UPDATE: Is Ideology About Status? was an unacknowledged inspiration for this post.

There have been many theories of conflict within a society that we may see reflected in politics. Most famous perhaps is that of economic classes, associated with Marxism. Some libertarians (and Calhoun) present their own of tax-payers and tax-eaters, though Bryan Caplan rejects that in favor of a jock vs nerd theory. As I’ve mentioned before, I agree with Ilya Somin that all of them are wrong. Many on both sides seem to think the divide is based on thinking with your head vs your heart (or “gut” as a tv personality popular with those kids these days puts it), but Caplan has found that is also wrong. A related divide many might imagine is the smart vs the stupid, but the ideologues on both sides tend to be similarly smart while moderates are generally quite dim (there may of course be temporary fluctuations, but they are not an enduring fundamental of politics). These may be on the right track though, because our political leanings are significantly heritable and the brain seems a good place to focus. This idea inspired Lee Sigelman to muse that rising rates of obesity among pregnant women may result in conservative dominance (hiss, the biological-yet-nurturist heresy). It’s also behind the Genetic Warfare site/post I pointed out a little while back. What I’m going to postulate is a divide between the intelligent based on the two immediate sub-components of g: verbal and visuo-spatial intelligence.

This occurred to me as I was reading the Bell Curve. I’ll excerpt the portion which set me thinking and should also explain the basic concepts: “A full-scale IQ score is the aggregate of many subtests. There are thirteen of them in the Wechsler Intelligence Scale for Children (WISC-R), for example. The most basic division of the subtests is into a verbal IQ and a performance IQ. In the white samples, the verbal and performance IQ subscores tend to have abot the same mean, because IQ tests have been standardized on predominantly white populations. But individuals can have imbalances between these two IQs. People with high verbal abilities are likely to do well with words and logic. In school they excel in history and literature; in choosing a career to draw on those talents, they tend to choose law or journalism or advertising or politics. In contrast, people with high performance IQs – or, using a more descriptive phrase, ‘visuospatial abilities’ are likely to do well in the physical and biological sciences, mathematics, engineering, or other subjects that demand mental manipulation in the three physical dimensions or the more numerous dimensions of mathematics.” The prototypical high-verbal IQ types sound a hell of lot like Mencius Moldbug’s bete-noire, the Brahmins. As an introverted nerd, I am drawn to other similar types who are quite prideful in their abilities and often resentful or contemptuous of the clever-talking flim-flammers and “people persons”. And it seems like everybody hates lawyers. I am reminded of the Charles Murray quote “Who wants to be an elephant?“, though without the geniality. They like to say “The numbers/physics don’t lie” and “You can’t argue with a computer/reality”. We can see the bitter narcissism of the engineer in comments like this one, directed at those Wall Street masters-of-the-universe that just screwed up the financial markets. I have to admit I indulge in this myself. In the back of my mind lurks a variety of “producerism” in which inventors are the prototypical producers and bullshitters are prototypical parasites. This is tempered with the knowledge that engineers and applied scientists are especially vulnerable to pseudo-science that Ivory Tower egghead theorists have to shoot down. The “practical man” is often enough led by “theory” as much as a theorist, only illogically, incoherently and unknowingly.

With that out of the way let’s move aside from any discussion of the relative merits of these intellectual abilities and who deserves to be on the holding or receiving end of the stick. Does my idea hold water any more than the theories I dismissed at the beginning? I don’t have data on hand that can analyze this directly, so I’m going to throw out some things that might raise your Bayesian assignments of probability (although if they are underwhelming they should lower it from where you had adjusted it before reading). Not all groups have the same profile when it comes to these sub-components. Men and women differ, for example. They are generally believed to have the same average IQ, but men are more visuo-spatial and women more verbal. As Larry Summers pointed out to his detriment, men also have a higher variance when it comes to IQ than women. In politics there is some gender-gap, though it shows up primarily among the unmarried. Unmarried women lean left. Might this be because of the variance aspect? I don’t think so. The GOP seems to be a more homogenous party, which draws its support from the middle class (not the working class, as some stupidly think nowadays) with some college education but not grad school. I dispelled myths about that here. The Dems draw their support both from the poor/working-class and from some quite wealthy coastal elites in the richer Blue States, such as the finance guys attacked above. I don’t actually know if finance (or programming, for that matter) attracts verbal rather than visuo-spatial sorts, but I figured I’d throw that out there. So if this was simply about variance we might expect men (who as Roy Baumeister explains are high-variance in general) to go with the high-variance party, but they don’t.

It’s not only genders that differ along these sub-components. Ethnic groups do as well. However, the difference in their average IQ scores makes this tricky. The highest intelligence is found among Ashkenazi Jews. They are especially gifted verbally (Barry Sonnenfeld in his commentary for Miller’s Crossing points out the odd unusualness of Jewish cinematographers in Hollywood). Jews in the past had a similar profile to other “white ethnics”: they arrived as immigrants without much to their name, were enlisted by the Democratic urban machine and worked their way up in the face of nativist hostility (likely exaggerated in the retelling). What’s odd about Jews is that they are the most succesful ethnic group in America in economic terms, but vote for the party of the poor. The saying goes that they “Earn like Episcopalians and vote like Puerto Ricans”. Except Hispanics with higher income would be significantly to their right. According to Ann Coulter in Slander (yes, I’ve read her, is that a crime) Jews bloc-vote at the highest rate of any comparable demographic group. Second highest in her list are blacks, the economically least succesful ethnic group. Far different than Jews when it comes to IQ, yes? Except their advantage is also in verbal abilities. There is an ethnic group that seems to have a higher IQ than gentile whites, but with the opposite advantage in sub-components. These are East Asians. At present they lean left, but this may the temporary result of Christian-inflected culture war. They went right in 1992. Even today they aren’t as left-wing as other minorities like Jews or blacks. This would be better for my case if they went right, but the general party of outsiders/alienated effect could be swamping it. I’ll add here that American Indians (related to Asians way back) have a similar visuo-spatial orientation (and although their average IQ is lower than whites, northern tribes get the closest among peoples outside Europe or Asia) and lean right. Despite having very high poverty rates and being mistreated to an arguably greater extent than any other ethnic group in the country, they are very patriotic and are the most overrepresented in combat casualties relative to their population share. Perhaps due to their extreme learning experience they are also the most anti-immigrant.

All this talking about groups may seem like too much of a reach when I should be focusing on cognitive specialties. So let’s look at academia. Razib has some charts here on political leanings by subject. Subjects like English, History, Philosophy and even Theology are among the leftiest, and all draw on verbal smarts. Engineering and the hard sciences which draw more on visuo-spatial abilities are to their right, although this only means a little more conservatives and mostly just fewer liberals leading closer to parity. I would have expected Music to be leftier (don’t famous musicians have Hollywood/”creative class” type politics), but maybe Agnostic’s argument about the visuo-spatial component is shining through.

I don’t place a terribly high confidence in this theory and it could easily be blown out of the water by some data directly analyzing the correlation between the sub-components and political leaning. Anyone that can do so is encouraged to fire at will.
Addendum: Razib used the GSS to analyze the same question but came up with more ambiguous results than Half Sigma.