Saturday, November 26, 2005

A Genetic and Cultural Odyssey   posted by Razib @ 11/26/2005 01:21:00 PM

As promised, I did read A Genetic and Cultural Odyssey, the biography of human geneticist L. L. Cavalli-Sforza. I have to say that it was a very uneven work, but neither of the two authors are writers by profession (an anthropologist and geneticist). The main thing I took away from the book was that I will be curious to check out Cultural Transmission and Evolution,1 which is a theoretical work that attempts to transpose the models of population genetics into a cultural framework, and Bryan Sykes is a grand-standing ass. Since there are other data points in regards to Sykes, I am prompted to believe Cavalli-Sforza's side of the story.2

There are short sketches of Cavalli-Sforza's early work in bacterial genetics that might surprise some (I was aware of this from reading I'd done on R.A. Fisher), but there seemed to be a disproportionate focus on his work in relation to anthropologists. In particular, his somewhat scientific-reductionistic attitude toward the discipline, in contrast with the zeitgeist of the field within the past generation which has focused on criticism and contextualization, almost to the neglect of model building or positive assertion. Of course, the race-does-not-exist chapters are somewhat hilarious in their attempt to dance around semantically. A key issue might be that race-does-not-exist in the way that it is conceived of colloquially, idealized discrete types, but it does exist as predictable lumps within the soup of clinal variation. Sometimes the authors make embarrassing statments, for instance, "Pygmies are short because of the wet tropical climate." The adaptive story is probably a lot more subtle than that. This is also pretty egregious in light of their emphasis on the neutral portion of the genome, as opposed to functionally relevant loci. Some of the unevenness might have been due to the fact that the two authors wrote alternating chapters, but that's an explanation, not an excuse.

Of course, how much can you compress into ~200 pages? The major problem is that I'm not sure the authors knew what they wanted the book to be, the mathematical footnotes were interesting, but anyone who would take an interest in such details would almost certainly wince at the superficiality of the prose used to impart the concepts.

Addendum: One good thing about this bio though is that it shows how much Cavalli-Sforza's life intersected with other prominent biologists, Joshua Lederberg, R.A. Fisher and Anthony Edwards. Also, Lynn Jorde has reviewed the book, similar general impression from what I can gather.

1 - What utility do mathematical models have in discussing something as amorphous and intractable as culture? The reality is that mathematics can often clarify exactly what you are trying to get at, so the upside might be precision in communication more than a utility in regards to predictive value (that is, predictions made from the models might be trivially obvious).

2 - Why do I link to The Seven Daughters of Eve? First, it is copiously cited in the semi-popular literature. Second, if you take it for what it is (rather than what Sykes tries to claim it), I think it is a decent survey of the H. sapiens mtDNA literature circa 2000.