize for assuming that we would disagree about this. You and I think that (approximately) 70% heritability is a scientific fact. Do all your fellow CT authors agree? I doubt it. In fact, I would wager that a majority of them would disagree strongly. Let’s ask them!

As far as I can see the range around this number hasn’t narrowed much over time. Obviously neither has the extent to which people get upset about it, nor the extent to which both sides of the nature-nurture debate are prone to premature claims of victory (see Galton, Watson, Skinner, Eysenck, Bell Curve etc etc).

I am confused again. Once you grant (as you seem to) that IQ is a meaningful construct, that it predicts all sorts of important life outcomes and that it is highly heritable, what aspects of the Bell Curve do you disagree with? (Let’s leave aside the racial chapters for now.) Just considering white americans, every factual claim made by the Bell Curve seems to be something that you agree with. So why bad mouth the book?

63

Alan Kellogg 04.15.07 at 11:41 pm

aa,

Oh poo. And here I was hoping for a knockdown, drag out, knuckle chomping brawl. :(

As for IQ testing. Maybe one day we’ll have a reliable method not subject to observer bias, participant vagaries, and cultural weighting, but I’d sooner rely on the timely arrival of a haelthy puppy from a Tijuana puppy mill.

64

tim B 04.16.07 at 12:23 am

53,58
It’s so incredibly laughable what you two (Matt McIntosh and Darth Quixote) are demanding from Hidaris. Both you of you are associated with GNXP where repeated charges of fraud and dishonesty have made about Jared Diamond, Stephen Jay Gould, etc. by GC and others and at no point in time did any of you made any demands for substantiations or retractions. Stop being hypocrites.

65

blah 04.16.07 at 1:26 am

repeated charges of fraud and dishonesty have made about Jared Diamond, Stephen Jay Gould, etc. by GC and others and at no point in time did any of you made any demands for substantiations

I don’t think you realize this, but there was quite a lot of substantiation for the fact that Jared Diamond is a fraud and that Stephen Jay Gould was wrong. They have original links to the paper in which Jared Diamond talked about how important it was to investigate why African testes were larger than European testes, which in turn were larger than Asian testes.

In contrast, Hidaris offers only baseless calumny.

66

John Quiggin 04.16.07 at 1:58 am

While the flow of vitriol certainly demonstrates my point, I’m going to call a halt now on accusations of dishonesty by partisans on one side or the other. Anything further along these lines will be deleted.

67

steve 04.16.07 at 2:27 am

John,

I’m still interested to know what fraction of CT regulars accept the 70% heritability figure, in the case where there is not severe environmental deprivation.

I’m not as sure as David that you also accept the utility of the IQ/g/general intelligence construct. Is that the case? I can already tell from the comments that at least some of your readers don’t.

Much of The Bell Curve had nothing to do with race and everything to do with how a quantity like IQ, which is largely heritable and readily measureable from a very short exam, could statistically predict a lot of scholastic as well as life outcomes.

All the ideological baggage prevents people from appreciating that IQ as a psychometric is almost like magic in its utility. We use it all the time to apportion scarce resources like elite education. Certainly there are weaknesses in this process, but it’s just amazing to me that SAT score alone is about as predictive of college performance as combined high school grade information. If you think about that, it means a 3 hr test yields as much information as the combined opinions of all your teachers throughout high school! (Please don’t respond that SAT just measures social class or family income. Affluent and poor kids come with the whole range of IQs, and it’s the IQ that is predictive, not the family background.)

68

John Quiggin 04.16.07 at 2:57 am

As mentioned above, I’m happy to accept 70 per cent as the best estimate of heritability of IQ test scores for any given cohort of white Americans, bearing in mind that
(i) there’s a range of estimates around this, and the range hasn’t narrowed greatly over time
(ii) heritability is a technical term regarding the attribution of phenotypic variance in a given population at a given time, which does not mean anything like what is commonly assumed – the Flynn effect being just one illustration of this

I don’t have a firm view one way or the other on the validity of the g/general intelligence construct – I haven’t looked hard at it. The expositions I’ve seen relied on a rather naive reification of the outputs of factor analysis – maybe you can point me to something more sophisticated.

I agree that IQ or SAT tests are good predictors of university performance. In fact, I’d say that these tests are best viewed as measures of academic aptitude and that this explains why median members of past cohorts who were not obviously stupid nevertheless did much worse on IQ tests than median members of later cohorts. Even in the absence of severe environmental deprivation, it seems plausible that the environment has become more favorable to academic skills as school completion, and then university attendance became standard expectations.

As regards the Bell Curve, my big beef is that they just ducked the issue of the Flynn effect, by saying they were concerned only with current inequality. I made this point here.

As I said then, this is like ignoring twin studies because most people aren’t twins.

69

ragadaisus 04.16.07 at 3:26 am

In #68 John Quiggin says
>I agree that IQ or SAT tests are good predictors >of university performance. In fact, I’d say that >these tests are best viewed as measures of >academic aptitude

But they are so widespread and so controversial because they work and work very well for almost everything under the sun, like how long you will live, whether you will have an auto accident, how much you and your lathe will produce on my assembly line, and so on. This is why the topic is interesting: if it were a narrow measure of academic aptitude no one would care about it except professors.

A good summary is

Robert Gordon, “Everyday Life as an Intelligence Test”, Intelligence 24:203-320, 1997.

70

steve 04.16.07 at 3:28 am

John,

I think I agree with your opinions (modulo perhaps the precise interpretation of the Flynn effect), but I suspect you are at odds with a lot of CT readers!

Regarding IQ as a predictor of non-academic outcomes, there is a lot of research (e.g., by psychologist Linda Gottfredson) on this. IQ correlates with many things which are not academic, like the chances you’ll be killed in a car crash, your ability to quickly navigate a restaurant menu, etc.

The US military found that below a certain threshold (like 85) troops are significantly less trainable and not effective as combatants. They received a special dispensation from Congress to filter by IQ (Armed Forces Qualifying Test (AFQT)), the only US organization to be exempt from Griggs v. Duke Power. Goldman Sachs, Microsoft and Google would love to have this dispensation :-)

71

David Kane 04.16.07 at 3:57 am

1) All the opinions expressed by JQ above seem reasonable. But are they shared by many of the other authors (forget about readers) of CT? I doubt it. Clarification welcome.

2) The key point of the Bell Curve with regard to the Flynn Effect is that, ignoring the chance of some miracle discovery, there is no evidence (nor has there been in the last 10 years since publication) that we can change the IQs of large numbers of people. Those IQs may change, but if the reasons are beyond are control, then it doesn’t much matter to the analysis in the Bell Curve.

3) Anyway, I still feel that I am in some Bizarro World CT. Most people who buy into the whole IQ/hereditary Bell Curve view (leaving aside the racial stuff) have a radically different view on social policy that that which is common (I assume!) among CT authors and readers. What does JQ think about, for example, affirmative action in university admissions?

72

blah 04.16.07 at 4:01 am

the only US organization to be exempt from Griggs v. Duke Power. Goldman Sachs, Microsoft and Google would love to have this dispensation :-)

Indeed, though the primary workaround today are academic pedigree (credentialism) plus proxies like this:

http://news.com.com/2100-1023_3-5263941.html

The main difference is that there is the fig leaf that such tests are “directly related to work on the job”. Of course they are, insofar as IQ is directly related to work on the job, but this is an end run around Griggs vs. Duke Power.

73

rikurzhen 04.16.07 at 5:47 am

John Quiggin wrote (i) there’s a range of estimates around this, and the range hasn’t narrowed greatly over time

This is mistaken. The appearance of a range of estimates is due to fact that the heritability of IQ is a function of age. According to the APA report (Neisser et al. 1996), If one simply combines all available correlations in a single analysis, the heritability (h2) works out to about .50 and the between-family variance (c2) to about .25 (e.g., Chipuer, Rovine, & Plomin, 1990; Loehlin, 1989). These overall figures are misleading, however, because most of the relevant studies have been done with children. We now know that the heritability of IQ changes with age: h2 goes up and c2 goes down from infancy to adulthood (McCartney, Harris, & Bernieri, 1990; McGue, Bouchard, Iacono, & Lykken, 1993). In childhood h2 and C2 for IQ are of the order of .45 and .35; by late adolescence h2 is around .75 and c2 is quite low (zero in some studies). Substantial environmental variance remains, but it primarily reflects within-family rather than between-family differences.

74

Hidari 04.16.07 at 7:28 am

In response to some of the more excitable commentators above (and in case any lawyers are reading) I, personally, have absolutely no opinions as to the ethical and ontological status of Bouchard’s data.

However, (and maybe I shouldn’t have passed this news on, although I’m not the first to bring this to the public’s attention) it is an objective fact that I have personally heard such issues raised. It’s not my imagination. I didn’t make it up. I really have heard these rumours. It’s also true that lots of other people have (in print, John Horgan and Oliver James have alluded to them).

OK so mea culpa maybe I shouldn’t have said it.

But here’s the key point. Bouchard could scotch all these rumours tomorrow, but he won’t because he won’t release his ‘raw’ data .

Now all I can say about this, with the laws against libel being what they are, is that I feel that Bouchard has made a mistake here. At the very least it would be common scientific courtesy to allow other researchers to access your raw data, don’t you think? And I will point out something else. ‘Human nature’ (!) being what it is, other people are going to continue to find it slightly mysterious that Bouchard refuses to release his raw data, in spite of repeated requests to do so.

Sorry but blame the ‘suspiciousness’ gene we doubtless all possess.

(Please note, I didn’t see John Quiggan’s note above, but just to restate: I’m not accusing anyone of anything, and I’m not going to reply to anything that assumes that I did).

75

abb1 04.16.07 at 7:32 am

Blah (65), I read your Jared Diamond post and I don’t see how it justifies the “Jared Diamond is a fraud” conclusion or its relevance to the discussion of cognitive stuff. Could you clarify, please.
Thanks.

76

Hidari 04.16.07 at 7:32 am

‘Cartesianism has a metaphysical component ( dualism ) and a psychological component ( description of the mechanism which we call mind ).’

Well that’s certainly how we see things now, although it’s not how Descartes saw it. But this presupposes that Cartesianism splits easily and cleanly into two easily separable ‘parts’, the ‘metaphysical’ and the ‘scientific’. But it’s not at all clear that that’s the case, although Pinker assumes it is.

In any case, Pinker doesn’t make this distinction. Pinker rails against Cartesianism in its entirety (my understanding is), not just the ‘metaphysical’ aspect, or at least, he fails to make clear what parts of Descartes’s thought he is accepting and rejecting.

77

John Quiggin 04.16.07 at 8:39 am

To clarify, no more discussion of fraud, including meta-discussion etc, please.

78

daniella 04.16.07 at 9:03 am

The primary observations of variation need extension – for example, do Yorubas have large testes as well as frequent dizygotic twins?

Wow. Wow. Jared Diamond wanted to take calipers and measure the balls of people of color. Wow. what next, we find out that Martin Luther King was a KKK member.

79

zdenek v 04.16.07 at 9:40 am

Hidari writes:

“In any case, Pinker doesn’t make this distinction. Pinker rails against Cartesianism in its entirety (my understanding is), not just the ‘metaphysical’ aspect, or at least, he fails to make clear what parts of Descartes’s thought he is accepting and rejecting.”

No this is not right. Pinker makes it clear that he is opposed to Descartes’ dualism ( i.e. his metaphysics of mind )when he uses pejoratively the ‘ghost in the machine’ slogan. This is standard criticism functionalists like Pinker ( same criticism made by Dan Dennett whose more technical dismantling of Cartesian dualism Pinker endorses ) make of dualism.

But even if Pinker wasn’t as clear as he should, it should be obvious why and what component of Caretesianism is being rejected because the entire 20th century effort in philosohy of mind to naturalize the mental is deeply anti dualist and hence anti Cartesian. Pinker is part of this broader intellectual effort ( to naturalize mind ).

This is why your criticism of Pinker is not only uncharitable but strikes me as made by someone who is not familiar with recent history in this area.

80

abb1 04.16.07 at 9:44 am

No, no discussion of fraud; I’m just curious what variations in testicles size among different populations have to do with cognitive abilities. After all, variations in skin pigmentation are plain obvious, so how’s the testicles article is earth-shaking?

I’m not a scientist, so I’m probably missing something.

81

zdenek v 04.16.07 at 10:00 am

re # 76 . Another little clarification . It is possible to be an innatist ala Descartes with respect to say language aquisition but embrace methodological and metaphysical naturalism and hence part company with both Descartes and Chomsky.

You seem to think that this is somehow inconsistent but as I point out this confuses metaphysics with psychology.

82

zdenek v 04.16.07 at 10:46 am

Hidari earlier on said :–

“In other words, Chomsky always explicitly based his opposition to Empiricism and Behaviourism on his Rationalist Cartesianism, and, in fact, so does Pinker.”

Yes but note that Chomsky unlike Pinker is sceptical about the posibility of naturalising that is providing Darwinian account of mind. Why is that ?

Answer to this is the explanation why and where Pinker and Chomsky part company. The answer is Chomsky is a dualist ( either substance or property dualist ).Pinker rejects dualism in favour of some sort of physicalist functionalism.( mind is either identical to brain or function of brain that supervenes on the brain )

More important though is that this is not just a dispute between Chomsky and Pinker but rather between Chomsky and modern phil of mind /cognitive science.

That is, it is Chomsky’s view which is the weird one because note if one cannot provide an evolutionary explanation of mind we are left with
two posibilities ( Descartes wouldnt mind ):

a) supernatural origin : mind created by God and somehow implemented inside our bodies. ( Roman Catholic view )
b) mind cannot be explained at all ( lovely obscurantist view ).

83

John Quiggin 04.16.07 at 10:55 am

Anecdotal response to #70: As someone who scores pretty well on IQ tests, I am keen to believe I would be good with a lathe. I doubt that my high-school woodwork teacher would share this expectation, though – I still have the occasional table he made for me after despairing of my attempts. OTOH, I am definitely among the 80 per cent of the population who are better-than-average drivers, so the prediction regarding car crashes seems plausible to me.

84

steve 04.16.07 at 3:24 pm

But that’s the thing: IQ has predictive power in areas far from the expected bookish pursuits. Take a group of machinists or production workers. Give them each a 20 minute “Wonderlic” test (a simple IQ test, also used by the NFL on potential draft choices). Voila! A correlation between individual productivity (output) and IQ. (It’s only a correlation — sure, some future professors like me were terrible at woodshop.)

It almost makes you want to believe in “g”! :-)

From Gottfredson:

For example, a 1969 study done for the U.S. Army by the Human Resources Research Office found that enlistees in the bottom fifth of the ability distribution required two to six times as many teaching trials and prompts as did their higher-ability peers to attain minimal proficiency in rifle assembly, monitoring signals, combat plotting and other basic military tasks.

Aren’t you *happy* that Griggs vs Duke Power doesn’t apply to the military (despite the disparate impacts on different racial groups)? How about your local power utility? Would you like it if they could improve efficiency dramatically by administering little tests to all their job applicants?

85

Pablo Stafforini 04.16.07 at 3:47 pm

Hidari writes:

Pinker’s ideas on linguistics ARE Chomsky’s ideas. The ONLY difference is that Chomsky has always been sceptical of evolutionary explanations for the Language Acquiring Device (or module or faculty or whatever you want to call it) and Pinker isn’t.

Chomsky is not sceptical of evolutionary explanations of the language faculty; his scepticism is limited to adaptationist explanations of such faculty. See his recent article with Hauser and Fitch, and also his earlier reply to Maynard Smith in the New York Review of Books.

86

Kevin Donoghue 04.16.07 at 3:52 pm

Does the marshmallow test continue to predict academic ability better than IQ tests?

87

steve 04.16.07 at 4:09 pm

There is actually good evidence for a personality component you might call conscientiousness (related to the ability to put off eating a marshmallow, I suppose) which is more or less independent of IQ. High school grades probably measure a mixture of IQ and conscientiousness, whereas the SAT is more correlated with IQ. When you combine grades + SAT you get improved prediction of college performance, although by themselves grades and SAT are roughly of equal value.

The point about IQ is that it seems to be a principal component that pops out when you look at a large range of cognitive tests — it correlates with all the other more narrowly defined cognitive abilities. This didn’t have to be the case — in that sense the result is a bit surprising. The 20 minute Wonderlic test accesses some deep aspect of your cognition (“general intelligence”) that is then indicative of more specialized capabilities (visualization, verbal ability, memory, etc.).

I don’t fully believe in g: there are people whose narrow capabilities are all over the map and not very well-indicated by g. But if you average over populations then it’s not unreasonable.

88

abb1 04.16.07 at 4:23 pm

I find it hard to believe that higher IQ translates into better driving or crane-operating. I mean, we’re talking about the ability to analyze abstract concepts here. I mean, you concentrate on analyzing abstract concepts while driving – you’re gonna miss the moment when the driver in front of you hit the brakes. Seriously. I’d definitely prefer a chauffeur with 80 IQ to one with 140.

89

steve 04.16.07 at 5:07 pm

abb1:

It may be counterintuitive, but there it is in the statistics. Perhaps there is more to IQ than the abstract component you are thinking of? Perhaps it also correlates with a construct we might call “general intelligence”?

A chauffeur with IQ of 80 might have a hard time following your instructions, figuring out traffic patterns, negotiating with doormen and parking attendants, maintaining the car according to the manual, etc. Granted, someone with an IQ of 140 might not *want* to be your chauffeur, but if we suppose that for some reason (personality quirk) they were enthusiastic about the job, I am sure that *averaged over populations* they would outperform the IQ 80 crowd.

90

abb1 04.16.07 at 5:49 pm

I used to work with a couple of Brazilian fellas from Rio. One was a run-of-the-mill programmer, the other was regarded as a genius, extremely smart, he later opened his own company, developed some voice-recognition something and settled in the US for good. The average guy was suffering from nostalgia all the time, complaining about quality of American meat and eventually moved back to Rio.

Now, here’s the punchline. Living in Rio, Navis (the smart one) had been attacked 18 or 19 times, while the average-intelligence guy hasn’t been assaulted or robbed even once, ever. See, the average guy had the common sense to watch what’s going on around, to cross the street or turn around if necessary. High IQ doesn’t help with that, quite the opposite. You know: the so-called street-smarts.

91

steve 04.16.07 at 6:17 pm

abb1:

Sure, there are absent minded geniuses. What happens when you average over the entire IQ 130 population and the entire IQ 80 population? Then who has more trouble with street crime?

As I wrote above:
I don’t fully believe in g: there are people whose narrow capabilities are all over the map and not very well-indicated by g. But if you average over populations then it’s not unreasonable.

92

ragadaisus 04.16.07 at 6:24 pm

Re the marshmallow test: it is perhaps not surprising that a test of time preference at age four is predictive. IQ tests for that age are not worth very much. Time preference is a very strong correlate of IQ and of social class. A good read is Greg Clark’s “Genetically Capitalist”, a chapter from a forthcoming book that is all over the web, proposing that medieval evolution in England created a new kind of person. One of his big and interesting pieces of evidence is the time course of the interest rate, which mostly reflects time preference.

93

abb1 04.16.07 at 8:18 pm

It’s not just absentmindedness, often it’s also severe lack of social skills, shyness/overconfidence/overanalyzing and stuff like that. We all have seen it, I’m sure, and maybe even suffered from it. Nothing comes without a price.

94

ragadaisus 04.16.07 at 8:33 pm

re #93: “Nothing comes without a price.”

But this is exactly why IQ is such a fascinating puzzle: there is apparently no price (except, perhaps, for myopia, maybe gout). The literature directly contradicts your impressions: high IQ predicts more not less athletic ability for example.

95

John Quiggin 04.16.07 at 11:27 pm

The absence of a price is a big problem for an evolutionary explanation of IQ differences. Evolution is supposed to optimise fitness.

96

Darth Quixote 04.17.07 at 12:12 am

The absence of a price is a big problem for an evolutionary explanation of IQ differences. Evolution is supposed to optimise fitness.

Humans are not at steady state. Think about that. Deeper than you think.

97

John Quiggin 04.17.07 at 12:46 am

I have thought about it, and I don’t think the absence of a present steady state helps much. Unless IQ gains are costly, where is the genotypic variation coming from? Why didn’t selective pressure push IQs up to 150 (or whatever upper bound you want to nominate) in our ancestral populations and eliminate all the genes that could reduce it?

98

David Kane 04.17.07 at 1:02 am

There is no free lunch in optimization, genetic or otherwise. I know that JQ doesn’t think that much useful research has been conducted in “decades,” but has he considered “The Natural History of Ashkenazi Intelligence” by Cochran et al.

Basic idea is that the Ashkenazi were under (greater) selective pressure and so their IQ rose faster. But, along with that rise came various genetic diseases, some quite serious.

Sure, if there was some magic button to press with no side effects which increased IQ then you might expect evolution to press that button. But there are always side-effects. A simple example is that increased head/brain size both increases IQ and (I think) increases maternal mortality. Evolution has to confront these trade-offs. The fact that we are all not 150 IQ is no more relevant to selection pressures on IQ then the fact that we are all not 6’6” (or super muscular or insert-your-favorite-desirable-genetic-trait) is relevant to selection pressures on height.

99

John Quiggin 04.17.07 at 1:15 am

DK, I didn’t say no useful research had been conducted, I said nothing fundamentally new had been found. The discussion over at GNXP confirmed this. After a lot of spluttering, the claim was that we now have more and better data of the same general kind that we had in 1970, not that I had missed something big.

More generally, your point seems to be directed at #94 and #96, not at me. It’s obvious there must be trade-offs – my question in #97 was of the rhetorical variety.

100

Pithlord 04.17.07 at 1:40 am

I think both Chonsky and Pinker’s metaphysical views are being oversimplified. Chomsky says somewhere that he is a materialist (I could dig it up). Pinker is a mysterian about consciousness/ qualia, much to the disgust of Dennett et al.

101

razib 04.17.07 at 2:37 am

Unless IQ gains are costly, where is the genotypic variation coming from? Why didn’t selective pressure push IQs up to 150 (or whatever upper bound you want to nominate) in our ancestral populations and eliminate all the genes that could reduce it?

IQ is a quantitative trait, and it is likely controlled by many loci, right? well, antagonistic pleiotropy & correlated response is one issue. selection on a quantitative trait in domestic animals can increase the mean value of a trait many standard deviations from the initial parent population (in part by generating new genetic combinations as low frequency alleles increase in proportion), but they are often met with negative side effects which one must buffer them from via special considerations. in the extreme fitness approaches 0 as the animals are simply infertile and incapable of reproducing (this sort of thing happens when you attempt to increase the size of chickens i believe).

p.s., and evolutionarily stable strategies can be mixed and preserve various morphs within the population via frequency dependence.

102

Ragout 04.17.07 at 3:12 am

So what exactly is this “heritability” everyone keeps talking about? How much connection does it have to our usual notion of children inheriting traits from their parents?

I take it we estimate heritability by the correlation in IQ between identical twins raised apart. But there is no logical connection between this fact and the idea that high-IQ parents have high-IQ children. For example, identical twins always have the same eye color, so I guess the heritability of eye color is 100%. But children don’t always have the same eye color as their parents.

Am I missing something, or is this term incredibly misleading?

103

ragadaisus 04.17.07 at 3:26 am

Re #95: James Lee at Harvard has IMHO come up with the answer to this. Since the end of the ice age human populations have multiplied a lot, and the probability of a new advantageous mutation showing up is directly proportional to population size (Fisher). There are lots and lots of mutations undergoing selective sweeps in the large continental populations, many of them affecting neuronal and brain function. These new transient polymorphisms should and probably do generate all the heritability that we see. For a reference search for “Moyzis Wang PNAS”: they identify 1800 sweeping alleles and a forthcoming version pushes that up to 2300 or so.

104

dennis 04.17.07 at 4:09 am

DK, I didn’t say no useful research had been conducted, I said nothing fundamentally new had been found. The discussion over at GNXP confirmed this. After a lot of spluttering, the claim was that we now have more and better data of the same general kind that we had in 1970, not that I had missed something big.

Are your feet sore yet? What exactly are you asking for from the relevant fields? We’ve spent the last two decades dismantling the wall of denial erected by Gould et al. (and are beginning to see the restoration of the reputations of those he unfairly libeled; being right still counts in the end).
It almost sounds like a creationist crowing about the inability to recreate speciation in the lab though, when discarding reams of relevant data all going in the same direction by pointing out the nonexistence of an unidentified breakthrough. And above all, it is oh so early still.

The revolution in genomics is harrowing, if one fears what it might reveal.
The past two decades have seen the confirmation of what Burt et al were marginalized for stating those many years ago, by way of all that “more and better data”, with the Great White Hope of the much used and abused (and still not satisfactorily explained) Flynn effect embodying more and more hope for those who desperately want it all to stop. As for what the Flynn effect means (not what we hope it means) one might want to read his recent Harvard paper to find out his thoughts on the matter.

Someone way up in this long comments thread has already pointed out that 70% heritability for IQ was once enough to end a career, now it is accepted by all but the last few diehards.

Once you accept heritability of IQ coupled with continuing evolution (discarding now discredited notions of stasis) you have to give up the ghost. Recent studies (Cochran and Harpending; Lahn) suggesting not only that evolution has not somehow stopped working on humans but that evolutionary pressures may have increased in recent history, in particular affecting the brain and behavior, the dread specter of significant racial differences in IQ must follow. It’s almost as if the “race as a sociological construct” canard was devised in anticipation of all this, like calling in an airstrike on your own position.

105

zdenek v 04.17.07 at 4:49 am

pithlord writes :–

“Pinker is a mysterian about consciousness/ qualia, much to the disgust of Dennett et al.”

You have to be thinking of someone else ( Colin McGinn or R. Penrose ?) because Pinker has an computational view of the mind ( mind is a neural computer that performs operations on strings of symbols ) but of course so has Dennett.

106

Pithlord 04.17.07 at 5:46 am

zdenek,

Pinker’s a mysterian about subjective experience. He says so in The Blank Slate and also in this interview with Robert Wright. As I understand it, he takes a computational view of mind-as-it-could-in-principle-be-observed-by-a-third party, i.e., beliefs, goals, emotions, etc.

107

zdenek v 04.17.07 at 6:13 am

Pithlord I am sorry but you are barking up a wrong tree , here is Pinker ( from the interview you mentioned ) :

“No this is I mean we’re we’re defiantly in agreement that there is a puzzle there and we’re both vulnerable of being accused of being mysterians of saying that this is consciousness will never be solved, it’s beyond the reach of neuro-science which is not what I believe and I think it’s not what you believe in the sense of consciousness meaning the difference between conscious and unconscious processing. ”

There is a difference between saying that the problem is hard to solve ( re qualia ) which is what P is holds and saying that it *in principle* cannot be solved .The ‘mysteryans ‘ hold the second view and it is clearly a different kettle of fish ( much stronger and much harder to defend ).

108

zdenek v 04.17.07 at 6:32 am

Pitlord here is a concession to you because P also says this :

“There is something maddening about it in that it’s on the one sense complete comprehensible and in the other sense bafflingly incomprehensible and what that makes me think is that we’ve run into some limitation in our own mind a kin to the one that make it very difficult or impossible for us to understand that time came into existence with the big bang. ”

Is this mysterianism ? it sounds like it but I would expect an argument to the effect why the problem cannot be solved ( that is what McGinn does ) so to the extent that P does not offer any arguments like that he is not a mysterian but merely someone who things the problem is v. hard to solve ( as I said that is a different claim a weaker claim ).

109

David Kane 04.17.07 at 5:21 pm

1) Thanks to JQ for participating in this thread. Very fun!

2) I have a suggested plan of action to move the discussion forward:

a) JQ should write up few sentences about his claim about what “significant new information” has come to light. I think that I understand his position, but clarifications always help.

b) The folks at GNXP could then write a literature summary arguing against this. (I am fairly sure that they will disagree with JQ.)

c) JQ could then restate his opinion and link (from a new CT post) to the GNXP literature summary. A fun discussion at CT would then ensue. I would especially enjoy hearing the views of other CT authors.

The best intellectual discussions feature smart people with different views. Putting CT together with GNXP seems a good way to accomplish this.

What could possibly go wrong?

110

Neil Morrison 04.17.07 at 10:24 pm

“The absence of a price is a big problem for an evolutionary explanation of IQ differences. Evolution is supposed to optimise fitness.”

That’s “genetic” fitness, ie increased presence of particluar genes in the gene pool at the expense of others. Although there is bound to be a cost (larger brain, more neurons, more energy consumption) it’s the competition amongst genetic variants that will drive the evolution of IQ. Fitness is hardly ever stable.

111

Neil Morrison 04.17.07 at 10:31 pm

Forgot to add – there will always be variants of particular genes within a population.

112

Ozma 04.17.07 at 11:09 pm

I haven’t read all these comments – but what an awesome review of Pinker’s book!

113

John Quiggin 04.18.07 at 1:51 am

DK, I’ve said my piece for now, but I’d be interested to read a GNXP view of the main developments in recent decades, taking account of the Flynn effect. If someone at GNXP wants to write something like this, I’ll certainly make some comments and may respond at greater length.

114

John Quiggin 04.18.07 at 10:33 am

Ozma, thanks for your kind comment.

115

abb1 04.18.07 at 11:50 am

I liked the review too, though I disagree that the nature/nurture controversy is frivolous, which is what the last paragraph seems to imply.

116

Michael Mouse 04.18.07 at 2:59 pm

Why didn’t selective pressure push IQs up to 150 (or whatever upper bound you want to nominate)

Err … because IQs are (were) normalised to 100? Or was this a trick question?

Comments on this entry are closed.