Yesterday’s Thoughts

June 20, 2005

Two Kinds of People – Genetics

This wasn’t something I knew about or was expecting when I started talking about two kinds of people, but tomorrow’s New York Times has an article on some political scientists’ recent discovery of genetics.

Using twin studies they have determined the heritability of various opinions. The graphic in the article reports on “Genetic Contribution to Variance in Opinion.” Some samples statistics, school prayer – 41%, property tax – 41%, moral majority – 40%, the draft – 38%, foreign aid – 35%, death penalty – 32%, segregation 27%, abortion – 25%.

Based on my earlier entry on two kinds of people it would have been useful if they had extracted any data on the questions of fear of crime or fear of the police, but they only report on single opinions, not the attitudes and beliefs on which those opinions are based.

I don’t make too much of this research based on what I can learn. The original article was published in The American Political Science Review, which I can’t access, so I am relying on the Times’ synopsis which is somewhat contradictory.

According to the Times, the researchers took a large data set that already existed and pulled out 28 questions that the they thought bore on political questions (the infographic only reports on 17). They measured the frequencies with which the identical and fraternal twins agreed on the questions, and calculated the percentage of the correlation which was from genetics alone.

On school prayer, for example, the identical twins’ opinions correlated at a rate of 0.66, a measure of how often they agreed. The correlation rate for fraternal twins was 0.46. This translated into a 41 percent contribution from inheritance.

To say that I don’t understand this paragraph would be a huge understatement. There must be a huge missing piece. Where does 41% come from? What I thought they were saying was that (0.66-0.46)/0.66 is 41%, but that isn’t right, nor is (0.66-0.46)/0.46 which is 43%, so there is some disconnect on the face.

That isn’t the most significant discrepancy in that paragraph. If you ask people any binary question, the odds that they will agree is 50%, so if the paragraph is accurate I would think that the big news from this story was the people who are raised together and not genetically identical (fraternal twins) are more likely to disagree than any two random people off the street. I don’t think is the story, so there must be something else going on.

The rest of the story gets hopelessly woolly for me: changing political parties, Zell Miller, a paragraph about a different study involving responses to Hilary Clinton and taxes.

Basically I accept the premise of the article, there are some genetic predispositions toward various opinions, even if I don’t understand how it comes from the data presented, or what the data is saying. The underlying components of these opinions are things like irritability, fear of strangers, tolerance for ambiguity and desire for novelty which have all had some strong role in human evolution and are probably significantly correlated to many of the things that we call culture.

I’m not sure I would go as far as the final paragraph of the Times article:

The researchers are not optimistic about the future of bipartisan cooperation or national unity. Because men and women tend to seek mates with a similar ideology, they say, the two gene pools are becoming, if anything, more concentrated, not less.

All Americans have some similarities. By hypothesis these similarities are different from those of most other nations and these similarities are tied to genetic factors. For instance, all American’s are descended from ancestors who migrated here for one reason or another, so the desire-for-novelty, desire-to-escape-from-authority, or whatever constellations of traits would be highly selected for in the founding stock. The claim from the article could be restated that there was a certain distribution of gene frequency in the founding populations and over time members on the extremes chose to preferentially mate with others on those extremes, which would lead to a bi-modal distribution in opinions and political discord.

This is an interesting claim. I’d like to see some evidence that wasn’t three levels of abstraction removed from the actual genetic makeup of the population over time. My hunch is that the phenomenon is probably pretty strongly tied in with the American context of the present. You can imagine that contrary opinions could be developed from the same genetic background, and the a single opinion could arise from discrepant genetic backgrounds. There is no gene which dictates that the bearer opposes increasing property taxes, but there could be a gene or combination of genes which would bear on general levels of fear of crime, strangers, or novelty.

What do we get if find a gene for conservatism or liberalism? Does politics just become a matter of reproduction rates? If conservatives have larger families, do they take over the national discourse? Is this the converse of the Freakonomics abortion research?

Sorry, comments for this entry are closed at this time.