Sunday, November 19, 2006

Bad Petition

A reader asked me to draw attention to this petition against LSE Professor Satoshi Kanazawa for his research claiming that low IQ is the cause of Africa's woes. Judging by the news story, his claims sound ludicrous. (If based on correlational data alone, it seems far more plausible that any causal arrow would go in the other direction!) But I can't support the anti-academic, politically motivated petition. It reads as follows:
We, the undersigned, feel that the academic paper 'Mind the gap…in intelligence: Re-examining the relationship between inequality and health' written by LSE lecturer Satoshi Kanazawa could be read as promoting racist stereotypes. Dr Kanazawa claims to show that the lower life expectancies and higher levels of infant mortality in less developed nations are the result of lower levels of intelligence, rather than economic inequality.

We also regret the negative publicity this has attached to the LSE, which prides itself on being a diverse and tolerant institution, through the media coverage surrounding the publication of this paper.

We ask that the LSE reaffirms its opposition to racist or racially reductionis[t] standpoints and condemns the publication of any research that may jeopardise students' safety and well being by providing ammunition for those seeking to promote a racist agenda.

Almost every single sentence of this petition betrays its anti-intellectual PC ideology.

1) The mere fact that something "could be read" in a racist way does not justify opposition to it. Surely we should instead oppose those things that really are racist. Despite what some leftists apparently believe, a mere accusation is not enough to establish guilt.

2) The petition then objects to what Kanazawa's research "claims to show". But it never addresses the crucial question of whether the research succeeds in showing this. Yet that is the only question relevant to assessing academic research. Academics are meant to pursue truth, not comforting conclusions. We should object not to Kanazawa's heretical claims, but to their lack of reasoned support (assuming they are in fact unsupported).

3) Okay, the third sentence is harmless enough. Though it seems a bit hypocritical to lament bad publicity in a public petition against the school!

4) The conclusion is a straightforward denial of academic freedom. The LSE's response has been entirely appropriate:
The LSE declined to offer any opinion on Kanazawa's conclusions but defended his right to publish controversial research. A spokeswoman said: 'This is academic research by Dr Kanazawa based on empirical data and published in a peer-reviewed journal. People may agree or disagree with his findings and are at liberty to voice their opinions to him. The school does not take any institutional view on the work of individual academics.'

The petitioners, in contrast, are asking the school to condemn unpopular research, based simply on the claims it makes, and without regard for its academic merit. That's so incredibly wrongheaded, and fundamentally anti-academic, that I'm struggling for words. (Maybe you could go read Butterflies and Wheels instead.)

By all means, refute Kanazawa's arguments. Even condemn him for irresponsibly drawing politically dangerous conclusions on such a flimsy basis. But for God's sake don't just complain about heretical conclusions without regard for their truth! Such ideological witchhunts give the Left a bad name. Aren't we meant to be the "reality-based community"? Let's please start acting like it...

Categories:

20 comments:

  1. Right on! Even if this research is empirically sound and the tests are fairly administered (something I find hard to believe), the data supports a different conclusion. If an entire continent scores much lower than another on a certain IQ test, it's pretty strong evidence that the test is not a fair measure of what we call intelligence. Of course, discovering empirically what is and is not a fair test for intelligence is a tall order, but if this doesn't count as evidence that we don't yet have it, I don't know what will.

    ReplyDelete
  2. I agree with your post.

    There is obviously a grain in truth in his argument (although I don’t know if he does it right) after all being smarter (particularly in a euphemistic way like culturally sensitive IQ) does generally tend to make you slightly more able to live longer, and this “Nobody could prove that there are racial or national differences in IQ” is just crazy talk.

    ReplyDelete
  3. I've only briefly skimmed the guardian article, but I'm not sure that the author denies this:
    "(If based on correlational data alone, it seems far more plausible that any causal arrow would go in the other direction!)"

    Isn't his basic point that their current levels of intelligence determine their life prospects, and therefore it's that we should aim to change, not the poverty, except insofar as the poverty affects the intelligence levels.

    Now that obviously sounds wildly implausible to me, but I don't see any suggestion that he's talking about innate levels of intelligence.

    On the free speech issue, it's worth pointing out that it's consistent with the ideal of academic freedom to insist that authors ought to be tactful when publishing controversial work, and ought to explicitly distance their work from harmful ideologies that it might wrongly be thought to support.

    Alex

    ReplyDelete
  4. Frank Ellis, a lecturer at Leeds uni, got suspended for the same type of research earlier this year.

    (http://education.guardian.co.uk/racism/story/0,,1738151,00.html)

    Our students union (warwick) almost unanimously passed a motion calling for his dismissal. I, along with about 3 others, were the only ones at that meeting to vote it down. The majority of the campus left voted with the motion. I still fail to see the logic in it.

    ReplyDelete
  5. As far as I can tell, the best work on IQ suggests very strongly that it is not innate. It is not learned in the way that college admissions tests in the U.S. like the SAT are. It tests an actual skill set rather than just knowledge. But these are developed skills, and they are developed in certain contexts more than others. There's a reason the raw score for IQ tests for black Americans has gone up significantly each generation. It's basically the same genetic stock doing better each generation. Jews in the U.S. went from pretty low before the first world war to pretty high after the second. It may be that genetics will provide a potentiality that gets realized as intelligence develops, but so many other things affect how far it develops.

    Nevertheless, IQ correlates fairly well with one kind of success, just as SAT tests correlate very well with college success. There is something that it tests, whatever you choose to call it. You can refer to something else as intelligence, but it's then worth figuring out what explanatory role intelligence plays. We have explanatory roles already for what IQ tests and SAT tests measure. If there's another thing we're going to call intelligence, it would be good to know what it is and what it explains, and we don't want to remove whatever we're going to call the thing we've been calling intelligence so far. I'm a big fan of the multiple intelligences theory, but it doesn't mean the classical view of intelligence is wrong, just perhaps that there might be other things worthy of the name.

    ReplyDelete
  6. Very strange. I just got an email saying that my comment failed. Yet it seems to be here. The email contained the entire text of my comment. I've never seen anything like that before. Is this some new feature that they haven't gotten working correctly yet?

    ReplyDelete
  7. (Jeremy - I occasionally get similar messages. I think it just means that your comment failed to post to my "recent comments" list. The latter is a special hack I use, not a standard Blogger feature.)

    Alex - good point, I was jumping to conclusions re: claims of innate racial differences. Though - as you also note - if the researcher was making a rather more innocent claim about contingent skills/intelligence, then he really should take care to distinguish this from the more pernicious claims that could easily be misattributed to him.

    ReplyDelete
  8. I think IQ is heavily related to things like what you eat as a very young child, what you were taught as a very young child and whether your mother had measles or drank too much etc.

    BUT the main tests are also strongly correlated to parental intelligence (I had a discussion with the MENSA fellows about how you can tell the difference if you are in an adopted family and how that can cause problems) and surely it is just unbelievably obvious that there is at least some genetic component to intelligence.

    Also you have to be a little willfully blind not to notice that people from certain countries tend to be different in ways that could be tested from many other countries regardless of wealth (whether IQ tests 'the right things' or not is of course debatable). One key measure is that Koreans tend to be more 'child like' in appearance while many Nigerians appear more 'masculine' and that that there is a similar (although it could be unrelated, although that seems unlikely) effect in their personality and learning habits. That might have only a small effect (or be in either direction although one direction is more likely), BUT to think that such things would have ZERO effect on IQ is rather like climate change denial.

    Also from an evolutionary psychology perspective, IQ profiles (and the average exam results in any NZ class) are just TOO similar to what you would expect for there to be no connection.

    But I do agree that it could be a dangerous thought. Possibly one worthy of opposing not because it is false but just because it is dangerous.

    ReplyDelete
  9. BTW actually propper IQ tests seem to be pretty reliable after a certain age (very surprisingly so).

    Which implies they are developed at a very young age and then not strongly effected by environment.

    ReplyDelete
  10. The proper word for this sort of emotional screaming is not Witch-Hunt but Heresy Trial .

    The author is guilty of heretical thoughts, and MUST be condemned - as quickly as possible, before any more real research is done on the problem.....

    ReplyDelete
  11. This type of research is always fascinating because it can be the corner stone in disproving a whole concept - such as the political correct idea that we are all equal.
    Anyone who is looking to stop research because it might prove something they didn't want proved is obviously no fan of free speech.
    Without this type of research, there will always be a nagging doubt of an unanswered question - enough to create a much larger problem of a flawed debate in the future.
    Personally, I think there is something in this research, although it must be done with culturally stable testing tools. IQ tests seem to be well suited to some cultures better than others and if that is the case then it will only natural that African scores will suffer.

    ReplyDelete
  12. The closing paragraph reads just like something you'd expect to hear from a religious fundamentalist.

    I'm tempted to start a petition demanding that LSE issue a statement denying that people with white skin absorb more vitamin D. After all, this could be used by racists. Some people also suggest that red-haired people tend to be more prone to freckles.

    ReplyDelete
  13. This irrational reaction is reaction to the perceived implications rather than the research. Science should not back down from rational examination of any possibility just because some implications make us uncomfortable.

    One problem is that some people seem to have based their entire argument for civil rights and legal equality on the assumption that all people are basically equal, which is a really tenuous rationale (it is much better to base right on the fact of their humanity, not their IQ)

    Is it possible that large groups are more or less intelligent? It's possible, but we simply don't know. Does it make a difference? It shouldn't. IQ is not (should not be) one's ticket into the human race. Besides, average intelligence says NOTHING WHATSOEVER about any particular individual

    It certainly is not controversial to realize that some groups are taller or shorter, or even that some groups are more subject to alcoholism or certain diseases. It's not controversial that breeds of dogs have somewhat different behaviors.

    People are entitled to rights and respect based on their humanity. Everything else is just variety.

    ReplyDelete
  14. Hello.
    I speak Spanish.
    I don't know write good the english, so so.
    How add your recent comments?
    I nedd that your help me, because I have a blog to help.
    Name of my blog is Ayuda para el Blog (Help to the Blog)
    Do you know spanish?
    To that help me con this post, it have but no understand because a error in recent comments.
    For example:
    Error: hearsay: could not process comment:
    And I want publicar this, but in spanish.
    http://ayudaparaelblog.blogspot.com
    Bye.

    ReplyDelete
  15. In a way, this dispute is kind of nice, since it can help provide people with leftish sympathies a glimpse into what's going on in the minds of all of those crazy folks who think that evolution shouldn't be taught in school (and the like).

    On the one hand, I have a gut feeling that this guy really shouldn't be doing this research, and a vague sense of forboding about what could come from it. On the other hand, I can see that even approaches like Richard's, "condemn him for irresponsibly drawing politically dangerous conclusions on such a flimsy basis" could be a way of defending our preconceptions. Without looking into his methodology in any detail, this response assumes that it must be fundamentally flawed because that just couldn't be the right conclusion. Sure, it's correlational research, but there are ways of doing correlational research that are pretty good at supporting causal hypotheses, or at least ruling out the most likely alternatives (e.g. by making use of time lags, or various kinds of modeling), and we have no idea what he did.

    By the way, Kanazawa really has claimed that there are genetic differences underlying observed racial differences in intelligence. (You didn't think that you could wriggle away from the discomfort that easily, did you?) Look at page 521 (aka p. 10) of this paper, for instance. Here's the relevant snippet:

    Since our ancestors spent most of their evolutionary history in sub-Saharan Africa, it is evolutionarily more familiar to the human brain than the rest of the world, which is more evolutionarily novel. If general intelligence evolved as a means to deal with evolutionarily novel situations [and that is precisely the theory that Kanazawa is advancing in this whole paper], then it follows that it should evolve more rapidly in the rest of the world than in the ancestral environment of the sub-Saharan Africa.

    Of course, he's a level-headed fellow (professor at the LSE, you know), so he goes on to emphasize "It is important to note that the geographical differences in national IQs are not entirely explainable by the differences between the races. ... The difference is ... at least partly geographic, not entirely racial." (His italics)

    ReplyDelete
  16. "Kanazawa really has claimed that there are genetic differences underlying observed racial differences in intelligence."

    Thanks for clearing that one up Blar. As I said, I'd only skimmed the guardian article.

    "this response [i.e. it's implausible] assumes that it must be fundamentally flawed because that just couldn't be the right conclusion"

    I think a stronger case than this can be made. I'm no expert on this area of history, but I think the following is true. People have been trying to "prove" for centuries that non-white ethnic groups are inferior in some respect to white groups. Given that every past attempt has failed, and given that the vast majority of those attempts appear, in hindsight, to be simple racism, there's good reason to give a presumption against any work of this kind.

    That justification is epistemic in nature, and I don't think it needs to be at all ideologically left.

    Of course, in the end, if there's absolutely no way in which to support that presumption, it should be rejected. But until then, it seems reasonable to think that it's likely that IQ is being measured in a western-centric manner, that environmental effects on IQ are being underestimated, or that the methodology was flawed, and so on.

    ReplyDelete
  17. > People have been trying to "prove" for centuries that non-white ethnic groups are inferior in some respect to white groups.

    Most old research is probably (reasonably) ignored based on it being racist or because they use outdated tools. Besides most modern research finds that white ethnic groups don’t have the highest IQ.

    > Given that every past attempt has failed

    I don’t know how one can say that. People who DONT work in that field automatically reject it in the same way many religious people reject evolution but I have seem almost nothing serious that counters the theory.

    As to (anti) work ‘in the field’ I read a book that was supposed to counter the 'bell curve' argument - and it was really just a whole lot of disagreeing with racism (which probably deserves attention) and asserting that there are other effects (which no sane person denies) and not tackling the theory.

    The kry problem I guess is the ‘anti’ side doesn’t even have a coherent theory and if it did it would be easy to rip apart from what we know about genetics.

    > that IQ is being measured in a western-centric manner, that environmental effects on IQ are being underestimated, or that the methodology was flawed, and so on.

    1) IQ seems to be tested in a western neutral way (since they get about average on average)

    2) No sane person thinks the environment has any les than a huge effect for example getting hit in the head by a car will probably take quite a few IQ points off. Genetics and environment have a ‘both 100%’ sort of effect where a human is smarter than an ant because of genetics but an ant is smarter than a dead human because of environment.

    3) Methodology is difficult and the data in those books is not entirely internally consistent (i.e. certain points disagree that shouldn't and there are an almost infinite number of variables that are not controlled for) But that’s a bit like denying evolution because you haven’t controlled for everything.

    ReplyDelete
  18. Sadly, these things are all too common. Political correctness permeates everything, even the scientific world... (sigh)

    ReplyDelete
  19. From david brin (sf novelist)

    http://www.blogger.com/comment.g?blogID=8587336&postID=116371777581926098

    "This is why I offer a very different answer to those “Bell Curve” guys who try to put forward evidence for small differences in average innate ability in various areas, among various races. In abstract, this matter may be of some small intellectual interest. But it has no conceivable beneficial use and potentially devastating effects upon a PROCESS OF REPROGRAMMING that our civilization has underway.

    That process involves the long, slow replacement of extremely vile habits... like the habit of assuming that a person who is a member of a classification group in one aspect, automatically should be judged according to the average of that aspect that is associated with that group. Or, worse, that such a person should have opportunities limited because of false-causal association with OTHER traits that are purely stereotypes of that group.

    There are clear reasons why our co magnon brains developed such habits. But, in a complex and fair civilization, there is no way that such a filthy ways of thinking can do anything but create injustice, resentment and vast waste of human potential, denying us access to an immense portion of out talent pool.

    Here’s the key point. In order to break such habits, it is necessary to romanticize and exaggerate, by insisting that all groups are inherently equal... an exaggeration that will almost certainly be refuted someday, in some way. But I am in no hurry for that day to come! Because, until we are mature, there will be millions among us who will leap upon an opportunity to return to the old ways."

    ReplyDelete
  20. Genius: Great David Brin quote.
    I have to question whether evolutionary psychology has anything to do with IQ. Tooby and Cosmides always emphasize that evolutionary psychology is the science of universal human nature, not of biological influences on human nature, except regarding gender. Also, I would expect hunter-gatherers to need more intelligence than agricultural people, though less foresight.

    My best guess is that northern climates select humans for sheer size like they do other clades, due to insulation benefits from lower surface area to volume and that large brains come along for the ride, bringing small IQ advantages
    http://www.loni.ucla.edu/~thompson/PDF/nrn0604-GrayThompson.pdf

    ReplyDelete

Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)