From the Journal of Cross-Cultural Psychology:
Michael A. Woodley and Edward Bell
This article examines the hypothesis that although the level of democracy in a society is a complex phenomenon involving many antecedents, consanguinity (marriage and subsequent mating between second cousins or closer relatives) is an important though often overlooked predictor of it. Measures of the two variables correlate substantially in a sample of 70 nations (r = –0.632, p < 0.001), and consanguinity remains a significant predictor of democracy in multiple regression and path analyses involving several additional independent variables. The data suggest that where consanguineous kinship networks are numerically predominant and have been made to share a common statehood, democracy is unlikely to develop. Possible explanations for these findings include the idea that restricted gene flow arising from consanguineous marriage facilitates a rigid collectivism that is inimical to individualism and the recognition of individual rights, which are key elements of the democratic ethos. Furthermore, high levels of within-group genetic similarity may discourage cooperation between different large-scale kin groupings sharing the same nation, inhibiting democracy. Finally, genetic similarity stemming from consanguinity may encourage resource predation by members of socially elite kinship networks as an inclusive fitness enhancing behavior.
A correlation of -0.6 is fairly strong in the social sciences.
Stanley Kurtz, Randall Parker, HBD Chick, and myself have been saying roughly this for awhile. Stephen Colbert's book I Am America (And You Can Too) touches upon cousin marriage.
It would be interesting to measure the effects of topography on the rates of cousin marriage. In the U.S., which has extremely low rates of cousin marriage and high degree of hostility to the very idea, it was notoriously most common among hillbillies. In general, there's been a lot to be said for living in broad river valleys, of which northern Europe has an abundance.
But even up in the hollers, consanguinity was much lower than in much of, say, the Middle East. And Iraq, the land between the rivers, has very high cousin marriage rates.
Here's a converse of this theory that cousin marriage undermines democracy off the top of my head. I don't have any evidence regarding it, but maybe somebody will do a solid study of it someday: a big war is followed by more than a few guys coming home and marrying an old war buddy's sister. And I suspect that the effect is the opposite of cousin marriage: it builds social capital by creating broader-based networks of in-laws.
Unfortunately, I can't even think of any anecdotal evidence for either part of this hypothesis. But it sounds kind of right to me.