« Galileo | Gene Expression Front Page | Handicap the SATs »
September 19, 2003

Genetics articles over @ Dienekes

Dienekes Pontikos has posted some interesting abstracts dealing with genetics over at this site:

  • European Levites share common haplotype says this article. Since the Levys are a priestly caste like Cohens this shouldn't be that big of a surprise. What is strange is that the Levites of Europe might have an extra-Levantine origin[1].
  • Another article states that there has been a shift from polygyny to monogamy-the latter is quite obvious, mostly men today are monogamous (officially), while the latter can be gleaned from the genetic data that indicates that a fraction of males of each generation contributed most of the Y chromosomes to the next generation. The abstract indicates that this applies across all continents, though the sample size (46) seems a bit small. Also, you might be curious about this old article inferring patrilocality from the geographic diversity of Y chromosomes & mtDNA as well as a more recent article (full PDF version) dealing with New Guinea.
  • This paper discusses the secular rise in IQ being the result of heterosis. It presents it as a hypothesis to test along with the conventional environmental explanations.
  • This paper claims to have falsified much of J. P. Rushton's theoretical framework & his conclusions about behavior of various races. See my comment for my opinion on the topic in the thread. In sum, I think we need more study, but I've always had qualms about the details of Rushton's r-K thesis, if less anger & outrage at the data that he presented.

fn1. Bizarro world, but right after I typed this, an orthodox Jewish guy walked into Starbucks!

Posted by razib at 02:38 PM

Actually, I was under the impression that the article *was* challenging Rushton's data. i.e. saying that they could not find cross-cultural racial patterns in sexuality, development, violent crime, etc., etc., in their data analysis.

Posted by: Jason Malloy at September 19, 2003 04:02 PM


you know, i've never put much stock in a lot of rushton's "behavioral" conclusions because they are more subject to personal, historical & social forces (ie; blacks must have had an "r selected strategy" so they had more children in their lifetime). i always focused on twinning rates or IQ data because that seems more concrete....

(not that i'm saying there aren't behavioral differences, but they are so buffeted by non-genetical factors that it is hard to disentangle some of the confounding facotrs-for instance, jews in 18th century poland had higher birth-rates than poles).

Posted by: razib at September 19, 2003 06:13 PM

The problem of course is that reading the abstract doesn't tell us which data of Rushton's they found fault with. Aside from the twinning rates, which seem pretty solid, there is also the Interpol crime data for a large number of countries, along with national income, etc. Now, I have seen some explanations that treat income as an independent variable (or done the equivalent in a fancier way, e.g. by taking income, education, literacy, etc. and treating them as less correlated than they are). Doing that, you can, if you don't look too hard at the data, conclude 'the African nations have high crime because they are poor. Race does not play into it'; but it is neither a simple nor particularly convincing demonstration.

Posted by: bbartlog at September 19, 2003 06:33 PM

bbart-yeah, true-but the r/K idea is probably not valid in the conventional way-and rushton's theories about WHY there might be racial differences have always struck me as ad hoc.

Posted by: razib at September 19, 2003 06:59 PM

Surely Norwegian fsrmersw circa 1850 were incredibly crime-ridden, because they sure were poor.

That's sarcasm.

Posted by: gcochran at September 20, 2003 08:34 AM

but it is neither a simple nor particularly convincing demonstration.
Why not? Reasons, please, reasons!

Surely Norwegian fsrmersw[sic] circa 1850 were incredibly crime-ridden, because they sure were poor.
Please explain away the criminality of Jewish and Italian immigrants in the early half of the 20th century, if poverty has nothing to do with crime - think Meyer Lansky, Bugsy Siegal, Al Capone, et al. And do you have any data indicating that Norwegian slum-dwellers weren't prone to criminality? It's a bit difficult to commit crimes when you're a farmer out in the middle of nowhere, isn't it?

Posted by: Hanno Buddenbrook at September 20, 2003 02:50 PM

Dunno about Hanno's last point - they can always bugger the sheep.

BTW, is there really someone out there (apart from me) who still reads Thomas Mann?

Posted by: David B at September 22, 2003 11:55 AM

I don't have to put much effort into explaining Jewish criminal activity in the first half of the 20th century, because there wasn't much. Those tenements in the Lower East Side had really low crime rates. I recall reading of a project in which some students looked at old New York crime records for that area - they were amazed that you seldom saw an offense more serious than running a vegetable cart without a permit.
It's perfectly possible to have poor city-dwellers with low crime rates. At one time, there was a census district in San Francisco that was the poorest in the state, had the highest incidence of TB and the lowest crime rate. Chinatown.
The idea that poverty causes crime is stupid - and at this point, boring.

Posted by: gcochran at September 22, 2003 07:34 PM

On a more interesting note, that Levite chromosomal study shows what can happen from drift. I don't that drift was very important in the Ashkenazi population as a whole (and I have the numbers to prove it) but the Levites were only about 4% of their population, and the effectvie population size for a Y chromsome is four times smaller than that of an autosomal chromsome - so the effective N for the levite y chromsome was 1/100th of the _effective_ Jewish population size, which was itself probably no more than a third of the Ashkenazi population at any given point in time.

Posted by: gcochran at September 22, 2003 07:35 PM