Saturday, January 03, 2009

More thoughts on The Economist & Sociobiology

I had  a few other thoughts about the recent run of sociobiological (or EP-influenced) articles in The Economist.

One was the prominent role played in them by the University of New Mexico evolutionary psychologist Geoffrey Miller.

Miller's main lines of thinking are 1)that the standard measure of intelligence (g) is a manifestation of some broader and more inclusive fitness; and 2) that the brain's growth--and the development of much of what we hold to be "human"--was driven by sexual selection rather than plain old natural selection.

I've taken a half dozen or so rigorous IQ tests in my lifetime, as well as a huge battery of other standardized tests that are built--more modestly--along the same lines, and frankly I have a fair deal of skepticism about how effective they are as a measure of anything. 

Because they are culturally skewed? Maybe, but in a way different than most people who write on this issue seek to claim. Standardized tests require a great deal of planning, concentration and effort to complete to the best of your ability--my own SAT-family scores fluctuated wildly from the disaster of the hung-over PSAT to the glory of the very well prepared GRE. My scores--all prior to the re-curvings of recent years--were hundreds of points apart, even on the same test. I improved a friends score on the GRE by 400 points by training her in basic test-taking method and running her through several timed mock-tests.

So, obviously, a lot more goes into the scores on the test than anything we would usually call intelligence--mental, emotional and physical preparedness; experience with similar tests; motivation . . .

The work of Miller & his colleagues is probably controlled for the wider variations in these factors, so i have no doubt that they may be on the trail of some more inclusive "fitness factor" and will follow future studies with interest. BUT the way this work will be applied by amateurs in the field--and there are LOADS of them, ranging from loutish loudmouths on the internet to editors for the Economist to scientists from related fields who wish to be able to say something about social policy, affirmative action, urban crime, whatever. But few of these amateurs will understand Miller et al's work well enough to see its limitations. And Miller, et al. will be rather reluctant to disabuse the amateurs of their illusions because those limitations tend to make the work rather uninteresting to a broader public (those very interested amateurs included).

In short, few people are interested in intelligence testing or in a more inclusive fitness factor unless it can be used as a tool in contemporary ideological and political discussion.

One place we'll probably soon see Miller's work employed is in combination with the work of Richard Lynn, who has used global IQ test scores to explain global economic disparities. Conveniently, he can draw on exceedingly low IQ scores for sub-Saharan Africans, which are based on tests which are the best we've got but, frankly, utter crap. Why utter crap?--because they are completely uncontrolled for factors like those I list above--familiarity with the test, motivation to do well, the effect of local rules of etiquette on face-to-face questioning, etc.

So, Miller et al.'s inferences--that their IQ scores may be related to more general fitness will be uncritically extended to, essentially, all existing IQ scores, no matter how crappy they may be. We will soon be being told that the economic outcome we have seen over the course of industrial modernization--rich west, rising east, decrepit Africa--is nothing more or less than the verdict of evolution on our species.

And Miller, if he holds true to form for the newly famous handsome young scientist, will smile, say nothing and negotiate a bigger book deal.

Thursday, January 01, 2009

Sociobiology again?

Wow, it's been a while . . .

More news on the Economist:

The magazine has recently been on a bit of a sociobiology kick with several recent unsigned articles chawing the collective ear about the biological basis of human behavior. Here's what I've found over the past year or so:

# LEADERS: Evolution
Of music, murder and shopping
It is time to turn to Darwin to explain human behaviour
Dec 18th 2008

Why we are, as we are
As the 150th anniversary of the publication of “On The Origin of Species” approaches, the moment has come to ask how Darwin’s insights can be used profitably by policymakers
Dec 18th 2008

# CHRISTMAS SPECIALS: Human evolution
Why music?
Biologists are addressing one of humanity’s strangest attributes, its all-singing, all-dancing cultureDec 18th 2008

# THE WORLD IN: Leaders
Shocking science
Geoffrey Carr expects scientists to provide a year of celebrations and screams
Nov 19th 2008 Web only

Balls and brains
The quality of a man’s sperm depends on how intelligent he is, and vice versa
Dec 4th 2008

# CHRISTMAS SPECIALS: Beauty and success
To those that have, shall be given
The ugly are one of the few groups against whom it is still legal to discriminate. Unfortunately for them, there are good reasons why beauty and success go hand in hand
Dec 19th 2007

And there's more naturally, much of it the work, one would suspect, of the magazine's science editor Geoffrey Carr.

I'm not at all opposed to the theoretical notion that our behavior, our morals our civilization & culture are all ultimately based on biology. I'm just skeptical that biology should be the thing we're focussing on.

The Economist makes a lot of the fact that murder rates, pretty much universally, peak at the time when there is the most competition for mates--teens to twenties--and rapidly fall after that. Interesting, but what does biology bring to it?

We already were perfectly well aware of the fact that violence was a young man's game--demographic research has contributed to social policy for quite some time--how does the notion that it may have an evolutionary explanation help policy makers?* I don't think it does.

(*Rather, we should say, the notion that evolutionary biologists can make up a story ex post facto that would seem to explain the phenomenon from an evolutionary perspective.)

And what is the evolutionary explanation for the more remarkable number on their graph--that the murder rate among males is 30 times higher in the US than Great Britain, across all age groups? 

The other big point the article propounds is that even in an egalitarian society, outcomes between different population groups--between men and women or blacks and whites--may not be equal. In other words men and women, or blacks and whites may turn out to be different even when treated fairly. One would suppose that this theoretical proposition is pretty uncontroversial as far as men & women go (many of us have first hand experience of enduring differences between the sexes). Many might say that the subject of enduring intellectual differences between the races is an "unsafe" notion to bandy about, but even staunch enemies of this line of thought--SJ Gould, for instance--acknowledged that this is a theoretical possibility.

(Virtual marginal note: A lot of commentators make a lot of hay out of this attitude regarding "unsafe" thinking and (tiresomely) cry "political correctness" whenever it comes up. But wherever you ultimately fall on the idea of censorship and self-censorship, the anti-political-correctness folks don't some to realize that this conversation isn't just an intellectual game for some people. There is a long legacy of majorities using this sort of discussion to suppress minorities, sometimes with the very gravest consequences. The upside of free and open discussion of even heinous ideas can seem pretty modest when compared to the downside. And certainly the utter lack of consideration of what the stakes are at blogs like gnxp fairly cries out the prolonged intellectual adolescence of those writers.)

But the strong implication of the Economist articles--that we have already arrived at a time when we cannot assume that strong differences in "outcomes" between population groups are evidence of unfairness is patently and obviously false. Just a few years ago, people were making strong arguments on behalf of Larry Summers by pointing out that it was obvious that there women were innately inferior at mathematics to men. Now it appears that that isn't the case--very egalitarian societies show little evidence of an "outcome gap" in math--we only thought we had reached a state of fairness with residual unequal outcomes. Which, frankly, was clear to anyone who looked at the situation without ideological blinders.

The bit about race in the same (Dec. 20th) article is rather bizarrely credulous. We are told that people generally register three things about a person they meet: sex, age and race.
But Dr Cosmides and Dr Tooby [two of the founding figures of the latest push to apply biology to policy] pointed out that before long-distance transport existed, only two of those would have been relevant. People of different ages and sexes would meet; people of different races would not.

The two researchers argue that modern racial discrimination is an overstimulated response to what might be called an “alliance” detector in the human brain. In a world where the largest social unit is the tribe, clan or what-you-will of a few hundred people, your neighbours and your other allies will normally look a lot like you, and act similarly. However, it is known from the study of modern hunter-gatherers, and inferred from archaeological evidence about ancient ones, that neighbouring tribes are often hostile.

Though an individual might reasonably be expected to know many members of his tribe personally, he would probably not know them all. There would thus be a biological advantage in tribal branding, as it were. Potential allies would quickly identify what marked them out from others, and what marked others out from them—and, because those differences would probably be small, the detector would need to be very sensitive.

So, in a world "where the largest social unit is the tribe, clan or what-you-will of a few hundred people" In a world where you spend every living moment associating with pretty much the same 2-300 people, we don't recognize them all? Admittedly, we wouldn't know them all well, but we would be able to recognize them. There would be no need for a super-sensitive "lineage detector."

Our awareness of race--and many of our other traits--has a more complex explanation that people like Cosmides and Tooby are willing to admit. they can't all be traced back to factors in some notional "early human environment." Some have roots that go back beyond early man, and beyond primates even. And many traits have a lot more cultural input than they'd like to admit.

In short, the origins of human behavious are complex and are not the place to turn for stunning insight about what we should be doing now. In fact, biological questioning of the the origins of our behavior seems to a) give us answers we already knew about, either because that's the story biologists wanted to tell us or because we all have an inherent ability to judge ourselves; or b) give us no good answers at all, as biologists radically simplify complex phenomena in order to imply they've got the goods when they don't. They tell us they know the etiology of social phenomenon X, when they don't really appreciate X as a phenomenon at all. How can they tell us how X arose, when they don't even know what X is?

All they really want to do is to say X is a function of my field of specialty, and you really should stop paying attention to X and start paying attention to me.

For most cases of X, X is more interesting than evolutionary psychology. And the study of social phenomena as they manefest themselves to us today is infinitely more likely to provide us with answers and progress than the study of evolutionary biology.

And as economics suffers from a crisis of credibility, it may seem only natural that the Economist would turn to economics' cousin evolution for a boost. But this totalizing system is not much better that the others we suffered through in the last century. One would hope the Economist would in the final instance make a stand for the particular against the sloppy totalizers.