Very interesting – thanks, Mijin. From the Wikipedia descrition of the Flynn effect (a “phenomenon” mentioned in the abstract of the article you cited): “Attempted explanations [for shifts over time in the average scores for people living within a nation-state’s borders on so-called “IQ” tests] have included improved nutrition, a trend toward smaller families, better education, greater environmental complexity, and heterosis[9]. Another proposition is greater familiarity with multiple-choice questions and experience with brain-teaser IQ problems.”
If we didn’t see genotype differences among populations, I’d get a little more interested in this sort of study. But since genes do vary in prevalence among populations, I’m less inclined to jump on the nutrition band-wagon. It’s nice to see the authors at least tacitly acknowledge an actual IQ difference among populations rather than attacking the idea of measuring IQ or pretending that there is no difference. That’s a start.
It bugs me (from an egalitarian viewpoint) that in the US (as a well-studied example), children of wealthy blacks and children from black families with highly-educated parents substantially underscore their white and asian peers, and barely score on par with white children from poverty-stricken families and white children from families where parents did not finish high school.
It makes sense that poor countries with rotten infrastructure have higher parasitic loads. It makes sense that that rotten infrastructure corresponds with lower average IQs. Beyond that I remain unconvinced; if such a hypothesis were correct, surely we wouldn’t see that gap persist between, say, blacks and asians emigrating to the UK. The gap does persist in those two groups and it’s very, very large. So what you end up needing is an entire hodge podge of various environmental and nurturing arguments according to where your populations are living. Moreover, it’s not as if parasitic infestation is an explanation for IQ differences within the same subpopulation living in the same region. That’s obviously genetic…
Genetic differences is a much cleaner and simpler explanation. Occam’s razor, and all that.
Actually, they can.
Suppose I have two populations of cows, and I want to see which can produce the most milk. Here’s what I do: I give subsets of each population similar environments and I control the variables. What’s left is genetic. I don’t have to decide which gene is coding for what, and how it works. I’ve normalized nurture, and what’s left is nature.
Much of the argument with these genetic debates revolves around the supposed impossibility of controlling for nurture, and I maintain that most of the dismissiveness is a result from an a priori assumption that we are all basically equal in potential except for some minor phenotypic appearance differences.
I find the enormous disproportionate representation of blacks in sprinting and the NBA (to use two examples) or Kenyans in particular in marathons (to use a third) very suggestive for nature. In all three of those instances, the enrollment pathway is clearly open to all groups and the desire to put those pursuits ahead of all others is also at least equal for all putative candidates. That is to say, no white kid gives up basketball until he is no longer competetive.
With academic pursuits, the gap has remained stubbornly persistent in higher education even when only the very top tier of blacks is selected for and pitted against all whites and asians. And even when very specific, very targeted efforts are made to boost opportunity for that sub-group of black students beyond just equalizing opportunity.
I recognize genetic egalitarians will never accept that opportunity is normalized–it would undermine their entire argument. That’s why the first chink in the armor is a demonstration that genes do vary in prevalence by racial groups. The second chink will be the demonstration that many of those genes are not just appearance genes (and why should nature limit them to that?). The third chink will be a demonstration that, even within the same broad population, gene prevalences vary according to a skillset–let’s say, for instance, that all Fast Sprinters have FSGX and Slow-Sprinters have FSGY. And the nail in the egalitarian coffin will be showing prevalence of those decoded genes between two different populations.
Bit by bit, on each of those advances, the egalitarians will retreat.
That will also be the nail in the coffin among the racialists.
In fact, we already have that situation, today. Sickle Cell Anemia, Thalassemia, Lactose Intolerance, Tays-Sachs, and a host of other genetic situations all occur in multiple populations. Grouping populations based on the prevalence of any of those conditions makes no sense, given the wide variety of other conditions that do not hold true for the smaller populations who share those traits. (Interestingly, a repeating refrain from among the racialists exactly buys into that error, with claims that Sickle Cell is a “black” disease or Tays-Sachs is a “Jewish” disease, and similar erroneous claims.)
Identifying specific populations that suffer certain conditions makes sense in order to target remedies. Lumping widely differing populations together because some subsets share conditions makes no sense at all. It tends to result in people trying to discover all the great marathon runners from Ghana and the marvelous sprinters in Kenya.
OTOH, it shouldn’t be ignored that the researchers compared other correlations such as with wealth and with race, and the correlation was weaker.
I’m not trying to sell the idea that childhood diseases are responsible for the lion’s share of IQ differences between nations. I haven’t made my own mind up yet.
I find it interesting that the response of many people on the web (e.g. on the Economist’s discussion forum. Not here, yet) is to voraciously attack this research, even as they continue to cite IQ and the wealth of nations.
…when the studies are essentially the same kind of analysis, but wealth of nations is older and uses dodgy data.
A book from 2002 is old?
Are you certain you’re not confusing yourself between that and The Wealth of Nations from 1776 (which, I don’t think, comments on race at all).
Older than the studies under discussion.
2002 is actually more recent than I thought…
But still, that’s older, so I don’t need to change what I wrote.
And as I’ve alluded, with IQWN, there’s no shortage of people arguing that it’s conclusion is true, and it’s just the political correctness brigade that refuse to accept it.
But suddenly with this study, it’s a case of “oh, you can support any silly conclusion with statistics” etc
Relevant recent study, referenced in the latest *Nature *- basically, it’s foolish to make statistical generalizations about other populations based on work overwhelmingly done on a small, eccentric subset.
Like IQatWoN, even before the data manipulations.
We aren’t grouping populations based on genetic diseases–we’re grouping genetic diseases based on prevalence within populations. And various populations DO have various prevalences for all sorts of physiologic conditions, whether it’s HbS disease or precocious puberty, or renal sodium handling or any one of any number of genetically-based differences that have varying prevalences according to the population category.
You keep getting confused about what is being argued here, and I’m starting to think you are deliberately obfuscating the point. It’s not that hard.
Suppose I wanted to argue that HbS disease (there are a variety of hemoglobins that cause sickling) is more prevalent in blacks because of genetic, and not nurturing reasons. I examine prevalence of HbS disease and find it’s more common in blacks. I examine gene prevalence coding for HbS and find the gene prevalence is more common in blacks (who–as a group–are more likely to be associated with areas where HbS has selection advantages).
I’m sorta done: HbS disease is genetically-based, and more prevalent in blacks, and the reason it’s more prevalent is nature and not nurture.
It’s a completely irrelevant red herring to complain that someone thinks it’s a “black” disease. They are either ignorant or using verbal shorthand. It’s completely irrelevant that someone thinks we should define blacks as "the group with a higher prevalence of HbS disease. Kind of stupid, that.
What it is, is a genetic condition more prevalent in blacks, as a group. This would be true if you took all blacks in the world and all Eurasians in the world, because the very high prevalence among some black sub-populations drives the average for the whole group; it might not be true for some very select sub-populations because those sub-populations might not run true to the average (say, for instance, ones in the Mediterranean area).
All of that is just obfuscating the key point: differences between two populations are genetic, where we control for nurture. The Kalenjin subpopulation might be lousy NBA players, but an extraordinarily high prevalence of genesets coding for running might be enough to cause some entire metagroup into which you plug them to be over-represented on the world stage.
I am much less interested in arguing HOW to group people than I am in arguing that differences are genetic. But IF you group them–if you decide there is a population known as “black” or “African American” or “Jew” or “Chicagoans” and THEN you normalize for nurturing differences when you compare that group to some other group–“Philadelphians,” maybe–the difference is genetic.
Genetic.
I don’t think anyone is arguing that it’s impossible for there to be inheritable genetic differences that will subsequently be more likely to appear within a certain related subpopulation. The argument is that at the moment, for the sorts of things you’re talking about, correcting for nurture is nigh-impossible. It may well be that some subpopulation of blacks have a particular bone structure or musculature that makes them better for sports than any subpopulation of whites in the US, or that the white subpopulation is smaller. But you’ve certainly not shown any evidence of that, and you’ve certainly not shown that you’ve corrected for nurture. And that’s dealing with the case of the US where you and I understand the social situation and where our two groups are relatively close in terms of diet, training, etc. If you can’t even make the argument for American basketball players, saying anything about the IQ or physique of people living in Africa is simply absurd.
Find me the genetic marker for a basketball player. Show me bone measurements that compare black and white subgroups. If you can’t do that, you can reiterate your rightness all you want, but insistence simply isn’t proof.
Nope. You keep shifting around in various threads. On the other hand, I will note that I was not directly addressing your claims–unless you now are claiming to be among the racialists, the group I explicitly addressed.
You continue to hammer away at “the egalitarians” regardless whether any given poster has actually expressed that perspective and have accused other posters of being part of a group for which you have come up with a cute name.
If you do not like the reverse argument applied to you, then you should back off and simply address the actual remarks of the posters with whom you are engaged without throwing around your labels or attacking postions that have not been expressed.
It seems to me that it makes a lot more sense to simply identify the groups that actually display that prevalence rather than gathering multiple groups encompassing an enormous number of people based on irrelevant characteristics and them declaring them a “group” about whom you are making a “genetic” observation.
There is no question that HbS is genetic. So what? When you decide to identify it as a “black” disease while ignoring the number of white populations who have it and the number of black populations who do not, you are creating an artificial claim that provides no worthwhile information while placing potentially harmful barriers around the information in ways that will misdirect people to ignore white sufferers or falsely include black non-sufferers in future decisions.
It is simply not that difficult to identify a person by whether their ancestry is Greek or Nigerian rather than to look at their skin color, thereby getting a more accurate basis for future investigation.
And yet, using slightly different words, that is exactly what you just did..
Which is a pretty silly way to use statistics.
Then perhaps you should simply stick to your favorite argument that genetics determines all things and stop saying anything about any groups? Every time you begin to discuss issues on the group level, you fall directly into the errors you have described, above.
Sickle-cell (or HbS, if you prefer) is an interesting example, because it in fact shows how it’s environment, *not *population genetics, that determines this sort of thing - or why would there be a strain stretching across Arabia and India, when the supposed-associated phenotype doesn’t?
Chief Pedant, after all the caveats and the redefining of populations for specific purposes and the exceptions and the limitations on interpretations and the within-group and between-group heredity this and the that, it’s not so much that what’s left of your argument is wrong, but rather that it is, objectively speaking, uninteresting.
(A bit like how Stephen Jay Gould, to prove a rather different point, showed how his own age and the price of gasoline were highly correlated in the years up to his writing The Mismeasure of Man. True, but who cares?)
Along the lines of what tomndebb suggested, your argument can only be considered “interesting” if you are hoping your ideas are misappropriated by your audience to mean more than they do, in ways which only led to bad things in the past (and present).
(Notice this is NOT the same as saying that “liberals” are willfully ignoring your “evidence” for fear of its “consequences”. They aren’t.)
The environment does help drive the genetic differences among populations. That’s what evolution is all about: Descent with modification. And modifications which are advantageous to reproduction are the ones which become increasingly prevalent.
But once those prevalences are established for a given set of populations being studied–i.e. when you take a snapshot of the populations to one another–it’s the genetic differences that create the phenotypic expressions once you’ve controlled for nurture.
In summary, in your view, is it “uninteresting” or is it “incorrect” that two comparison populations are different in average IQ because they have different prevalences of genes?
If you want to call it simply “uninteresting” (because, for instance, you don’t like lumping people into racial populations), then I have no quarrel.
If you want to call it “unsupported” (because, for instance, you don’t think we can ever normalize for nurture) then I have a quarrel, but I doubt it will be resolved by pulling out the same arguments. No success so far, anyway 
.
We’ll both wait for more genotyping and elucidation of specific gene functions, I guess.
Surely you jest…there are thousands upon thousands of peer-reviewed articles over decades of study showing physiologic differences between black and white cohorts.
See herefor one by Kaiser Permanente for bone density.
See here for an example study looking at pelvic floor musculature in nulliparous asymptomatic women.
See here for differences in central blood pressure and vascular function in healthy young men.
See here for a comment about race differences in reference range for the muscle enzyme, creatine phosphokinase. “Black men have higher CK levels than non-black men or black women, who have higher levels than white or Asian women. Differences in muscle mass are believed to account for these sex and racial differences.”
You aren’t suggesting that–as a group average–there aren’t consistent, measurable, physiologic differences among race groups, are you?
When we talk about the NBA, aren’t we talking about outliers? I don’t think any of your IQ testing curves say very much about the outliers
You keep saying this about normalization. This has never happened so why do you keep saying it?
Quite the opposite: vast differences among racial groups in outliers is a good indication of different group averages; see my post earlier.
If you really wanted to protect the no-difference hypothesis, you could also claim inherently greater variance among the more-outliers group, but that would 1) defeat the purpose of claiming no genetic differences, and 2) require more outliers at the weakling end of the spectrum. There’s a good case for males having higher IQ variance than females; hence more geniuses and more idiot criminals, but for other groups and traits there is less evidence.