quote:Originally posted by rivka: Also, the story is 4 years old. Why bring it up now?
It's an interesting story that I was not aware of four years ago. Was it discussed here? I did a search on the key terms and only got this thread.
Posts: 532 | Registered: Feb 2009
| IP: Logged |
posted
There's a problem with any sort of research into correlations between ethnicity and intelligence: Namely, it is likely to make the researcher very unpopular, unless the result is "no correlation", in which case why bother? Even a one-sigma increase in IQ for some particular group would lead to a hydra-headed uproar as people criticised the data, the methodology, the intentions of the researchers, their private lives and thoughts, the journal that published them, the university that paid them... Just take a look at what happened with the Bell Curve. (Which, incidentally, is a fine example of why this sort of research is so suspect.) So the only people who take on this kind of thing are the ones with an axe to grind - nobody else wants the hassle - which only feeds the suspicion that their research is greeted with. By this time, we're so deep into this cycle that the critics are usually right; ethnic-IQ research really is pretty unreliable! So even if any particular study is actually correct, there's no way to tell it from the shizzle of people who found what they set out to find.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
posted
So I gathered from my sociology studies in college (only a few semesters of required breadth courses).
Even if there were a definable genetic trait that tended towards "higher intelligence," the outcome would be impossible to observe objectively. It would be impossible even to establish a baseline level of average intelligence, much less develop an effective instrument for measuring individual capacity for it. Every single possible test carries enough social and cultural variables to throw the results off by margins wide enough to discredit any result. And with good reason- human intelligence is effectively unmeasurable, and the belief that it can be defined is just a romantic notional holdover from colonial days.
quote: By this time, we're so deep into this cycle that the critics are usually right; ethnic-IQ research really is pretty unreliable!
I don't think the current politics of research changes anything about the original outlook. It was always a foolish endeavor, based on wishful thinking and ethno-centric pseudoscience. Do you think that at any time in the past, we were actually more equipped to deal reasonably with the subject? Because as I recall much of the scientific thinking on race in years passed was deeply connected with political and cultural prejudice against non-European cultures- that's where we get the basis for eugenics and phrenology, the misapplied justifications for cultural prejudice couched in scientific observation.
Posts: 9912 | Registered: Nov 2005
| IP: Logged |
posted
Ok, it's a reasonable point. But in the 1930s, let's say, when such research was respectable, it would have been easier for someone who didn't have an axe to grind to look into this. Just for starters, he could have talked about his research at cocktail parties without becoming a pariah! And whatever he found wouldn't be immediately suspect as the product of axe-grinding, or at least not as much as it is today. Thinking about it, perhaps such research from the thirties exists, shows no measurable differences between whatever groups they looked at, and you never hear about it? Such a scenario would at any rate not surprise me. But notice: If we look up what was done back then, and we find that it shows Caucasians more intelligent than ethnic group X, we won't believe it; we'll chalk it up to bad experimental design. What's more, we'll likely be right to do so.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
quote:Originally posted by rivka: I think that a story that starts with so-simplified-they-become-nonsensical claims like this:
quote:But ultimately, they became known as Ashkenazim, a variation on the Hebrew word for one of Noah’s grandsons.
has, at best, not done its research.
Also, the story is 4 years old. Why bring it up now?
That story is. I remember reading it on Jerry Pournelle's site. link,link
Those are from 2002 and 2003.
I think it's an interesting idea, but there's no way on earth to really know if there's anything to it.
Posts: 12266 | Registered: Jul 2005
| IP: Logged |
posted
Why not? We understand how to measure IQ, and we understand how to extract genetic from cultural influences - twin studies, for starters. In an ideal world where neutral researchers could work on such questions without suspicions on their motives, we should certainly be able to see if there's any genes correlated with intelligence.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
quote:Originally posted by King of Men: Ok, it's a reasonable point. But in the 1930s, let's say, when such research was respectable, it would have been easier for someone who didn't have an axe to grind to look into this. Just for starters, he could have talked about his research at cocktail parties without becoming a pariah! And whatever he found wouldn't be immediately suspect as the product of axe-grinding, or at least not as much as it is today. Thinking about it, perhaps such research from the thirties exists, shows no measurable differences between whatever groups they looked at, and you never hear about it? Such a scenario would at any rate not surprise me. But notice: If we look up what was done back then, and we find that it shows Caucasians more intelligent than ethnic group X, we won't believe it; we'll chalk it up to bad experimental design. What's more, we'll likely be right to do so.
I think the results don't really matter that much- the area is taboo now largely, as you've said, because it is such a wasteland of actually useful or reliable data. But we have to remember that we arrive at our currently accepted scientific and sociological understanding of race as a somewhat a priori conclusion. Empirical observation of ethnic traits and general racial features is too subjective to allow us to define with any kind of reasonable reliability, the specific abilities of any racial group. And often in the past we've assigned certain traits to races, rather than more properly to social or linguistic groups that were actual responsible for the traits, ie: Asians have generally higher math scores because of their linguistic backgrounds, African Americans are dominant in professional sports because of socio-economic conditions in many black neighborhoods, etc. You can't start with the fully grown athlete, and deduce something important from his genetic structure that necessarily caused him to be an athlete, just like you can't take a spelling bee winner and isolate the gene that caused her to spell perfectly, so any study claiming to link genetics to general aptitudes must necessarily ignore 10 more prescient factors from the real world in order to focus on one hazy and impractical generality. If 9 out of the top 10 world chess champions are Russians, can you conclude that Russian genes are linked to chess ability? Of course not. The conclusion of general intelligence among Ashkenazi Jews is even more dubious, given that we're talking about an international ethnic group with an insular religion, and a strong emphasis on higher education that dates back millenia- given that members of this group have long been involved with the creation of and preservation of western theories of intelligence and enlightenment in the first place, that some instruments devised to test intelligence favor this group generally would not be surprising.
The blunt analogy is writing a universal intelligence test in Catalan, administering it to the whole world, and being fascinated with the conclusion that Catalans are the smartest people in the whole world, followed by the Spanish, Provencali, the French, Italians, etc. There's simply no test possible that obviates cultural bias when testing between different cultures. Hell, even just in America we know that our own standardized academic tests are advantageous to certain social groups, even if only because those are the groups who supply the largest number of test writers.
Posts: 9912 | Registered: Nov 2005
| IP: Logged |
quote:Originally posted by King of Men: Why not? We understand how to measure IQ, and we understand how to extract genetic from cultural influences - twin studies, for starters. In an ideal world where neutral researchers could work on such questions without suspicions on their motives, we should certainly be able to see if there's any genes correlated with intelligence.
Motives notwithstanding, all of these scientists, absolutely without exception, would have certain biases. There is no one person who speaks and understands every language in the world, or has even a working knowledge of every culture. There is no test possible that would avoid cultural bias, and as far as I know there is no effective way of testing intelligence without involving culture. Even IQ tests require basic cultural knowledge that is not universal by a long shot- IQ tests only really measure a person's facility with language and culture- the other elements are superficial items of general knowledge or methodology.
Posts: 9912 | Registered: Nov 2005
| IP: Logged |
posted
Ok, firstly, you appear to be underestimating the effect of IQ tests. They really do measure something that's not purely cultural, although I'm not claiming that they are zero percent culture. Secondly, your argument that
quote:You can't start with the fully grown athlete, and deduce something important from his genetic structure that necessarily caused him to be an athlete, just like you can't take a spelling bee winner and isolate the gene that caused her to spell perfectly, so any study claiming to link genetics to general aptitudes must necessarily ignore 10 more prescient factors from the real world in order to focus on one hazy and impractical generality.
is a bit of a straw man. Of course you can't deduce anything from single people. The study of genetics is always the study of populations. That doesn't mean you can't learn something about the statistical effects of alleles.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
posted
I would be more likely to put the blame on cultural things than genetic ones, because I think it's likely that the difference in intelligence/capability is very minor when compared to cultural/experience factors.
Posts: 8473 | Registered: Apr 2003
| IP: Logged |
posted
On what basis? Nobody has ever studied this in a trustworthy way! All you're saying here is that you have a prejudice against genetic-ethnic explanations. Again, we know how to differentiate between culture and genetics, that's what twin studies and regression analyses are for. For all my disdain for social science, you really can learn things about genetics if it's done right. This is not rocket science!
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
posted
Um, no. You do not need to know anything about the gene-brain or brain-mind interfaces to learn about correlations of IQ and genes, any more than knowledge of the nuclear forces was needed for macro-level descriptions of radioactivity. You can learn quite a lot of useful things at high levels of abstractions without having any micro-level explanations.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
posted
How can you distinguish between cultural and genetic differences using studies of identical twins, unless identical twins are genetically identical? As Scientific American reports, it seems they are not. That would at least seem to make such studies more complicated.
posted
By measuring the difference in variation between twins and non-twins. Look, I know science is complicated. There is always some complicating factor that doesn't make it into the press release. But the other point is that scientists are, as a general rule, really smart. They really do find ways to deal with these complicated issues, and they really do come to conclusions that are genuinely true. It takes two seconds to post a link about some difficulty; it may well take two years to figure out how to get around it; but nonetheless it does get figured out. This is why a PhD thesis takes five years, but problems actually do get solved.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
quote:Originally posted by King of Men: Um, no. You do not need to know anything about the gene-brain or brain-mind interfaces to learn about correlations of IQ and genes, any more than knowledge of the nuclear forces was needed for macro-level descriptions of radioactivity. You can learn quite a lot of useful things at high levels of abstractions without having any micro-level explanations.
But those correlations don't *mean* anything. They don't at least mean anything terribly useful.
Thus my point- our assumptions about the nature of intelligence are a priori- the connection between genes and measured intelligence can't be shown to be causal without that link- we would actually have to show that the brain functions differently, more effectively, more precisely, whatever, instead of just being able to show that a person's cognitive process is most similar to what we arbitrarily consider the optimum. What exactly makes you think that there wouldn't be a plethora of aberrations in the data, and that all those would be conveniently linked to a significant alteration in background for the individual. What makes you think that intelligence is useful at all as a statistic? I can show you statistics that connect wealth with English ancestry, so I suppose I'm genetically predestined to be wealthy. I'm aware that's a strawman, but this that you're talking about isn't much better, really.
I just think this is all a fool's errand.
quote: This is why a PhD thesis takes five years, but problems actually do get solved.
We both know studies pertaining to "cognitive potential" and genetic heritage have been kicking around for decades. As I said, some of the earlier versions of these theories formed the basis for eugenics and phrenology, which were popular pseudo-sciences. The problem is that the problem of studying this particular question is insurmountably large. It's not like finding genetic links for diseases or physical features- it's a couple of orders of magnitude removed from that level of observation. Maybe I'm wrong, maybe next year someone will build a machine that accurately measures everyone's ability to do anything with their brains, and it's 100% predictive. I seriously doubt it.
Posts: 9912 | Registered: Nov 2005
| IP: Logged |
quote:What makes you think that intelligence is useful at all as a statistic?
This seems to me the core of your objection. Are you seriously suggesting that intelligence is not a factor in life outcomes? Because if it is, that looks like a damn useful statistic to me. If you can increase the intelligence of the human race by even a tenth of a sigma through tweaking one gene, or supplying one particular protein during childhood, or whatever, are you really going to say that's not worth doing?
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
posted
Well, of course. My point is that "intelligence" is not a genetic feature. It's not even properly quantifiable at all. It's related to genetics in many ways, but so many ways that it would be a gargantuan task to even begin to map them. And where would you start to determine a "positive" sigma change? Aren't different genetic makeups also responsible for the ways in which our intelligence develops in our given environments? That's too many factors to control for, at least for me to sign off on the idea that it can be done at all.
I mean, maybe I'm being dense and stubborn here, but I know a guy who learned to speak Czech, with perfect fluency, in four months. The guy can barely hold an intelligent conversation in his own language. Do I want his genes?
Posts: 9912 | Registered: Nov 2005
| IP: Logged |
quote:My point is that "intelligence" is not a genetic feature.
I think we have here a serious mismatch in the way we are using language, because if this sentence is true, then either "intelligence" or "genetic feature" in the sense that I use these words is meaningless. Of course intelligence depends on your genes. So does body size, muscle development, and hormone distribution. That all of these can also be influenced by your environment does not change that they depend on your genes.
quote:It's not even properly quantifiable at all.
I believe that on this point you are just plain mistaken as a matter of empirical fact. IQ tests at age sixteen do in fact measure incomes, you know. I do not claim that this is everything that is meant by intelligence, but it is clear that something is being measured which is at least a component of intelligence.
quote:And where would you start to determine a "positive" sigma change?
IQ is normally distributed with average 100 and standard deviation 15. (Some tests use a different sigma.) What is the difficulty? And yes, I'm aware of the issues with IQ as a measure of intelligence; but do not throw the baby out with the bathwater. Once again, IQ does measure something useful.
quote:I mean, maybe I'm being dense and stubborn here, but I know a guy who learned to speak Czech, with perfect fluency, in four months. The guy can barely hold an intelligent conversation in his own language. Do I want his genes?
If you are able to identify the genes for language-learning, and avoid the ones for bad communication skills, then yes, you surely do. I'm aware, of course, that the phrase "genes for X" is a vast over-simplification; but really, we're not going to learn anything by throwing up our hands and saying "It's too difficult"! You have to start somewhere.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
I believe that on this point you are just plain mistaken as a matter of empirical fact. IQ tests at age sixteen do in fact measure incomes, you know. I do not claim that this is everything that is meant by intelligence, but it is clear that something is being measured which is at least a component of intelligence.
IQ is also a reliable determiner of social status, especially at an early age. I understand why you're using IQ as a reference, but I'm not at all comfortable with the idea that higher IQ is necessarily desirable across the board, and even less sanguine about the notion that any specific racial group actually could have an identifiably higher IQ as a rule. I'm not just throwing my hands up- I'm being reasonably cautious about a subject that neither of us is expert in. I remind you that it is these questions in general which have led to many socially destructive acts, mostly caused by irresponsible use of questionable data, and even more questionable theories of its meaning.
Posts: 9912 | Registered: Nov 2005
| IP: Logged |
quote:But I'm not at all comfortable with the idea that higher IQ is necessarily desirable across the board
Yes, of course any society needs some Gammas and Deltas to be janitors and whatnot. Are you sure you want to make this argument?
quote:[E]ven less sanguine about the notion that any specific racial group actually could have an identifiably higher IQ as a rule.
Why? Here are several things to consider before you answer that. First, truth is truth. Second, you don't object to the idea that some racial groups have higher muscle/fat ratios, or more lung capacity, or better lactose tolerance. Third, if we knew how a racial group was getting its higher IQ, we could fix it in everyone else! Fourth, it is clear that any effect of race is going to be small; going back to IQ, it cannot be more than say five points, or we would certainly know about it by now. That's way smaller than the variation across individuals, which is on the order of a hundred points - at IQ 80, you're well within the normal range, while IQ 180 is one in several million. If there is any injustice in IQ distributions, the known fact that some people have 180 while some have 80 is a lot worse than the conjectured possibility that some ethnic groups average 103 and some average 98! And really, this known fact actually is a truly dreadful injustice. My observation and feeling is that smarter people are more alive, in the sense of taking more joy in life, than average people. They are more able to ignore small things and enjoy big things. They have (almost by construction) much better abilities at entertaining themselves out of their own resources, just by plain thinking about things. And, apart from these subjective interpretations, they are able to make enough money for a comfortable lifestyle without much drudge-work. That people are denied these advantages is terrible!
quote:I remind you that it is these questions in general which have led to many socially destructive acts.
No. The socially destructive acts are caused by people believing they have answers for these questions, and not bothering to do real checks; and then reasoning from is to ought, or more accurately, from is to their existing prejudices. The Nazis had at least the courage of their convictions; it's rare to see modern race theorists argue that the low-IQ people of their own race should be sterilised. But no question has ever caused a socially destructive act.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
quote:Originally posted by King of Men: Why not? We understand how to measure IQ, and we understand how to extract genetic from cultural influences - twin studies, for starters. In an ideal world where neutral researchers could work on such questions without suspicions on their motives, we should certainly be able to see if there's any genes correlated with intelligence.
IQ is nonsense. There are so many different types of intelligence that the whole idea becomes silly. I scored 150 in kindergarten and 135 in 5th grade. Did I get dumber in those 5 years? And those tests are timed. Those of us who are good test takers have a huge advantage over very intelligent people who aren't good test takers.
To say nothing of all the cultural bias that's built into so many of these tests.
I had a roommate once who needed to give an IQ test to someone for her grad work, so I let her test me. And there were vague things on the test that simply can't be any kind of objective gauge of intelligence.
Posts: 12266 | Registered: Jul 2005
| IP: Logged |
Why not? Children develop at different rates, and the rates change. IQ for children is defined a bit differently than for adults, being relative to their age group rather than the entire population; why shouldn't you have been a fast developer in your earlier years, then slowed down a bit and allowed others to catch up? And apart from that, how certain are you that it was the same test? There are many different IQ tests, with different standard deviations. For all I know, the 135 and the 150 are actually equal.
Finally, consider once again that IQ at 16 does, in fact, predict incomes at 40, even after accounting for social status - to answer Orincoro's objection from above. If the idea is nonsense, how can it have predictive power?
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
posted
I don't think it's nonsense, I just haven't been convinced that it actually measures intelligence- that it doesn't more accurately measure other factors, like cultural awareness, grasp of language, etc.
Posts: 9912 | Registered: Nov 2005
| IP: Logged |
quote:Originally posted by King of Men: There are many different IQ tests, with different standard deviations. For all I know, the 135 and the 150 are actually equal.
If that's the case, the numbers are virtually meaningless.
quote:Originally posted by King of Men: Finally, consider once again that IQ at 16 does, in fact, predict incomes at 40, even after accounting for social status - to answer Orincoro's objection from above. If the idea is nonsense, how can it have predictive power?
In the first place, there may be a corrolative relationship, rather than a causative one. In the second place, I recall a study where one set of parents were told their children had high IQs and another set were told their children had low IQs. The children of the first set of parents did better than the childre of the second set. Expectations can make a big difference.
Posts: 12266 | Registered: Jul 2005
| IP: Logged |
posted
IQ sure ain't what it was made out to be, but it's not nonsense. Some people just have better genes when it comes to their potential for actualized intellectual power, and we do have ways of testing to determine which people are highly intelligent, and which people are not.
Posts: 15421 | Registered: Aug 2005
| IP: Logged |
quote:In the first place, there may be a corrolative relationship, rather than a causative one.
Yes, yes. There are three options: Either 16-IQ causes 40-wealth, or vice versa, or both are caused by a third factor - your 'corrolative' relationship. Unless you believe in time travel, we can rule out the second option. And if there's a third factor - please notice, socio-economic class has been ruled out - then I'd say it's one that's worth knowing about! Since IQ is apparently a proxy for it, it's a quite useful number.
quote:If that's the case, the numbers are virtually meaningless.
No, it just means you have to think for a minute. If one test is sigma 17.5, and the other is sigma 25, then the numbers are telling us exactly the same thing: This subject is two sigma above the average. Now these are fairly unlikely numbers for IQ test sigmas, so I think the explanation that you slowed down in your teenage development more likely; but having to know the sigma that the nubmers are relative to does not make the measurement meaningless.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
posted
If it were true that for genetic reasons Jews had higher average intelligence than surrounding Middle Eastern tribes, would that be evidence that the Jews are God's chosen people? >_> (I saw an article about this phenomenon in Commentary a few years ago that concluded this way. Hopefully the author was being facetious.)
Posts: 781 | Registered: Apr 2005
| IP: Logged |
quote:Originally posted by Mucus: Israel does indeed look kinda surrounded
Ok, now that, I'll give you, is a great demonstration of the weaknesses of IQ tests. That's plain cultural effects.
Posts: 10645 | Registered: Jul 2004
| IP: Logged |
posted
Though the article makes the whole idea seem really biased and unscientific, and though I believe it's almost impossible to look at such questions with any rigor at all, so that the question as it's posed seems almost meaningless, still, my son is by heritage an Ashkenazi Jew, and he happens to be very intelligent. So maybe there's something to it, though I suspect the best thing is to ignore that and go on treating people as individuals with traits and quirks that belong simply to themselves and not their genetic or ethnic or geographic group.
Posts: 6246 | Registered: Aug 2004
| IP: Logged |
quote:And if there's a third factor - please notice, socio-economic class has been ruled out - then I'd say it's one that's worth knowing about!
IQ tests basically seem to me to test the ability to be good at tests. Break down a problem, solve it, not spend it too much time on things you can't solve but rather move on.
And that ability does influence future income, yeah.
But I'm not sure we'd necessarily want to increase the human race's ability to take tests, if it comes with the price of having the human race be antisocial or less creative.
IQ tests don't test how capable you're of writing a moving story. Or how intelligent you behave in a relationship. Or how capable you are to persuade others.
These all seem to me to be aspects of intelligence that matter lots and lots, but aren't measured even a tiny bit in IQ tests.
Posts: 676 | Registered: Feb 2003
| IP: Logged |
Printer-friendly view of this topic
![[Razz]](tongue.gif)
![[Wink]](wink.gif)
