Monday, November 30, 2009

"Old Europe"   posted by Razib @ 11/30/2009 09:43:00 PM

A Lost European Culture, Pulled From Obscurity:
The little-known culture is being rescued from obscurity in an exhibition, "The Lost World of Old Europe: the Danube Valley, 5000-3500 B.C.," which opened last month at the Institute for the Study of the Ancient World at New York University. More than 250 artifacts from museums in Bulgaria, Moldova and Romania are on display for the first time in the United States. The show will run through April 25.

At its peak, around 4500 B.C., said David W. Anthony, the exhibition's guest curator, "Old Europe was among the most sophisticated and technologically advanced places in the world" and was developing "many of the political, technological and ideological signs of civilization."

The Horse, the Wheel, and Language: How Bronze-Age Riders from the Eurasian Steppes Shaped the Modern World is a very interesting book. One of the problems with pre-literate civilizations is that they're only accessible via archaeology, which is a field averse to system-building or theorizing. But it is likely from what we know of pre-literate cultures which Europeans encountered that lots of stuff happened. Perhaps ancient DNA will help resolve some of these questions, at least establishing whether peoples or just pots were on the move.


Sunday, November 29, 2009

Are over-leveraged counties seeing an increase in food stamp usage?   posted by Razib @ 11/29/2009 11:33:00 AM

Since The New York Times put up the csv file which they used to generate their maps of food stamp usage, I thought I'd look at the data a little closer. In particular, look at this graphic of change in food stamp usage by county (dark equals more usage):

I was curious about this part from the story below::
While use is greatest where poverty runs deep, the growth has been especially swift in once-prosperous places hit by the housing bust. There are about 50 small counties and a dozen sizable ones where the rolls have doubled in the last two years. In another 205 counties, they have risen by at least two-thirds. These places with soaring rolls include populous Riverside County, Calif., most of greater Phoenix and Las Vegas, a ring of affluent Atlanta suburbs, and a 150-mile stretch of southwest Florida from Bradenton to the Everglades.

Thanks to the Census I happen to have 2007 housing value and household income data. Also though it would be interesting to compare with obesity and diabetes rates. Scatterplots & correlations (r) below.

It does indeed seem that food stamp usage has been increasing in higher income and property value counties. The Census data I used above were collected between 2005-2007, during the height of the late great property bubble. But when I took the ratio of property value by income as a rough proxy for being over-leveraged it didn't seem to add much.

When I took the partial correlation of home value and increase in food stamp usage controlling for income, it was only 0.11. Here are some other correlations controlling for income:

% on food stamps - obesity = 0.33
% on food stamps - diabetes = 0.44
% of whites on food stamps - white diabetes rates = 0.36
% of whites on food stamps - white obesity rates = -0.05

There's an obvious correlation between black proportion in a county and food stamp utilization. r = 0.43. So using proportion of blacks as a control:

% on food stamps - obesity = 0.43
% on food stamps - diabetes = 0.51
% on food stamps - white diabetes rates = 0.43
% on food stamps - white obesity rates = 0.06
% on food stamps - median household income = -0.71

It does seem to be correct though that food stamp utilization has been shooting up in more affluent communities. But if it is true that well over 90% of those eligible in places like Missouri are already using food stamps, while only 50% of those eligible in California are, it makes a bit more sense. In wealthier communities likely more people go in and out of eligibility and so never need to make recourse. In contrast, in regions where people are immobile and poverty is chronic there isn't as much scope to increase the program because most people who are eligible are already on it. That probably explains the triangular geometry of the scatterplot, very low on the affluence latter social services seem to have soaked up all eligible individuals, leaving little room for increase with the recession.

Note: Estimates are white obesity are based on state level variation. Estimates of white diabetes rates are based on national level variation. These two variables need to be appropriately down-weighted in terms of confidence of their accuracy, especially the second.

Update: By coincidence, a reader noted this similarity of maps this morning:

Labels: , , ,

The grain dole of America   posted by Razib @ 11/29/2009 10:17:00 AM

Ben points to the a new article in The New York Times, Across U.S., Food Stamp Use Soars and Stigma Fades. The county-by-county data are of interest. I've just snatched the csv file, which they made available. Andrew Gelman has a modest critique of the assertion that 50% of children are on food stamps at some point in their childhood. The variance in utilization rates of the program by region (50% in California vs. 98% in Missouri) of those eligible, as well as the near saturation of utilization in much of the Black Belt and highland South (the Appalachians and the Ozarks), implies to me that while in some American subcultures the program is seen as a stop-gap in others it is a background condition of life. A minimum income guarantee or grain dole basically. Also, I recently heard a radio interview with Kevin Concannon, an under secretary of agriculture. In response to criticism of misrepresentation of the results of reports of hunger in America his stance was basically "statistics, schmamistics."

The reason that I'm fixating a bit on the issue of hunger in America is that we're also told that there's an "obesity epidemic" in this country, in particular among the lower classes. Often from the same policy elites who point to long lines at soup kitchens as evidence of a surfeit of food! To be hungry sometimes is uncomfortable, I know this personally, I am hungry sometimes. Though for me it has to do with the fact that I don't think that the immediate response to hunger always has to be food to satiate the pangs (I don't like to eat past a certain hour). Nutritional belt tightening isn't necessarily a bad thing, remember that the Great Depression saw an increase in life expectancy.


Saturday, November 28, 2009

The white vote for Obama, by county & correlates   posted by Razib @ 11/28/2009 03:01:00 PM

A friend of mine who was looking at the distributions on obesity and diabetes wondered about their political correlations. To do that and add anything new it seems that it would be best to estimate the white vote for Barack Obama in 2008 by county. This is how I did it:

1) I looked at the exit polls for each state, which has breakdowns by race for each candidate.

2) Since the white vote probably varies more county-by-county than the minority vote, especially the back, I used the state level exit polls and assumed that the minority vote in every county could be predicted by the state level exit poll. So for example, in New York the exit poll suggest that 100% of blacks voted for Obama. So I would weight appropriately.

3) I also weighted by national turnout numbers. In other words, whites were a little overrepresented in the electorate, blacks equal to their demographic weight, and Asians and Latinos underrepresented. So:

% Obama in county = (White turnout)(White %)(White proportion) + (Black turnout)(White %)(Black proportion) + (Latino turnout)(Latino %)(Latino proportion) + (Asian turnout)(Asian %)(Asian proportion)

Many states did not have results for ethnic minorities in the exit polls, so the white vote estimate is identical to the real results in many counties (the correlation between my estimate and the real returns is on the order of 0.99-0.98 north of 85% or more non-Hispanic white). In places like Mississippi where most everyone is either black or white, we can probably be sure that blacks voted well in excess of 90% for Obama, I think the estimate for whites is probably pretty good. The main issue is with Latinos, who I suspect seem to vary quite a bit more than blacks (in fact, they probably tend to follow whites in voting except that they're more Democratic all variables controlled (again, I had to discard some counties were negative proportions pop up because Latinos are more Republican locally than on the state level).

Fist some maps, then some correlations. Again, note that red is below and blue above whatever threshold I'm using (usually median).

For the correlations, "est" means my estimate. Reduce the confidence in those correlations accordingly, as my data analysis hasn't gone through peer review! (until you comment)

Here are the summaries for Obama vote estimate:

1st quartile = 0.2240
median = 0.3591
mean = 0.3587
3rd quartile = 0.4754

Since Democratic votes are concentrated in a few highly populous counties the low proportions are not a surprise. Lots of counties with few people are anti-Obama.

White Obama Vote (est)- White Diabetes Rate (est) = -0.26
White Obama Vote (est)- White Obesity Rate (est) = -0.29
White Obama Vote (est)- White Birth Rate = -0.17
White Obama Vote (est)- College Degree = 0.42
White Obama Vote (est)- Median Household Income = 0.28
White Obama Vote (est)- Median Home Value = 0.40

(for whites ancestry are proportion of whites, i.e., Irish/White = Irish proportion)
White Obama Vote (est)- Origins in Britain & Ireland = -0.24
White Obama Vote (est)- English = 0.08
White Obama Vote (est)- Irish = 0.37
White Obama Vote (est)- Scots Irish = -0.13
White Obama Vote (est)- American = -0.50
White Obama Vote (est)- German = 0.38
White Obama Vote (est)- Scandinavian = 0.30

Partial correlations controlling for college degree rate:

White Obama Vote (est)- White Diabetes Rate (est) = -0.30
White Obama Vote (est)- White Obesity Rate (est) = -0.29
White Obama Vote (est)- White Birth Rate = -0.20
White Obama Vote (est)- Median Household Income = 0.00
White Obama Vote (est)- Median Home Value = 0.17
White Obama Vote (est)- American = -0.46
White Obama Vote (est)- German = 0.36

Partial correlations controlling for median household income:

White Obama Vote (est)- White Diabetes Rate (est) = -0.36
White Obama Vote (est)- White Obesity Rate (est) = -0.33
White Obama Vote (est)- White Birth Rate = -0.21
White Obama Vote (est)- Median Home Value = 0.30
White Obama Vote (est)- American = -0.52
White Obama Vote (est)- German = 0.35

The correlation between the white Obama vote and the proportion of blacks within a county is in the range of -0.30 to -0.40 (on the high end), even controlling for income and such (the blacker the county, the fewer whites voted for Obama). Interestingly when I control for black proportion the German correlation for voting for Obama drops a bit to 0.26, and the American correlation drops from the other direction, -0.39. Race can explain some, but definitely not all of these inter-ethnic differences in the white vote.

Poking through demographic data, a few things always seem to crop up:

1) Texas isn't quite like the rest of the South. It is more Republican on the federal level than racial polarization into a white and black party would predict.

2) The Latino counties in Texas are hard to fit into a model which is derived from conditions in the rest of the country. They have lower morbidity and are somewhat more conservative than Latinos elsewhere (in fact, their morbidity is lower than whites in many regions of the country). I often have to discard these counties because estimates using state level parameters are weird (in the case of white voting patterns or diabetes rates, negative values).

3) There's stuff going on in Appalachia which needs to be explored. I'm going to analyze Appalachian counties specifically in the near future. I had assumed that aside from outliers like Asheville Appalachia was relatively homogeneous. Not so.

Labels: ,

Reality check on American "hunger"   posted by Razib @ 11/28/2009 12:12:00 AM

Hunger here vs. hunger there:
There has been a fair amount of buzz lately (examples here, here, here, here) about "food insecurity" in the U.S. According to the Reuters headline, one in seven Americans is short of food. In looking into the data, what has surprised us is how different the meaning of "hunger" is when we're talking about the U.S. vs. the developing world.

Developing-world hunger: 30% of children underweight


The "food insecurity" categories are derived from people's answers to questions like "We worried about whether our food would run out before we got money to buy more" and "We couldn't afford to eat balanced meals" (full list on pg 3). The details of the answers are found on page 45:


Note in particular the difference regarding children. In the developing world, as shown above, severe child hunger is rampant. In the U.S., even in "food insecure" families, it's extraordinarily rare for children to go hungry even temporarily. And indeed, World Bank data estimates that 1.3% of U.S. children under 5 are "underweight" - less than the 2.3% that would be expected in a fully normal distribution.

On the one hand the poor supposedly live in "food deserts" and so get fat. On the other hand, there's a lot of hunger in America. Something doesn't make sense. As someone whose family is from Bangladesh I have seen plenty of hungry people face to face. They look really hungry. If you're really chronically hungry you can't mask it with a stiff upper lip, you just look starved out, and a bowl of rice with salt is a luxury. They're really short too. When I went to Bangladesh in the late 1980s for a visit I was much taller than many adult beggars despite being a pre-teen, and I was always around the 50th percentile on the height distributions in elementary school.

The fact that fewer American children are very light than would be expected under a normal distribution is also interesting. Assuming weight is a quantitative trait, like height or IQ, one would expect the deviation from the normal distribution to produce a "fatter tail", not an attenuated one.


Friday, November 27, 2009

Diabetes and obesity   posted by Razib @ 11/27/2009 01:09:00 PM

Update: I made a major error in the algebra of estimating "white diabetes rates" per county. So the last set of correlations was junk. I think fixed the issue. Thanks to "bayesian" who noted that something was off with them.

The CDC provides data on diabetes by county as well as obesity.

Some Correlations:

Diabetes-Obesity = 0.72
Diabetes-Black = 0.65
Diabetes-Latino = -0.14

What's going on with the last? Latinos, in particular Mexican Americans, are more susceptible to diabetes than whites. So it must be that in counties where there are many Mexican Americans, white have particularly low prevalence of diabetes.

Other correlations:

Diabetes-Obama Vote = -0.01
Diabetes-College Educated = -0.46
Diabetes-Median Household Income = -0.45
Diabetes-Median Home Value = -0.42

I'm struck by the fact that the correlations are higher than for obesity (if you think about it in terms of r-squared, the square of the correlation explaining the variance of Y by X, it's even more striking). Probably has to do with the fact that only a subset of the obese are diabetic, as diabetes is a more extreme manifestation of morbidity. Let's control for the % black in a county with partial correlations:

Diabetes-Obesity = 0.63
Diabetes-Obama Vote = -0.28
Diabetes-College Educated = -0.52
Diabetes-Median Household Income = -0.41
Diabetes-Median Home Value = -0.43

Not much change in the correlations really. Also, now there is a modest correlation between political liberalism and lower levels of diabetes now that the black proportion is controlled (the correlation with black proportion controlled for obesity and Obama vote is -0.24, same magnitude and direction).

I also tried to estimate white diabetes prevalence by county. The national data suggest that blacks are 1.7 times more likely to be diabetic, and Latinos 2 times more likely. Obviously there's going to be some variance for these two groups, so I don't know how useful this estimate for whites is going to be. But, it should put into stark relief the negative correlation between the proportion of Latinos and white diabetic rates (note: again, Latinos seem to vary quite a bit and there are many counties along the Mexican-American border, as well as on the East Coast, where Latinos are so far deviated from the aggregate risk that I had to dump the data).

Here are some correlations (again, white county proportions are estimates):

White diabetic proportion-White obesity rate (estimate from previous post) = 0.47
White diabetic proportion-College Educated = -0.46
White diabetic proportion-Obama Vote = -0.18
White diabetic proportion-Median Household Income = -0.39
White diabetic proportion-Median Home Value = -0.44

OK, enough with correlations. Maps. Diabetes for all groups:

Now, my estimates for whites:

I think the assumption of an invariant relationship between white and non-white rates (i.e., blacks = 1.7 X whites) is causing problems. The white areas underneath the median suspiciously concentrated in the Black Belt.

So let's just focus on counties which are 85% or more white:


Where the fat folks live   posted by Razib @ 11/27/2009 12:44:00 AM

Since it's after Thanksgiving and I'm feeling bloated, I figure a follow up to the post on obesity and diabetes might be apropos. I want to focus on obesity. I have the raw county-by-county data, but obviously it isn't broken down by race. But, I do have the proportions for reach race by county, and, the CDC provides state-by-state breakdowns of the proportion of obese by race. So I decided to "estimate" the proportion of whites obese by county.

1) By "white," I mean "Non-Hispanic white." I'm going to say "white" from now on exclusive of Hispanics.

2) Some states, such as Vermont, do not have a large enough sample to estimate the obesity proportion of blacks. I just used a neighboring state to fill in the numbers. This guesstimate is really not much of an issue because the proportion of blacks is so low in the states I had to estimate that the estimate of obesity for whites and estimate of obesity for all races is the same in these counties anyhow.

3) Simple algebra. Total Obesity Percent In County = (Obesity Percent Whites) X (Percent Whites) + (Obesity Percent Blacks) X (Percent Blacks) + (Obesity Percent Latinos) X (Percent Latinos)

For the obesity percent of blacks and Latinos I only have state level data, so this is going to be a rough estimate. And it's going to result in the variation exhibiting state-to-state discontinuities, since the county variable is dependent on a state level variable. Also, I discarded some counties where the usage of state level data caused really big distortions. Along the Mexican border Latinos are not nearly as obese as they are further into the United States, so I end up with numbers where whites have negative obesity percentages to make the math work out. These are counties which are 90% or more Latino with relatively low obesity numbers.

I did the map shading the way I normally do. Blue is above the median value, and red below the median value, with the scale being set to their max and mins respectively. Unfortunately this causes a problem in the scaling in terms of an asymmetry because one side of the distribution will tend to have a more extreme outlier (usually the above median is where the skew is).

Here's the map with all the populations:

This is basically the earlier map except shaded differently. Here are the summary statistics for obesity by county:

min = 12.40
1st quartile = 26.60
median = 28.40
mean = 28.25
3rd quartile = 30.20
max = 43.70

Now for my estimate of whites only:

As you can see, the use of state level is causing some distortions. Also, you see something peculiar in the summary statistics:

1st quartile = 25.54
median = 27.62
mean = 26.71
3rd quartile = 29.47
max = 58.11

These averages don't align with the CDC values aggregated. But that's because I'm looking at county level data, and not weighting by population. Lots of low density counties with few people have many obese people. Instead of looking at national averages, we're looking at regional variations.

On the estimates, Texas probably jumps out at you. To my surprise it turns out that whites in Texas are a touch lighter than the national average for whites! For me the big thing that sticks out is that Appalachia seems to be split in two, along the Appalachian Trail (I feel funny mentioning the Appalachian Trail....). Some areas, such as New England, Colorado and California do not surprise in terms of whites who are below the national median. But again there is a pattern of some pockets in the Upper Midwest being relatively under the norm in the proportion of obesity. Some of you might be surprised by the Pacific Northwest, but this region is characterized by urban-rural polarization.

What are the correlations by ethnicity? Here are the correlations with white obesity in terms of ancestral proportion (the proportion of ethnicity X as a proportion of whites):

English = -0.17
German = -0.02
American = 0.07
Scots Irish = -0.13
Irish = -0.19

These are very modest correlations. Probably mostly explained by geography. How about voting?

Obama vote = -0.21

Again, modest. Median Family Income? Only -0.14! That surprised me. Interestingly, Median Home Value had a -0.26 correlation with obesity. Of course the "Dirt Gap" tracks this; in places where people are thinner property values are higher, and rose higher in the past decade. The proportion who have a college degree is like home value, a correlation of -0.25.

None of this is really surprising, on the aggregate level you know that wealthier and more educated people are thinner. So I might as well do something that's not totally predictable. Most of the variance of obesity on the county level isn't predicated by educational levels, but a non-trivial fraction is. I decided to fit a loess curve to the plot of obesity (white) who are college educated. Then I simply took the residuals above and below the line and shaded them blue and red respectively. In other words, blue areas have a lot of fat people for the number of college graduates, while red areas have relatively few fat people for the number of college graduates.

Labels: , ,

Thursday, November 26, 2009

Liberty or Libel?   posted by DavidB @ 11/26/2009 04:02:00 AM

There has been much discussion in the blogosphere (for example by Olivia Judson here) of the current libel case between the science writer Simon Singh and the British Chiropractic Association. Most of the comments have supported Singh and criticised both the BCA and the trial Judge, Sir David Eady. Science writers complain that the libel laws are stifling fair criticism of unscientific claims (which makes this at least marginally relevant to gnxp).

I have no interest, of any kind, in chiropractic, and I support freedom of speech, so you might expect me to join the chorus of Singh-lovers and Eady-haters. Unfortunately, much of the commentary has been ill-informed or self-interested (since journalists and bloggers view the libel laws much as turkeys view Christmas). The British press has other motives for attacking Judge Eady, who has extended the legal right of privacy against paparazzi and tabloid journalists. So protestations of concern for 'free speech' need to be taken with a hefty pinch of salt...

Some red herrings to dispose of. First, there is a legitimate debate over the practice of 'libel tourism' or 'forum shopping'. But this issue does not arise in the Singh case, where a British writer made comments about a British organisation in a British newspaper. There is no question that a British Court is entitled to try the case.

Second, on libertarian grounds I would be willing to argue for complete freedom of speech, with no restrictions on libel. But that is not where we start from. Every country has some kind of libel law. The details vary, and the balance between freedom of speech and protection of individual reputation is struck in different ways. It is arguable that American law leans too far in favour of the libeller, while English law leans too far in favour of the libelled. But Eady's critics argue that even within the general framework of English libel law his rulings are dangerous to freedom of speech. I will therefore take that general framework as given.

What then are the issues?

Here is the key passage from Singh's article, which prompted the libel action:

The British Chiropractic Association claims that their members can help treat children with colic, sleeping and feeding problems, frequent ear infections, asthma and prolonged crying, even though there is not a jot of evidence. This organization is the respectable face of the chiropractic profession and yet it happily promotes bogus treatments.

Before going any further it is necessary to set out the various stages of a libel action under English law, since some of the critical commentary seems to misunderstand this. A case can be divided into four main stages:

Stage 1: It must be established what was said or written, who said it, and who was its 'target'. In the present case this is straightforward.

Stage 2: It is necessary to decide whether what was said is defamatory. Roughly, this means whether or not it is damaging to the reputation of the complainant. At this stage, under English libel law, the truth or falsity of what was said is irrelevant. [Note 1] Much of the comment on the case has failed to grasp this. A true statement may be defamatory, and a false statement may be non-defamatory. The point at issue is not its truth, but whether it is damaging.

Stage 3: If it is decided that a statement is defamatory, the person responsible for the statement may then defend it. Except in certain special circumstances, the defence is either that the statement is true (the defence of 'justification'), or that it constitutes 'fair comment'. In the English system it is usually for a jury to decide whether the defence is convincing.

Stage 4: If the jury finds in favour of the complainant, a decision is then needed on the amount of damages or other remedial action. Damages are decided by the jury. All costs of the case are usually paid by the losing side. It has been suggested in some commentaries that it is cheap to bring a libel action, because the complainant can hire a lawyer on a no-win no-fee basis. But this is only true if the complainant has a strong case; otherwise no lawyer will touch it.

The basic complaint of the BCA is that Singh's article accuses them of dishonesty, by promoting treatments which they know to be 'bogus'.

Judge Eady was asked to give preliminary rulings on two issues: what Singh's words meant; and whether they amounted to an assertion of fact or merely an expression of opinion. On the first point, he decided, agreeing with the BCA, that Singh's article accuses them of dishonesty, saying: '[the quoted passage] is in my judgment the plainest allegation of dishonesty and indeed it accuses them of thoroughly disreputable conduct.' After this, it was straightforward to take the further step of deciding that the passage is defamatory, since an accusation of dishonesty could hardly not be. On the second point, Judge Eady concluded that the passage amounts to an assertion of fact. The importance of this is that if the defamatory passage is an assertion of fact, the defence of 'fair comment' is not available, and the only defence (usually) is to show that the assertion is factually true, or 'justified'. This defence remains open to Singh.

The case so far therefore raises two issues:

1. Was Eady right to conclude that Singh had accused the BCA of dishonesty?

2. Was Eady right to conclude that the accusation was an assertion of fact, rather than merely an expression of opinion?

On the first point, the matter is perhaps not as clear-cut as Eady's ruling suggests, but on a common-sense reading of Singh's passage it is at least a very reasonable interpretation. Singh's words are strong: he says there is 'not a jot of evidence' for the BCA's claims, and that while it is the 'respectable face' of chiropractic, it still 'happily promotes bogus treatments'. Whether or not Singh intended this to be an accusation of dishonesty, it is a natural inference for the reader to draw. The word 'bogus' by itself usually has an implication of dishonesty; the dictionary gives synonyms such as 'sham', 'spurious', and 'counterfeit'. To say that someone promotes bogus treatments therefore might in itself be taken as implying dishonesty. This interpretation is reinforced by the contrast Singh draws between the 'respectable face' of the BCA and its 'happily' promoting 'bogus treatments'. The contrast between 'respectable face' and 'bogus' seems to imply that the BCA is not, after all, as respectable as it may appear. If Singh did not intend an imputation of dishonesty, he expressed himself carelessly. An alternative possibility is that he did intend to impute dishonesty, but chose his words so as to insinuate that conclusion without making it explicit. In any case, under English libel law, Singh's intention is irrelevant: what matters is the interpretation that reasonable readers are likely to put on his words.

On the second point, namely whether the defamatory claim was a matter of fact or opinion, the issues are more technical, and I do not pretend to understand all the legal subtleties. According to Eady's ruling:

It will have become apparent by now that I also classify the defendant's remarks as factual assertions rather than the mere expression of opinion. Miss Rogers reminded me, by reference to Hamilton v Clifford [2004] EWHC 1542 (QB), that one is not permitted to seek shelter behind a defence of fair comment when the defamatory sting is one of verifiable fact. [Note 2] Here the allegations are plainly verifiable and that is the subject of the defence of justification. What matters is whether those responsible for the claims put out by the BCA were well aware at the time that there was simply no evidence to support them. That is an issue capable of resolution in the light of the evidence called. In other words, it is a matter of verifiable fact. That is despite the fact that the words complained of appear under a general heading "comment and debate". It is a question of substance rather than labelling.

Given the assumption that there was an accusation of dishonesty, this seems a reasonable enough decision. The defence of 'fair comment' is more narrowly circumscribed than the layman might imagine. The test of whether something is 'opinion' depends on the substance of the alleged disreputable conduct, and not on the form in which the allegation is made. It does not become a matter of opinion just because the author uses the words 'in my opinion' or some other verbal dodge.

Clearly the whole case (so far) hinges on the question whether a reasonable reader would interpret Singh's words as containing an accusation of dishonesty. Much of the commentary has either missed this point, or strained to find alternative interpretations. For example, the words are interpreted as imputing mere gullibility or ignorance, rather than dishonesty. In some circumstances that might be the most natural interpretation of the same or similar words. For example, it might be said that exorcism is a 'bogus' treatment for mental illness, yet that some religious sects 'happily promote' this bogus treatment. In this case it might plausibly be argued that the implied accusation is one of gullibility or ignorance rather than dishonesty. But this interpretation relies on the background knowledge than religious sects are commonly ill-informed and gullible. In the case of the BCA, the contrast that Singh himself makes is between the BCA's position as the 'respectable face' of a medical profession, and its willingness 'happily' to promote 'bogus' treatments for which there is 'not a jot of evidence'. It is difficult to regard this merely as an accusation of gullibility. According to Judge Eady's ruling:

It is alleged that the claimant promotes the bogus treatments "happily". What that means is not that they do it naively or innocently believing in their efficacy, but rather that they are quite content and, so to speak, with their eyes open to present what are known to be bogus treatments as useful and effective. That is in my judgment the plainest allegation of dishonesty and indeed it accuses them of thoroughly disreputable conduct.

The critics complain that this is reading too much into the word 'happily', which could have a variety of other meanings. But again the question is not what the word might conceivably mean, but what a reasonable reader is likely to take it to mean. The meaning of words often depends on their context. In this case, the word 'happily' does not have its literal meaning as a description of an emotional state. The word must in some way describe the collective state of mind of the BCA in promoting 'bogus' treatments, and in the context it does (it seems to me) have a strong suggestion of dishonesty. The alternative is to suppose that it has a weaker connotation of recklessness or irresponsibility, but not of conscious dishonesty, or that it leaves several possibilities open, meaning (roughly) 'dishonest or gullible or reckless or irresponsible...'. These interpretations are not impossible, but Singh himself has made it more difficult to accept them by saying that there is 'not a jot of evidence' for the 'bogus' treatments. If this were true, then the BCA, as a body of specialists in the field, could hardly be unaware of it, and their promotion of such treatments would go beyond mere recklessness into conscious dishonesty. Judge Eady's interpretation is therefore not unreasonable.

Nor does the case have the far-reaching implications for freedom of speech or scientific research that some critics claim. No-one is suggesting that it is improper to criticise chiropractic or other alternative therapies. The only lesson to be drawn is that if you wish to accuse someone of dishonesty, at least in England, you must be ready to back up your accusation with evidence; and if you do not wish to accuse someone of dishonesty, you should choose your words with care.

Note 1: This is the position in most of the Common Law world. It was also the position in the United States until a series of Supreme Court decisions shifted the burden of proof onto complainants, where they are 'public figures', to show that the words complained of are not only defamatory but deliberately false. A useful account of American libel law is here.

Note 2: Out of curiosity I looked up this case. British readers may recall the incident when the former MP Neil Hamilton and his wife were accused of having raped a woman. The accusation was investigated by the police and disproved. The accuser was subsequently prosecuted and jailed for making false accusations. But before this, she had sold her story to the tabloids, using the PR consultant Max Clifford as intermediary. During the police investigation Max Clifford had gone on television to defend the woman's claims, and among other things said he personally believed the claims were true. This was what led to the libel action, as the Hamiltons claimed that by endorsing the woman's accusations Clifford was himself in effect accusing the Hamiltons of rape. Clifford argued in his defence that he was merely expressing an opinion, but the Judge ruled that he was making an assertion of fact, and could not shield behind the defence of 'fair comment'. And who was the Judge? - step forward, Mr Justice Eady!

Added on 27 November: it has been pointed out that Simon Singh has recently been granted leave to appeal on some of the issues raised by the case. The Appeal Court may well reverse Judge Eady's rulings on some or all matters. In my post I did not suggest that Eady was necessarily right, just that his rulings were a lot more reasonable than some commentators have claimed. As I said at the outset I have no interest in chiropractic. I have only commented on the case because I was getting tired of misrepresentations of it, which recur in an article in the (London) Times yesterday. Two things in particular have irritated me. One is the one-sided presentation of the case by the commentators. I have not seen a single comment which recognises that the BCA might just have a legitimate complaint when they are, arguably, accused of dishonesty. You can argue about the precise meaning of the words used by Singh, but no-one can sensibly deny that they could be used to make an accusation of dishonesty. Second, I am concerned that scare-mongering about the effects of the case on free speech and scientific enquiry could be a self-fulfilling prophecy. If scientists and science writers (including bloggers) are led to believe that they cannot make strong criticisms of pseudo-science without facing a libel action, freedom of speech and enquiry really will be inhibited. For the reasons given in my post, I do not think that the Singh case has these implications, and those who claim that it does are harming the cause they wish to defend.

I am also happy to acknowledge that I obtained the text of Judge Eady's ruling through JackOfKent's blog, via Olivia Judson's blog, which is linked in my post. I would also stress that my criticism of 'ill-informed' commentators does not include JackOfKent. I don't agree with his assessment of the case, but he is certainly well-informed about it - far more so than me.

Added on 29 November: I hold no brief, in any sense, for the BCA, but it seems to me that in fairness one should not accuse them of 'litigiousness', without at least checking their own statements of position. Here is one of their press notices on the Singh case. I do not know (obviously) whether the quote they attribute to Simon Singh at the end of their statement is true, but if it is, it puts Singh in a very different light from that presented by his cheerleaders.

Wednesday, November 25, 2009

No support for birth order effects on personality from the GSS   posted by ben g @ 11/25/2009 08:47:00 PM

In researching for a review of The Nurture Assumption, I read over the debate between Harris and Sulloway over birth order effects on personality. Sulloway's thesis, explained in Born to Rebel, is that last-born children have more rebellious, agreeable, and open-minded/liberal personalities, and that this manifests itself in history with revolutions spearheaded by last-borns. This runs in contrast to Harris's theory that the family environment has no lasting impact on personality, so she spends a good deal of time in her books and articles critiquing it.

The whole debate makes my head dizzy. A seemingly simple empirical question has produced years of arguing over methodology. I'm not going to go over the tedious back and forth here, except to say that you can see what both sides have to say with a Google search.

Large, controlled studies have not been kind to Sulloway's thesis. Freese, Powell, and Steelman (1999) looked for a relationship between birth order (controlled for family size) and a variety of political measures on the nationally representative General Social Survey (GSS). They found no significant associations, contrary to Sulloway's predictions.

I decided to look at the GSS myself, this time to see whether questions that tapped into personality characteristics outside of politics showed any relationship with birth order (SIBORDER), when sibship size (SIBS) was controlled for. I excluded only children. I used the Multiple Regressions feature on the Berkeley SDA tool. I found no significant associations between birth order and any of the four variables I looked at:

  • MEMLIT (proxy for openness/creativity)- "Here is a list of various organizations. Could you tell me whether or not you are a member of each type? m. Literary, art, discussion or study groups"

  • TRUST (proxy for agreeableness) - "Generally speaking, would you say that most people can be trusted or that you can't be too careful in life?"

  • WORLD4 (proxy for agreeableness) - "People have different images of the world and human nature. We'd like to know the kinds of images you have. Here is a card with sets of contrasting images. On a scale of 1-7 where would you place your image of the world and human nature between the two contrasting images? 1. Human nature is basically good. 7. Human nature is fundamentally perverse and corrupt."

  • OBEYLAW (proxy for rebelliousness) - "In general, would you say that people should obey the law without exception, or are there exceptional occasions on which people should follow their consciences even if it means breaking the law?"
I wouldn't say that we should write off the idea of birth order influences on personality and intelligence, only that we should be very skeptical of them. To the extent that they do exist, they're probably not very significant.


GWAS, population structure and the Han Chinese   posted by Razib @ 11/25/2009 01:47:00 PM

Two new articles in AJHG, Genomic Dissection of Population Substructure of Han Chinese and Its Implication in Association Studies:
To date, most genome-wide association studies (GWAS) and studies of fine-scale population structure have been conducted primarily on Europeans. Han Chinese, the largest ethnic group in the world, composing 20% of the entire global human population, is largely underrepresented in such studies. A well-recognized challenge is the fact that population structure can cause spurious associations in GWAS. In this study, we examined population substructures in a diverse set of over 1700 Han Chinese samples collected from 26 regions across China, each genotyped at ∼160K single-nucleotide polymorphisms (SNPs). Our results showed that the Han Chinese population is intricately substructured, with the main observed clusters corresponding roughly to northern Han, central Han, and southern Han. However, simulated case-control studies showed that genetic differentiation among these clusters, although very small (FST = 0.0002 ∼0.0009), is sufficient to lead to an inflated rate of false-positive results even when the sample size is moderate. The top two SNPs with the greatest frequency differences between the northern Han and southern Han clusters (FST > 0.06) were found in the FADS2 gene, which associates with the fatty acid composition in phospholipids, and in the HLA complex P5 gene (HCP5), which associates with HIV infection, psoriasis, and psoriatic arthritis. Ingenuity Pathway Analysis (IPA) showed that most differentiated genes among clusters are involved in cardiac arteriopathy (p < 10−101). These signals indicating significant differences among Han Chinese subpopulations should be carefully explained in case they are also detected in association studies, especially when sample sources are diverse.

And, Genetic Structure of the Han Chinese Population Revealed by Genome-wide SNP Variation:
Population stratification is a potential problem for genome-wide association studies (GWAS), confounding results and causing spurious associations. Hence, understanding how allele frequencies vary across geographic regions or among subpopulations is an important prelude to analyzing GWAS data. Using over 350,000 genome-wide autosomal SNPs in over 6000 Han Chinese samples from ten provinces of China, our study revealed a one-dimensional “north-south” population structure and a close correlation between geography and the genetic structure of the Han Chinese. The north-south population structure is consistent with the historical migration pattern of the Han Chinese population. Metropolitan cities in China were, however, more diffused “outliers,” probably because of the impact of modern migration of peoples. At a very local scale within the Guangdong province, we observed evidence of population structure among dialect groups, probably on account of endogamy within these dialects. Via simulation, we show that empirical levels of population structure observed across modern China can cause spurious associations in GWAS if not properly handled. In the Han Chinese, geographic matching is a good proxy for genetic matching, particularly in validation and candidate-gene studies in which population stratification cannot be directly accessed and accounted for because of the lack of genome-wide data, with the exception of the metropolitan cities, where geographical location is no longer a good indicator of ancestral origin. Our findings are important for designing GWAS in the Chinese population, an activity that is expected to intensify greatly in the near future.

Labels: , , ,

Maps of diabetes & obesity   posted by Razib @ 11/25/2009 12:35:00 PM

Hope readers have a happy Thanksgiving. I assume this is also a day when you're not going to think too much about your diet and eat what you want to eat. But I thought this map on diabetes and obesity for those age 20 and up was interesting. These are estimates, which I think explains the rather sharp boundaries at state lines (since state level data was probably used to predict county values, see the methods here). To my knowledge the cuisine of the Upper Midwest and New England gets about as much props as that of England (vs. "Southern home cooking"), but hotdish can't be all that unhealthy? :-) H/T Ezra Klein.


Discussion of CRU Materials   posted by dkane @ 11/25/2009 06:53:00 AM

The debate over global warming is relevant to GNXP (previous posts here and here) not so much because our readers are interested in climate science but because the dynamics of the debate --- scientific "consensus" versus politically incorrect minority view --- have relevance to various debates over human biodiversity. With that background, I would be curious to know what posts/arguments/sites other GNXPers find most relevant and compelling. My three favorites are:

1) Willis Eschenbach on his attempts to use the Freedom Of Information Act to access CRU data and model details.

2) Eric Raymond on details of the released code. When serious software experts like Raymond start accusing climate scientists of "blatant data-cooking," the "consensus" started to looks much weaker.

3) Gavin Schmidt at Real Climate in defense of the scientists at CRU. Whatever the criticisms of Schmidt from folks like Steve McIntyre, you have to be impressed with his willingness to take on all comers.

In the spirit of (United States) Thanksgiving, I am thankful that the human genetics community seems to have a much broader diversity of opinion and greater committment to transparency and reproducible research than the climate science community.

The Biggest Loser and Indian obesity   posted by Razib @ 11/25/2009 12:27:00 AM

After reading this article about the The Biggest Loser, I checked out the Wikipedia page for the show. There are international versions. Through that I found out that there is an Indian version of the show. I thought this was weird. I mean, it's India, right? Well, around 2% of Indian women are obese (BMI 30 and above). That's 20 million obese! In India obesity still seems to correlate with wealth and higher status. Nearly 10% of women in Delhi and Punjab are obese (vs. 0.5% in Bihar, the heart of the "Deep North"). Nearly 10% of Jains are obese. 6.5% of women who have completed secondary school are obese, vs. 1% of illiterates. 6% of people in urban areas vs. 1% in rural areas.


Tuesday, November 24, 2009

R1a1 and the peopling of Eurasia   posted by Razib @ 11/24/2009 07:58:00 PM

A few weeks ago people in the comments were nagging me a bit about some new papers on the haplogroup R1a1. This Y chromosomal lineage is found at very high frequencies from East-Central Europe into India. Initially, researchers such as Spencer Wells assumed that R1a1 signaled the arrival of Indo-Aryans to the Indian subcontinent, its frequencies decline in a northwest-to-southeast gradient, and from high to low castes. In Europe the modal frequencies are among Slavic groups, with a high representation among Germanic-speakers. The frequency of R1a1 declines sharply in Western and Southern Europe. It is very common in Central Asia as well as eastern Iran and Afghanistan. One parsimonious explanation would be that R1a1 spread with Kurgan males, along with Indo-European languages, on the order of 4-5,000 years ago.

There is a problem with this model though. One of the new papers reiterates the finding that the coalescence of the European and South Asian lineages is on the order of 10,000 years ago: Separating the post-Glacial coancestry of European and Asian Y chromosomes within haplogroup R1a (R1a1 is the dominant clade within R1a). A second paper reports the finding that R1a1 is very diverse in India, indicating deep time depth: The Indian origin of paternal haplogroup R1a1* substantiates the autochthonous origin of Brahmins and the caste system. For both R1a1 &"Ancestral North Indians" (ANI) in Reich et al.: the frequency seems intuitively way too high among tribal populations, even in South India. Remember that the low bound for ANI was ~40%. R1a1 is found at frequencies as high as 25% or so among some South Indian tribals. If this lineage arrived with the Indo-Aryans it is peculiar that it is found in such high frequencies in populations which were marginal and isolated from the dominant non-Indo-Aryan populations of South India. Back to Europe, here is a section from the abstract of the first paper:
Conversely, marker M458 has a significant frequency in Europe, exceeding 30% in its core area in Eastern Europe and comprising up to 70% of all M17 chromosomes present there. The diversity and frequency profiles of M458 suggest its origin during the early Holocene and a subsequent expansion likely related to a number of prehistoric cultural developments in the region. Its primary frequency and diversity distribution correlates well with some of the major Central and East European river basins where settled farming was established before its spread further eastward. Importantly, the virtual absence of M458 chromosomes outside Europe speaks against substantial patrilineal gene flow from East Europe to Asia, including to India, at least since the mid-Holocene.

The Holocene started 11,700 years ago. We are living in the Holocene. So the means that gene flow can't be any later than 6,000 years ago. The paper which focuses specifically on Indian lineages reports a coalescence time on the order of 10,000 years in the past for South Asian R1a1 branches. Additionally, they confirm earlier findings that of caste ranking of R1a1 in terms of frequency, as well Brahmins having the most diversity of all groups in terms of haplotypes (ergo, the title of the paper).

Both Dienekes and Polish Genetics and Anthropology suggest that the calibration is wrong on these coalescence times. They argue that one should reduce the time to a common ancestor by a factor of 3. This would of course make a huge difference. In regards to the Reich et al. paper which argued for a plausible two-way admixture between ANI and "Ancestral South Indians" (ASI), the linkage disequilibrium has decayed too much from the time of admixture to peg a date. This was a method used to calculate the emergence of the Uyghurs as a hybrid population, on the order of 2-3 thousand years ago (admixture between two very different populations generates linkage disequilibrium which decays over time due to recombination). In terms of Fst the ANI have a value in relation to Northern Europeans which is about 3 times larger than the mean between population differences in Europe. This is somewhat greater than the pairwise values between any European populations except for the Baltic peoples (in particularly, the swath from Karelia to Lithuania) to the groups of Southern Europe. The degree of Neolithic Middle Eastern ancestry within Europe under debate, but I think one can assume that Southern Italians and Karelians are likely at opposite ends in terms of frequency of this contribution to the pre-Ice Age demographic substratum of Europe. From this I offer that it is not totally unreasonable to posit that the ANI contribution to South Asian ancestry was closer to the margins of the last Ice Age, rather than the period of the Indo-European expansion, and that its Fst values are not unreasonable in relation to modern European groups.

The main issue that is confusing is the diversity of R1a1 in South Asia. A first order model going from just this data would be that R1a1 derives from India, and spread to the Eurasian plain. But Reich et al. show data that imply little likelihood of South Asian contribution to European ancestry. The only possibility would be if ANI and ASI were totally separated when a branch of ANI left South Asia for the Eurasian plain, and which point the process of admixture between ANI and ASI began. Another possibility is that the distribution of R1a1 in Eurasia is a palimpsest. Recent work in ancient DNA is suggesting that inferring past distributions from contemporary ones may lead us astray. It could be that R1a1 was once far more diverse in Europe and Central Asia, but that subsequent demographic events eliminated most of that diversity, while such events did not occur in Europe. Y chromosomal lineages may be particularly likely to be wiped out by the expansion of new tribes as old elites are killed or marginalized. The current distribution of a particular branch of R1a1 in Europe, associated in particular with Slavs, may be an expansion of the lineage which managed to survive elimination at some point in the mid-Holocene.

Though do note I put little weight in my speculations. It seems rather confusing. But since I was asked....

Labels: ,

Why whales get no bigger   posted by Razib @ 11/24/2009 05:33:00 PM

Carl Zimmer reports that it might be a function of physics. Bigger whales have proportionality bigger mouths, but at some point the biological engineering runs up against constraints:
s they report today in the Proceedings of the Royal Society, Goldbogen and his colleagues found that big fin whales are not just scaled-up versions of little fin whales. Instead, as their bodies get bigger, their mouths get much bigger. Small fin whales can swallow up about 90% of their own body weight. Very big ones can gulp 160%. In other words, big fin whales need more and more energy to handle the bigger slugs of water they gulp. As their body increases in size, the energy their bodies demand rises faster than the extra energy they can get from their food.


If the scientists are right, they may have discovered one of the big ironies in evolution. Lunge-feeding may have allowed whales to become the biggest animals ever to roam the planet. But this was not an open-ended invitation. Once whales got large enough, lunge feeding itself became so costly it prevented them from getting any bigger. Perhaps some day another animal will evolve a new strategy that will let it get even bigger than a blue whale. But for the animal kingdom as we know it, we may be sharing the planet with the biggest species it can offer.

Given enough time and a large population one can imagine that evolution might be able to figure out a solution, or back out of the adaptive dead end.

Labels: , ,

Monday, November 23, 2009

Prediction markets   posted by Razib @ 11/23/2009 05:01:00 PM

It's News On Academia, Not Climate:
Yup, this behavior has long been typical when academics form competing groups, whether the public hears about such groups or not. If you knew how academia worked, this news would not surprise you nor change your opinions on global warming. I've never done this stuff, and I'd like to think I wouldn't, but that is cheap talk since I haven't had the opportunity. This works as a "scandal" only because of academia's overly idealistic public image.

It is a shame that academia works this way, and an academia where this stuff didn't happen would probably be more accurate. But even our flawed academic consensus is usually more accurate than its contrarians, and it is hard to find reliable cheap indicators saying when contrarians are more likely to be right.

If you don't like this state of affairs join me in trying to develop a more reliable consensus mechanism on such topics: prediction markets. It just takes time or money. Prefer instead to act shocked, just shocked, when the other side is shown to do this stuff, while reserving your side’s ability to do the same? Then I have little respect for you.


1 million SNPs to bind us all   posted by Razib @ 11/23/2009 01:41:00 PM

A a new paper in PLoS ONE, Genetic Variation and Recent Positive Selection in Worldwide Human Populations: Evidence from Nearly 1 Million SNPs:
Our analyses both confirm and extend previous studies; in particular, we highlight the impact of various dispersals, and the role of substructure in Africa, on human genetic diversity. We also identified several novel candidate regions for recent positive selection, and a gene ontology (GO) analysis identified several GO groups that were significantly enriched for such candidate genes, including immunity and defense related genes, sensory perception genes, membrane proteins, signal receptors, lipid binding/metabolism genes, and genes involved in the nervous system. Among the novel candidate genes identified are two genes involved in the thyroid hormone pathway that show signals of selection in African Pygmies that may be related to their short stature.

They seem to have looked at about twice as many SNPs by combining the sets of Illumina and Affymetrix chips as the norm. But they looked at only around 1/4 the number of individuals as other studies which used the HGDP panel. To a first approximation the Affy and Illumina chips are really close in the patterns of variation which they detect, but, the Illumina chip had a significantly higher heterozygosity (this is evident in some of the supplementals just by inspection).

I reformatted a figure which shows ancestral contributions to the individuals in their sample at K = 6 (6 hypothetical populations which contribute to genetic variation). In the paper they discuss the fact that the Uyghur and Hazara resemble each other, and that the Uyghur seem to have a non-trivial Central/South Asian component, and finally that the Russian and Adygei have East Asian and Central/South Asian ancestry. None of this is surprising, all this was evident in other papers which used the same sample.

First, in regards to Russians, analysis of genetic variation among East European populations sometimes show a "long tail" of variation which leads toward East Asia among Russians. That is, Russians tend to cluster with other Europeans, but a minority of individuals are deviated in the direction of East Asians, that minority shrinking in proportion to distance from Europeans. The historical reason for this presents itself plainly: a significant minority of ethnic Russians have Tatar antecedents in the recent past, and of those who do not such ancestry may be derived from Slavicized Finno-Ugric populations who may have ancient connections to the populations of Siberia. The Russian Orthodox priest who was murdered last week known for preaching to Muslims was himself an ethnic Tatar by origin.

Second, one should expect the Uyghur and Hazara to resemble each other. The Hazara likely emerged during the period of Mongol rule of Iran and Afghanistan, and are descendants in part of Mongols and Turks from greater Mongolia who settled down in Afghanistan. The Uyghurs are a Turkic-speaking people, but historically the Tarim Basin was inhabited by Europoid populations. The emergence of the Uyghur and Hazara mimic each other almost perfectly. In particular, the East Asian component of their ancestry is from the same region. The non-East Asian aspect differs a bit, but not too much when set next to the East Asian component. Interestingly, the Uyghur speak a Turkic language, while the Hazara speaking Dari, the Persian dialect. One can probably chalk that up to distance from the Turco-Mongol ur-heimat.

Third, the Central/South Asian component among the Uyghur should not be too surprising, there is significant evidence that the Tarim basin was influenced by Indo-Iranians, as well as the Tocharians. Buddhism arrived in East Asia via the Tarim Basin after all, and there have always been trade routes from the southern edge of the Tarim down into northern India. But what about the Russians and the Adygei? I think that this signal has something to do with what we've termed elsewhere as "Ancestral North Indians" (ANI), who were closely related to European populations, and probably emerged from somewhere in Eastern Europe to Central Asia. I've been told that the Fst number for ANI-Northern European populations is on the order of the distance between Baltic peoples and southern Italians. So this group may have emerged on the margins of Europe, and expanded mostly within Asia.

There's also an interesting chart showing patterns of selection, or at least what they detected, across geographies. Even if most of the signals are false positives one may hold that the real signals within this subset will still recapitulate the geographic relationships shown to the left. The patterns of selection mirror overall phylogenetic relationships. Note the overlap patters of Central/South Asians with Europeans and East Asians, some of both, but dominated by the former.

Citation: Lopez Herraez D, Bauchet M, Tang K, Theunert C, Pugach I, et al. 2009 Genetic Variation and Recent Positive Selection in Worldwide Human Populations: Evidence from Nearly 1 Million SNPs. PLoS ONE 4(11): e7888. doi:10.1371/journal.pone.0007888

Labels: ,

Lying with the GSS, easy, but not necessary   posted by Razib @ 11/23/2009 12:37:00 AM

About year ago I thought perhaps more bloggers would start taking up the GSS. That really hasn't happened. Sometimes I wonder why, and the other day I had one idea: it isn't necessary for what most bloggers want to do, confirm what they already believe. Google is sufficient. The GSS can be massaged pretty easily. I try and be careful about how I limit the variables so that I don't get the results I "want," bad sign if you keep fiddling & fiddling. That sort of thing made me realize it wouldn't be that hard to squeeze some results you wanted out of the data (though there's no way you're going to get some correlations and trends of course). But why bother? With a sufficiently precise Google query you can get all the data & results you want from the top of the stack which supports your own thesis (it is often not too difficult to reproduce the queries of commenters who engage in this sort of thing to prove their point, just put in a few variants of queries and the link that they provided shows up on the first page of results).


Sunday, November 22, 2009

The mosaic of North American populations   posted by Razib @ 11/22/2009 08:51:00 PM

A few months ago an interesting paper connected the historical demographics of New Hampshire with genetic variation. One of the notable features of North American history and culture is that it is a mosaic of different populations, and, that mosaic has come about in very different ways. For example, the millions of Italian and Jewish Americans descend from hundreds of thousands of Italian and Jewish immigrants. By contrast, millions of Yankees and Quebecois descend from tens of thousands of ancestors, who arrived in the 17th and 18th centuries. In 1992 a Census demographer, Gibson Campbell, calculated that 49% of the population of the United States in 1990 was descended from those whose ancestors were resident withi nthe United States in in 1790 in "The Contribution of Immigration to the Growth and Ethnic Diversity of the American Population" (inclusive of blacks and whites). 51% were descended from those who arrived after 1790. Put it another way, 127 million Americans in 1990 were attributable to the net 50 million immigrants who arrived after 1790. The remainder of the population would be attributable to the 4 million U.S. residents in 1790.

Note: Looking at the immigration records more than 1 million Italians and Jews remained in the United States (around 4 million Italians arrived between 1820 and 1920, but the majority seem to have gone back to Italy). But reproductive variance being what it is, I think it is plausible to assuming that fewer than 1 million may contribute most to the current generations of these two groups.

Labels: ,

A tale of two nations   posted by Razib @ 11/22/2009 07:18:00 PM

One of the mantras of the new age is that European nations have to deal with diversity, something that's new to them. This actually ahistorical. Some military units in the Austro-Hungarian Empire actually used English as the lingua franca because of their ethnic diversity (due to those who returned from the United States, see 1848: Year of Revolution), while the French language and identity as dominant within the political unit of France is an artifact of the 19th century (see The Discovery of France: A Historical Geography). But one point which was brought home to me was the diversity of the United Kingdom in 1800. I knew of course that Ireland before the famine was 1/3 of the population of the United Kingdom, but I hadn't thought through the political ramifications. In The Cousins' Wars: Religion, Politics, Civil Warfare, And The Triumph Of Anglo-America Kevin Phillips highlights an important set of facts about the United Kingdom:

1) The constant migration of dissenting Protestants in the 17th to 18th century to America resulted in a firming up of the position of the Anglican mainstream within England.

2) The massive migration (as well as deaths due to famine) of Irish Catholics in the 19th century solidified the Protestant character of the United Kingdom.

Below the fold is a comparison of England and Ireland in terms of population between 1800 and 1910.

Here is the biography of Patrick Kennedy, great-grandfather of John F. Kennedy:
By the time Patrick reached adulthood, both his parents were apparently dead and the family homestead was controlled by his older brother John Kennedy, more than a dozen years Patrick's senior, who was already married and the father of four children. The eldest son normally inherited whatever claims existed to the family's farm. Because of the life-threatening scarcity of food and resources, the rest of the children, such as third son Patrick Kennedy, usually were expected to leave for the New World. Patrick also had a brother James and a sister Mary.

Patrick's life as a farmer in Dunganstown consisted mainly of cutting and tying bundles of grain by hand, and planting and tilling potatoes for his family's consumption. This routine varied only when he ventured into the nearest town, New Ross, with supplies of barley, and when the family attended mass about a mile away.

At the age of 26, Kennedy decided to leave Ireland. It is assumed this was for reasons of starvation related to the Irish Potato Famine, illness, or because he knew that a third-born son had virtually no hope of running his family's farm. His good friend at Cherry Bros. Brewery in New Ross, Patrick Barron, who taught Kennedy the skills of coopering, had come to that conclusion months earlier and left for America. In October 1848, in love with Barron's cousin Bridget Murphy and with a plan to wed, Patrick Kennedy decided to follow.

Note: I am aware of the fact that some of the decrease in Ireland's population and England's rise is due to emigration from the former to the latter. Removing this would not change the qualitative relation much.


Data and social networks   posted by Razib @ 11/22/2009 12:19:00 AM

Does anyone know of a free source of county level presidential results going back to the 19th century? I want to compare correlations in voting across time. I did find some data from Pennsylvania, and noted that the Great Flip seems not to be evident in that state for the 1856 or 1860 election (that is, the correlation between Democrat and Republican voting patterns by county between 2008 and those years is around zero). Here's a map of the 1960 presidential election results by county, red for Nixon and blue for Kennedy:

The Yankee dominated regions of northern New England remained Republican strongholds in 1960, just as they were during the ascendancy of Franklin Roosevelt. In Albion's Seed David Hackett Fischer argues for a "First Settler Effect" which echoes down across the centuries. This sort of paradigm would ask us what substantive similarities underpin the common support of Vermonters for Hoover in 1932 and liberal Democrats and Republicans in the 2000s (remember that northern New England still has a much larger fraction of Yankees). But I wonder if what is really maintaining regional coherency across time are social networks which share ideas and evolve together over time. It is peculiar to imagine it now, but during the early republic the Yankees were the segment of the population most fixated on Christian orthodoxy (evident by the fact that New England states were last to disestablish their churches), and were the driving forces of the Second Great Awakening (which did spread across the country). Today Yankees are the most secular segment of settler descended subcultures. Conversely, in 1800 the South was relatively lax in matters of religion, republican in politics, and pro-French in sympathy. John C. Calhoun was a Unitarian.

So that's why I want to get county-by-county data sets.


Saturday, November 21, 2009

Models of IQ & wealth   posted by Razib @ 11/21/2009 05:48:00 PM

Steve Hsu has been interesting of late (interesting like Steve, not Malcolm). So, IQ, compression and simple models and If you're so smart, why aren't you rich?. For a theoretical physicist I find Steve to be eminently clear in his exposition of abstract topics (perhaps he has practice from having to talk to experimental physicists?).


Friday, November 20, 2009

Latin America is not panmixia   posted by Razib @ 11/20/2009 03:35:00 PM

A new provisional paper, Ancestry-related assortative mating in latino populations. Here are the results:
Using 104 ancestry informative markers, we examined spouse correlations in genetic ancestry for Mexican spouse pairs recruited from Mexico City and the San Francisco Bay Area, and Puerto Rican spouse pairs recruited from Puerto Rico and New York City. In the Mexican pairs, we found strong spouse correlations for European and Native American ancestry, but no correlation in African ancestry. In the Puerto Rican pairs, we found significant spouse correlations for African ancestry and European ancestry but not Native American ancestry. Correlations were not attributable to variation in socioeconomic status or geographic heterogeneity. Past evidence of spouse correlation was also seen in the strong evidence of linkage disequilibrium between unlinked markers, which was accounted for in regression analysis by ancestral allele frequency difference at the pair of markers (European versus Native American for Mexicans, European versus African for Puerto Ricans). We also observed an excess of homozygosity at individual markers within the spouses, but this provided weaker evidence, as expected, of spouse correlation. Ancestry variance is predicted to decline in each generation, but less so under assortative mating. We used the current observed variances of ancestry to infer even stronger patterns of spouse ancestry correlation in previous generations.

The correlations are to the left. An interesting point is that the correlations of total genome content seem too high to be explained by assortative mating for salient physical features (skin color, hair form, etc.) alone. From the text:
Another possibility involves physical characteristics, such as skin pigment, hair texture, eye color, and other physical features. Certainly, these traits are correlated with ancestry and are likely to be factors in mate selection. However, the spouse correlation for these traits must be high and the correlation of these traits with ancestry must also be high to explain the observed ancestry correlations....


If the spouse trait correlation is 0.6 (a reasonably high value), then for a spouse ancestry correlation of 0.3 (Puerto Ricans), the trait-ancestry correlation is 0.7; for a spouse ancestry correlation of 0.4 (Mexicans), the trait-ancestry correlation is 0.8. Previous studies on assortative mating in Latin American groups have retrieved correlation coefficients of 0.29 to 0.46 for education level, 0.48 for skin reflectance, 0.07 to 0.18 for eye and hair color, and 0.16 to 0.24 for different anthropometric measurements

As noted above, they controlled for SES and geography, and the correlation remains. Looking at the correlations within the genomes of these individuals they also inferred that assortative mating in the past was actually greater than it is today (they also have a historical citation which suggests this). I wonder of the correlation of ancestry is due to sorting by many traits which are subtle and nuanced, and relatively difficult to capture in surveys of the coarse salient traits are used to categorize phenotypic races. Looking at many traits, as opposed to a few, and one would have a better sense of total genome content. When it comes to mating one might look to a range of traits which in other circumstances are not noted, or fall below the threshold of reflective awareness. I'm assuming there might be something here which is Gestalt and subconscious. Kind of like the various studies which attempt to correlate mate preferences by HLA polymorphism.

Labels: ,

Thursday, November 19, 2009

Fake fact, America is not secularizing   posted by Razib @ 11/19/2009 06:20:00 PM

How Will Religion Evolve?, asks John Tierney. He notes:
If there is a religious instinct, how do we make sense of the declining church attendance in western Europe? As an agnostic myself, I've tended to see the European trend as a harbinger of a general move toward secularism as societies become richer and more educated. But you don't see that trend in the United States, where church attendance is still robust, and Nicholas told me that he sees a long future for religion: "The extent to which people practice religion in modern states may wax and wane, depending on social circumstances like war or privation, but religion is unlikely to disappear entirely."

The fake fact is that church attendance is not declining. It is. Compared to secular regions of Europe any region of the United States is very religious, but, the number of Americans who declare No Religion has doubled between 1990 and 2008, from 8% to 15%. Perhaps one might label this the "Silent Secularization," in contrast to the 1960s when the power of Mainline Protestantism as a cultural arbiter was broken in a very public manner through the Counterculture, along with the subsequent prominence of the Christian Right from the late 1970s on.

I think there are several reasons that the secularization of the of the 1990s, when more than 1 million per year joined the ranks of those with no religious affiliation, has been a silent phenomenon. First, the period from 1980 onward has been one where conservative politics has set the tone, and the Christian Right has been a major power broker. Despite the fact hat 600,000 people a year were joining the ranks of those with no religion in the 2000s, the president was a conservative Protestant, and Congress was dominated by figures who acknowledged the legitimacy of the Religious Right (though I tend to lean toward the proposition that economic conservatives still control the Republican party, or did, for most of this period).

Second, the secularization of the 1990s, in contrast to the 1960s, was relatively low key and banal. It wasn't flashy. It probably mostly involved nominal Christians who finally severed their vestigial ties to religion (the American Religious Identification Survey 2008 suggests that nominal white Catholics in particular defected in massive numbers). With the strong skew to the youth, these were probably Gen-Xers who decided to go hiking or jamming with their garage band on Sundays, but maintaining normal jobs and lives. By contrast, the growth of the megachurch made much better copy. In many ways conservative Protestants are the modern Counterculture, going against the dominant currents of the society (e.g., the "True Love Waits" movement). But interestingly, despite growing at the expense of Mainline Protestants, Evangelical Protestants are not making Protestantism more theologically conservative.

In any case, contrary to John Tierney's assumption, church attendance has been declining. GSS variables ATTEND and YEAR show the trend. It's all a function of the doubling of those with "No Religion" though. If you limit the sample to Protestants & Catholics, there's little change over the decades....


Hagarism, revision, and everything we think is wrong (?)   posted by Razib @ 11/19/2009 02:18:00 AM

This is more a question for readers who know this stuff, what do you think about Patricia Crone & company in their revision of the early history of Islam? I'm more of a Hugh Kennedy guy because I don't know much about this field and would prefer to stick to the mainstream, but a few years ago I read a short monograph on representational art in mid-Umayyad Syria, and it just didn't "feel right" in the context of the traditional narrative. The book didn't really talk much about history, but rather more the Late Antique cultural influences on the Umayyad's. But what I encountered seemed more like a conventional society of the post-Roman Near East than anything I would recognize as "Muslim." Of course it's all impressionistic, and I don't have a good feel of the lay of the land, so I dismissed it. But how about those of you who know the primary sources? I can't find Daniel Larison's opinion on this sort of revisionism via Google, and I would be curious has to his views (since he knows Byzantine history and its sources, and had an interest in Islam at some point as well).

Update: OK, probably crap.

Update: OK, Larison might be talking about a somewhat different model.

Labels: ,

My review of The Faith Instinct   posted by Razib @ 11/19/2009 02:03:00 AM

Over at ScienceBlogs now. It's a dense book and I only focused on a few major elements. Like the God of the philosophers sometimes it seems like attempts to analyze religion always have to face up to the fact that the phenomenon is awesomely complex, and we look through the glass darkly.


Wednesday, November 18, 2009

The Netherlands & SDA   posted by Razib @ 11/18/2009 11:17:00 PM

In case you didn't know, the SDA Archive has more than the GSS. For example, something called the Dutch Prejudice Survey 1998. Poking around, I confirmed a general trend you see in the GSS, more educated people tend to be ideologically polarized:

Though I am skeptical that more education makes one more intelligent, I do think that education can make one more reflective about one's beliefs and align those beliefs more coherent with one's political preferences. Since everyone in the mainstream seems to agree that college is something that more and more young people should do, we can expect fewer totally incoherent swing voters, but also more ideological polarization.

Or look at this, a massive increase in the "None" category in regards to religion down the age cohorts, and in particular a collapse of Catholicism. In the youngest age cohorts Catholics and Protestants are at parity once more.

I assume that this is a function of latency in Catholic secularization vis-a-vis Protestants. Now looking at religious intensity across the two confessions for those under the age of 40, and the difference is stark:

A much larger number of young Protestants are frequency church-goers. Looks like Protestantism went through secularization first, but adapted and bounced back, or is just institutionally more robust in a secular-dominated environment (Dutch Protestants are divided between various groups, according to the degree of liberalism, orthodoxy, etc.). A quick spot check with the WVS in Germany shows that this dynamic is not true in there, where Catholics seem moderately more observant than Protestants.

Labels: ,

Against infotainment   posted by Razib @ 11/18/2009 07:06:00 PM

Steve has an interesting column up this week:
Who will win the Super Bowl? Well, two minutes on Google leads me to a betting site that says the New Orleans Saints are +360, while the Indianapolis Colts are +385. (I don't even know what those numbers are supposed to mean.) Here's another site that has the Colts at 3:1 and the Saints at 4:1, which at least I understand.
So, there you have my fearless forecast: the Saints will meet the Colts in the 2010 Super Bowl, and one of them will win.
You heard it here first.

If you want political predictions, I can check the Intrade market to see that … hey, what do you know? Sarah Palin, Mitt Romney, and Tim Pawlenty are neck and neck for the 2012 GOP nomination.

So, that's my 2012 conjecture: taking a page from the late Roman Republic, the GOP will nominate Palin, Romney, and Pawlenty to run against Obama as a triumvirate.
Do you have a better guess?

I suppose I could obsessively study the political tealeaves to learn the minutia of upcoming elections (such as who this Pawlenty person might be). But how much would I be adding to the sum total of human wisdom?

Not much, I suspect. One thing the press does well is cover political horse races....

In terms of politics and sports, I think there is some juice which sites like FiveThirtyEight, The Audacious Epigone and Applied Statistics can squeeze out through quantitative analysis. Additionally, more qualitative analysis like Kevin Phillips (though Phillips does do a lot of exploration of voting records, the output tends to be verbal and not in percentages) have interesting things to say. Unfortunately, over the past year of reading American history it has become clear to me that it's really hard to evaluate the qual analysts who add genuine value because very few people operate with the appropriate data base to comprehend allusions and implicit pointers they are making.* To be marketable you really have to just reflect conventional wisdom, and play on its margins.

* More specifically, without the historical data base it's hard to detect the more subtle bullshit artists.


Tuesday, November 17, 2009

The Faith Instinct in National Review   posted by Razib @ 11/17/2009 09:08:00 PM

John Derbyshire has review of The Faith Instinct up. He hits the major points well. I should elaborate on something. In Darwin's Cathedral David Sloan Wilson outlines two dimensions of religion, the horizontal and the vertical. The vertical is pretty straightforward, supernatural agents and forces. The cognition of religious ideas. The horizontal is the communitarian aspect of religion which sociologists such as Emile Durkheim focused on. That is, religion's functional role in society. The two are somewhat related of course, but I think it's a neat division which is useful.

I think the vertical aspect probably is a byproduct of cognitive biases we have. In other words, pleiotropy, whereby selection for agency detection, social intelligence, and innate theories of how the world works (folk biology and physics), generate intuitions which we bracket in the category "supernatural" as a response (this ranges from animism to astrology to theism). In contrast, I can see quite clearly how the horizontal aspect can foster group-level success, and so might be a target of selection. But, I don't necessarily think that it is really religion as such which is the target of selection; instead, they are collective and communal impulses. They may be channeled in a religious manner, but clearly can manifest in other ways. This is why I think organized religion, which is hooked into the horizontal dimension, seems to be collapsing more than "spirituality" in many nations. Many of the intuitions which generate religious impulse are strongly biologically specified, so will persist even after indoctrination ceases. By contrast I suspect that the collective and ritualistic impulses can manifest in ways we perceive as secular. Of course, this last point might be a matter of semantics, as evident by the term "political religion".

Labels: ,

Band of Brothers   posted by Razib @ 11/17/2009 03:20:00 PM

At Cognitive Daily, Men often treat their friends better than women do:
The researchers say these three studies show that men are more tolerant of their friends' failings than women. Does this mean that men are more "sociable"? That's less certain. After all, it could be that women value the friendships more, and so are harsher judges when they perceive a betrayal. Regardless of your interpretation of these results, however, it seems that the stereotype of "men harsh, women friendly" is not always valid. In many cases, it can be said that women are less tolerant than men.

The research focused on college roommates. The only area where males were harsher than females in evaluating their roommates was in hygiene. In any case, there's other research which I've drawn upon to suggest that males are much better are scaling up in terms of social units capable of "collective action" than females.


Elite ancient Egyptians had heart disease   posted by Razib @ 11/17/2009 03:05:00 PM

Heart Disease Found in Egyptian Mummies:
"Atherosclerosis is ubiquitous among modern day humans and, despite differences in ancient and modern lifestyles, we found that it was rather common in ancient Egyptians of high socioeconomic status living as much as three millennia ago," says UC Irvine clinical professor of cardiology Dr. Gregory Thomas, a co-principal investigator on the study. "The findings suggest that we may have to look beyond modern risk factors to fully understand the disease."

"Every man a king" in these days indeed.


Monday, November 16, 2009

Germania   posted by Razib @ 11/16/2009 11:20:00 PM

Follow up post on the Isles, the distribution of German Americans.

25th quartile = 0.10
Median = 0.20
75th quartile = 0.30

Correlation(German, English) = -0.16
Correlation(German, American) = -0.74
Correlation(German, Irish) = 0.11
Correlation(German, Scots-Irish) = -0.31

Now let's exclude the South, where there are the fewest Germans.

25th quartile = 0.21
Median = 0.27
75th quartile = 0.39

Correlation(German, English) = -0.55
Correlation(German, American) = -0.47
Correlation(German, Irish) = -0.30
Correlation(German, Scots-Irish) = -0.37

FYI, the correlation between the frequency of German Americans as a proportion of the non-Hispanic white population and voting for Barack Obama is 0.21. The strong inverse relationship between the proportion of "Americans" and German Americans is in part a function of region. Germans are underrepresented in the Southeast quadrant of the country, where Americans are overrepresented. But that still punts the question as to why Americans define themselves in this way.

I think this is plain history. Though a minority of German Americans have ancestors who arrived in the 18th century (including Dwight Eisenhower), the German American presence in the United States dates to the period between 1840 to 1890. This is recent enough that it probably explains the vociferousness of Germanophobia during World War I, when there was still a German language school system extant in the United States. Lawrence Welk was born in the German community of South Dakota, and German was his first language, explaining his slight accent (whether this was affected or not is controversial).

By contrast, as documented in Albion's Seed, most of the ancestors of British Americans arrived in the 18th century. In fact, in New England it may be that most of the English origin population (and their descendant who spread into upstate New York and the Midwest) descend predominantly from 20-30,000 Puritan men and women who arrived in the Great Migration of 1620-1640 (when the religious climate in England was hostile to Puritanism). The ancestors of the Scots-Irish arrived in the 18th century or earlier, peaking in the decades before the American Revolution.

Of course, as is obvious in previous maps, Americans tend to concentrate in the South, whose predominant wave of settlement was a century after New England. So antiquity is not all that is at work. It seems possible that the "Englishness" of New Englanders is in part a function of the fact that the immigration of Irish Catholics made their Protestant English identity more salient. By contrast, in the South the large numbers of blacks, or the relative nearness of blacks, allowed for the hybridization of the "Anglo-Celtic" white identity, which can be labelled as "American" (i.e., "real Americans"). It is interesting that the Germans in Texas still tend to identify as Germans. For Texas:

Correlation(German, American) = -0.60
Correlation(German, English) = 0.16
Correlation(German, Scots-Irish) = -.24

Correlation(American, English) = -0.37
Correlation(American, Scots-Irish) = -0.36

Remember that Germans who settled Texas were generally recent immigrants from Germany. By contrast, the Anglos who settled Texas were secondary immigrants from the South, often regions of later settlement such as Tennessee, settled from the Atlantic states such as South Carolina in the first place.

Addendum: As some commenters have noted, there is also a likely bias in terms of the most recent immigrant lineage. So in the many individuals with both German and Anglo-Celtic ancestry, the former is likely to have been more recent and memorable.


South Park does its homework   posted by Razib @ 11/16/2009 07:46:00 PM

Years ago there was a South Park episode which commented on the primitive nature of Canadian transportation. Turns out that there was some truth to the jibe (via Tyler).

Labels: ,

The Isles in America   posted by Razib @ 11/16/2009 01:51:00 AM

It's easy to find maps of American ancestries, but I wanted to play around with the data, and in particularly the visualization myself. So I went to the Census and got the county level numbers. The first thing I wanted to do was look at non-Hispanic white ethnicities as a proportion of non-Hispanic whites. That would for example increase the Anglo-Saxon character of the lowland South because it would remove African Americans from the equation.

All the data was from the 2000 Census, and I simply divided the % of each European ancestry group by the non-Hispanic white percentage to reweight appropriately. Here are some correlations I found:

English X Scots-Irish = 0.34

English X Irish = 0.30

English X American = -0.20

Scots-Irish X Irish = 0.37

Scots-Irish X American = -0.25

Irish X American = -0.45

I left the Scottish and Welsh out of this because their numbers were relatively small. One of the main issues with look at the "Irish" and "American" category is that both of these are probably heavily loaded with Scots-Irish. Below the fold are some maps I generated.

Blue = above the median for the frequency of that group nationally (the median being calculated again with non-Hispanic whites only included).

Red = below the median.

The distributions of frequencies by county tend to be positively skewed, so the shading is covering a larger spectrum of frequencies in the blue than the red.

Min = 1.6%
25% = 8.5%
Median = 11%
75% = 14%
Max = 48%

Min = 0%
25% = 1%
Median = 2%
75% = 3%
Max = 10%

Min = 2%
25% = 10%
Median = 12%
75% = 14%
Max = 37%

Min = 0%
25% = 7%
Median = 14%
75% = 22%
Max = 70%

"Isles" includes Scottish & Welsh, as well as "American."

Min = 9%
25% = 39%
Median = 44%
75% = 51%
Max = 85%

Finally, here's a map where those of "Isles" origin are 50% or more of the non-Hispanic white population.

The shading for the "Isles" doesn't look right. But here's the histogram:

The median is 0.45. So that's probably why the blue is relatively homogeneous, the distribution is negatively skewed.


A simple framework for thinking about cultural generations   posted by agnostic @ 11/16/2009 12:44:00 AM

In this discussion about pop music at Steve Sailer's, the topic of generations came up, and it's one where few of the people who talk about it have a good grasp of how things work. For example, the Wikipedia entry on generation notes that cultural generations only showed up with industrialization and modernization -- true -- but doesn't offer a good explanation for why. Also, they don't distinguish between loudmouth generations and silent generations, which alternate over time. As long as a cohort "shares a culture," they're considered a generation, but that misses most of the dynamics of generation-generation. My view of it is pretty straightforward.

First, we have to notice that some cohorts are full-fledged Generations with ID badges like Baby Boomer or Gen X, and some cohorts are not as cohesive and stay more out of the spotlight. Actually, one of these invisible cohorts did get an ID badge -- the Silent Generation -- so I'll refer to them as loudmouth generations (e.g., Baby Boomers, Gen X, and before long the Millennials) and silent generations (e.g., the small cohort cramped between Boomers and X-ers).

Then we ask why do the loudmouth generations band together so tightly, and why do they show such strong affiliation with the generation that they continue to talk and dress the way they did as teenagers or college students even after they've hit 40 years old? Well, why does any group of young people band together? -- because social circumstances look dire enough that the world seems to be going to hell, so you have to stick together to help each other out. It's as if an enemy army invaded and you had to form a makeshift army of your own.

That is the point of ethnic membership badges like hairstyle, slang, clothing, musical preferences, etc. -- to show that you're sticking with the tribe in desperate times. That's why teenagers' clothing has logos visible from down the hall, why they spend half their free time digging into a certain music niche, and why they're hyper-sensitive about what hairstyle they have. Adolescence is a socially desperate time, not unlike a jungle, in contrast to the more independent situation you enjoy during full adulthood. Being caught in more desperate circumstances, teenagers freak out about being part of -- fitting in with -- a group that can protect them; they spend the other half of their free time communicating with their friends. Independent adults have fewer friends, keep in contact with them much less frequently, and don't wear clothes with logos or the cover art from their favorite new album.

OK, so that happens with every cohort -- why does this process leave a longer-lasting impact on the loudmouth cohorts? It is the same cause, only writ large: there's some kind of social panic, or over-turning of the status quo, that's spreading throughout the entire culture. So they not only face the trials that every teenager does, but they've also got to protect themselves against this much greater source of disorder. They have to form even stronger bonds, and display their respect for their generation much longer, than cohorts who don't face a larger breakdown of security.

Now, where this larger chaos comes from, I'm not saying. I'm just treating it as exogenous for now, as though people who lived along the waterfront would go through periods of low need for banding together (when the ocean behaved itself) and high need to band together (when a flood regularly swept over them). The generation forged in this chaos participates in it, but it got started somewhere else. The key is that this sudden disorder forces them to answer "which side are you on?" During social-cultural peacetime, there is no Us vs. Them, so cohorts who came of age in such a period won't see generations in black-and-white, do-or-die terms. Cohorts who come of age during disorder must make a bold and public commitment to one side or the other. You can tell when such a large-scale chaos breaks out because there is always a push to reverse "stereotypical gender roles," as well as a surge of identity politics.

The intensity with which they display their group membership badges and groupthink is perfectly rational -- when there's a great disorder and you have to stick together, the slightest falter in signaling your membership could make them think that you're a traitor. Indeed, notice how the loudmouth generations can meaningfully use the phrase "traitor to my generation," while silent generations wouldn't know what you were talking about -- you mean you don't still think The Ramones is the best band ever? Well, OK, maybe you're right. But substitute with "I've always thought The Beatles were over-rated," and watch your peers with torches and pitchforks crowd around you.

By the way, why did cultural generations only show up in the mid-to-late 19th C. after industrialization? Quite simply, the ability to form organizations of all kinds was restricted before then. Only after transitioning from what North, Wallis, and Weingast (in Violence and Social Orders -- working paper here) call a limited access order -- or a "natural state" -- to an open access order, do we see people free to form whatever political, economic, religious, and cultural organizations that they want. In a natural state, forming organizations at will threatens the stability of the dominant coalition -- how do they know that your bowling league isn't simply a way for an opposition party to meet and plan? Or even if it didn't start out that way, you could well get to talking about your interests after awhile.

Clearly young people need open access to all sorts of organizations in order to cohere into a loudmouth generation. They need regular hang-outs. Such places couldn't be formed at will within a natural state. Moreover, a large cohort of young people banding together and demanding that society "hear the voice of a new generation" would have been summarily squashed by the dominant coalition of a natural state. It would have been seen as just another "faction" that threatened the delicate balance of power that held among the various groups within the elite. Once businessmen are free to operate places that cater to young people as hang-outs, and once people are free to form any interest group they want, then you get generations.

Finally, on a practical level, how do you lump people into the proper generational boxes? This is the good thing about theory -- it guides you in practice. All we have to do is get the loudmouth generations' borders right; in between them go the various silent or invisible generations. The catalyzing event is a generalized social disorder, so we just look at the big picture and pick a peak year plus maybe 2 years on either side. You can adjust the length of the panic, but there seems to be a 2-year lead-up stage, a peak year, and then a 2-year winding-down stage. Then ask, whose minds would have been struck by this disorder? Well, "young people," and I go with 15 to 24, although again this isn't precise.

Before 15, you're still getting used to social life, so you may feel the impact a little, but it's not intense. And after 24, you're on the path to independence, you're not texting your friends all day long, and you've stopped wearing logo clothing. The personality trait Openness to Experience rises during the teenage years, peaks in the early 20s, and declines after; so there's that basis. Plus the likelihood to commit crime -- another measure of reacting to social desperation -- is highest between 15 and 24.

So, just work your way backwards by taking the oldest age (24) and subtracting it from the first year of the chaos, and then taking the youngest age (15) and subtracting it from the last year of the chaos. "Ground zero" for that generation is the chaos' peak year minus 20 years.

As an example, the disorder of the Sixties lasted from roughly 1967 to 1972. Applying the above algorithm, we predict a loudmouth generation born between 1943 and 1957: Baby Boomers. Then there was the early '90s panic that began in 1989 and lasted through 1993 -- L.A. riots, third wave feminism, etc. We predict a loudmouth generation born between 1965 and 1978: Generation X. There was no large-scale social chaos between those two, so that leaves a silent generation born between 1958 and 1964. Again, they don't wear name-tags, but I call them the disco-punk generation based on what they were listening to when they were coming of age.

Going farther back, what about those who came of age during the topsy-turvy times of the Roaring Twenties? The mania lasted from roughly 1923 to 1927, forming a loudmouth generation born between 1899 and 1912. This closely corresponds to what academics call the Interbellum Generation. The next big disruption was of course WWII, which in America really struck between 1941 and 1945, creating a loudmouth generation born between 1917 and 1930. This would be the young people who were part of The Greatest Generation. That leaves a silent generation born between 1913 and 1916 -- don't know if anyone can corroborate their existence or not. That also leaves The Silent Generation proper, born between 1931 and 1942.

Looking forward, it appears that these large social disruptions recur with a period of about 25 years on average. The last peak was 1991, so I predict another one will strike in 2016, although with 5 years' error on both sides. Let's say it arrives on schedule and has a typical 2-year build-up and 2-year winding-down. That would create a loudmouth generation born between 1990 and 2003 -- that is, the Millennials. They're already out there; they just haven't hatched yet. And that would also leave a silent generation born between 1979 and 1989.

My sense is that Millennials are already starting to cohere, and that 1987 is more like their first year, making the silent generation born between 1979 and 1986 (full disclosure: I belong to it). So this method surely isn't perfect, but it's pretty useful. It highlights the importance of looking at the world with some kind of framework -- otherwise we'd simply be cataloguing one damn generation after another.

Labels: , , ,

Saturday, November 14, 2009

Faith as an adaptation   posted by Razib @ 11/14/2009 06:11:00 PM

Nicholas Wade has an article up in The New York Times, The God Gene, which serves as a precis of the central arguments of The Faith Instinct, his new book. The title is catchy, but it should really be "The God Phene." Depending on how you measure it, religiosity is a heritable trait, with its variance being controlled by variance across many genes. There is as likely to be a "God Gene" as a "Smart Gene" or "Height Gene." In other words, not too likely.

I have been putting off putting up a review of The Faith Instinct because there's a lot of ground to cover. The portions which emphasized the role of common belief, "imagistic arousal" and ritual in cementing common bonds among men and allowing for maximal force of collective action were persuasive to me. As someone who has never served in the military I am not personally familiar with the "band of bothers" dynamic, but the role of chanting, posing and synchronous mindfulness & action in sport is obvious. It's no coincidence that high stakes athletics and religion tend to go hand & hand. Wade's references to William McNeill's Keeping Together in Time: Dance and Drill in Human History were very intriguing, and I have to check that book out at some point.

Though it is clear to me that there is utility in tribal gods binding a deme together to engage in collective action, I am more skeptical of the central function which Wade places upon religion as a driver of the cognitive biases which are likely to predict religion. Women are more religious than men. One plausible explanation for this is more men than women are socially retarded, and it is social retards who find supernatural agents less intuitively plausible, and are also liable to admit to this belief and not conform with the modal norms of society. The thesis in The Faith Instinct is that group level selection, on the level of tribal units, selected for those demes where religiosity was more pronounced, as those groups could engage in more effective collective action. Much of the argument is derived from Samuel Bowles from what I can tell. The problem of course is that the sex engaged in the warfare which is the specific manifestation of intergroup competition and subject to natural selection, males, seem to be less predisposed to belief in supernatural agents. Of course sex differences should be slow to evolve, so it suggests that if selection was operative upon religion as a trait it hasn't swept away all the various cobwebs of evolutionary history in terms of the lower-level traits which come together to form the religious phenotype.

An alternative model for why religion is universal in humans from the adaptationist one is that it is a byproduct of various other cognitive traits which are useful, just as heat is produced during work. More specifically, in books like Religion Explained & In Gods We Trust cognitive anthropologists Pascal Boyer & Scott Atran argue that basic intuitions which naturally lead one to supernatural inferences derive from extremely useful cognitive features; agency detection, theory of mind, and flavors of folk psychology. Supernatural intuitions don't constitute religion, and Wade et al. are not suggesting that it is simply theism which confers a selective benefit, but rather the entire cultural package of religious belief & practice, the "integrative" as well as the supernatural aspect. The problem that seems to emerge from these overlapping models is that I do not see why group selection dynamics operating upon biological traits are necessary to explain religious instincts as we see them today. Religion just doesn't seem that tightly integrated of a feature, but a more diffuse phenotype (as evident by the novel fusion of philosophy with religion which occurred during the Axial Age). Rather, it seems a cultural adaptation which hooks into previously extant and ubiquitous psychological intuitions.

But a fuller review at ScienceBlogs soon.

Labels: ,

Friday, November 13, 2009

Height doesn't always matter....   posted by Razib @ 11/13/2009 02:39:00 PM

How universal are human mate choices? Size doesn't matter when Hadza foragers are choosing a mate:
It has been argued that size matters on the human mate market: both stated preferences and mate choices have been found to be non-random with respect to height and weight. But how universal are these patterns? Most of the literature on human mating patterns is based on post-industrial societies. Much less is known about mating behaviour in more traditional societies. Here we investigate mate choice by analysing whether there is any evidence for non-random mating with respect to size and strength in a forager community, the Hadza of Tanzania. We test whether couples assort for height, weight, BMI, percent fat and grip strength. We test whether there is a male-taller norm. Finally, we test for an association between anthropometric variables and number of marriages. Our results show no evidence for assortative mating for height, weight, BMI or percent fat; no evidence for a male-taller norm; and no evidence that number of marriages is associated with our size variables. Hadza couples may assort positively for grip strength, but grip strength does not affect the number of marriages. Overall we conclude that, in contrast to post-industrial societies, mating appears to be random with respect to size in the Hadza.

Here's some stuff from the discussion:

Overall, however, our analysis suggests size and strength are not greatly important when Hadza are choosing a mate. This lack of size-related mating patterns might appear surprising, since size is usually assumed to be an indicator of health, productivity and overall quality. But health and productivity may be signalled in alternative ways in the Hadza, who are a small, relatively homogeneous population. An individual's health history may be more important than size, for example, and this may be relatively well known in a small, mobile population. Additionally, there may be some disadvantages to large size in food-limited societies, where the costs of maintaining large size during periods of food shortage may be high. Such disadvantages will not be seen in food abundant societies, so that large size may be a better indicator of quality in postindustrial populations. Finally, research on another African forager population found that height is negatively correlated with hunting returns (Lee 1979), suggesting that tall height may not be an indicator of productivity in such economies.

Here's a chart which shows the proportion of females-taller-than-male marriages by culture:

In a previous post I suggested that the shift from small-scale societies to agricultural societies witnessed a transition from an emphasis on innate individual level social intelligence toward rules and heuristics (in other words, wisdom embodied in the preferences of society and its institutions). External physical characteristics are correlated with "health," so they're useful. And those who are not physically attractive can signal their own status and abilities in other ways, ugly fat men can for example buy material signalers to show that they have something going on. It strikes me that the Wisdom of Seinfeld is most appropriate for large urban areas with some degree of anonymity. Quick & dirty signalers to filter and influence one's choices are critical in the incredibly large number of human interactions possible in these urban agglomerations. By contrast, if George Costanza lived in a village one would know enough about his persona to dismiss a random "pairing" with an attractive woman as an aberration (or, one would know the back-story to this bizarre pairing).

As our modern post-industrial society shifts toward information transparency perhaps we'll become less "shallow"? Remember the 1995 film Species, the attractive alien character met a handsome male at a night club. She assessed his fitness through his looks to make the initial choice. But later she killed him when she found that he was a diabetic. If she'd been able to access his health profile on her iPhone perhaps he would have been able to live for another day?

Labels: ,

Thursday, November 12, 2009

A quantitative ecologist looks at world history (again)   posted by Razib @ 11/12/2009 07:30:00 PM

Doing a literature search on the Price Equation for some weblog posts I found that Peter Turchin had written a new paper on world history using Price's formalism explicitly. A quantitative ecologist by training, Turchin has already written a series of books attempting to model human history in a more formal fashion than is usually the case. Though his work has a tendency to overlap with economic history, in particular cliometrics, Turchin brings a more robust theoretical toolkit from the natural sciences to the table. An ecologist once told me that the ultimate aim of his career was to "count stuff," and that professional expertise is handy when it comes mapping the distribution & abundance of the human species over time. What David Sloan Wilson is to multilevel selection theory, Peter Turchin is to "cliodynamics", his attempt to grapple with the general dynamics which characterize the cycles of human history.

Specifically, Turchin focuses the agricultural societies which mark the span between the age of the hunter-gatherers, and the industrial revolution. What I term the "traditionalist transient." Traditionalist because from our "modern" viewpoint we perceive many of the customs, institutions and values of this age as timeless and traditional. This despite the fact that they emerged in a specific and relatively brief historical transient after the Ice Age & before the Great Divergence. But this period is still important in our modern age, because the basic building blocks of contemporary identities draw from the traditionalist transient. Higher religions invariably date back to this period (Salafi Muslims and some ultra-traditionalist Roman Catholics look back to particular periods during the traditionalist transient as golden ages to be emulated), as do modern lingua francas, and the basic terms of political organization (democracy, republic, etc.). Early modern thinkers of the Enlightenment may have rejected or superseded the orthodoxies of the traditionalist transient as the Quarrel of the Ancients & Moderns was finally resolved to the satisfaction of the latter, but the context of the refutation was still to a large extent on the basis of traditionalist transient assumptions. For example, contemporary secularism, laicism and disestablishmentarianism are intelligible only in light of the fusion of the sacred and temporal which played out in the agricultural societies after the rise of Sumer. Turchin always notes that his conclusions may not, likely do not, apply to the dynamics extant in the present. But I suspect that much of what does go on in the present is intelligible only in light of the phenomena of the past. So his models are not purely abstract intellectual exercises.

Whether you think the project as a whole is worthwhile or not (see Massimo Pigliucci's skepticism), I am intrigued by the fact that Turchin focuses on Inner Asia because this is one region of the world which has long been "at the center of it all," both literally as well as more metaphorically. Some of the inferences from Turchin's framework illuminate rather well the broad observations and hints presented in Christopher Beckwith's Empires of the Silk Road. Since Beckwith is a philologist he lacks Turchin's more robust theoretical toolkit, but he naturally exhibits both more depth and granularity when it comes to the details of the history and ethnography. Setting both against each other is fascinating as one can make out binocular intellectual vision with more subtly than if just one narrative is considered.

As I said, in the paper, Warfare and the Evolution of Social Complexity: A Multilevel-Selection Approach, Peter Turchin uses the Price Equation as a theoretical framework. The reason for this is that Turchin believes that human societies within the past 10,000 years can be viewed as functional units subject to selection; in other words, they're adapting entities, organisms. The Price Equation allows one to partition variation between and within groups, variation being necessary for selection to operate upon collections of entities. Though the economist Samuel Bowles has suggested that between group genetic variance (FST) and selection (through warfare) may have values high enough in "small-scale societies" to allow for non-trivial biological group level selection, most seem to accept the contention of those who suggest that between group variance only makes cultural group selection plausibly common. Turchin is in the latter camp, in particular because his focus is not on small-scale societies, but larger polities which characterize what we would term "civilization." In the world of civilization it is clear that between group variance can be much greater culturally than biologically. Consider the example the case of Transylvanian Hungarian Protestants who could get by in late 16th century Oxford by virtue of their common fluency in Latin, combined with shared Calvinist religious precepts with many English Protestants (example from The Reformation: A History). Yet genetically Hungarians are closer to their neighbors, the Orthodox Romanians, than to the British. By intuition and impression it is clear that in relation to gene frequencies religious and linguistic identity tend to exhibit a less clinal pattern of variation. Though genes are discrete units, genetic variation approaches blending dynamics more easily than religious and linguistic variation, where pidgins and syncretisms are often marginalized or absorbed into one of the "parent" traditions.* Because languages and religions vary less gradually, it is easier for one to conceive of a clear and distinct group coherency in a selective framework. Where one entity ends and another begins is not arbitrary, the genes of France blend in to the genes of Germany more gently than do the dialects of French to those of German. It seems that selection between group genetic differences (not reducible to individual level selection) runs up against the problem of "gene flow" overwhelming divergences in frequency (I imagine in pre-modern times this gene flow consisted predominantly of the assimilation of the breeding-age women of conquered tribes). Here's an example from a primitive people:
31:9 And the children of Israel took [all] the women of Midian captives, and their little ones, and took the spoil of all their cattle, and all their flocks, and all their goods.
31:13 And Moses, and Eleazar the priest, and all the princes of the congregation, went forth to meet them without the camp.
31:14 And Moses was wroth with the officers of the host, [with] the captains over thousands, and captains over hundreds, which came from the battle.
31:15 And Moses said unto them, Have ye saved all the women alive?
31:16 Behold, these caused the children of Israel, through the counsel of Balaam, to commit trespass against the LORD in the matter of Peor, and there was a plague among the congregation of the LORD.
31:17 Now therefore kill every male among the little ones, and kill every woman that hath known man by lying with him.
31:18 But all the women children, that have not known a man by lying with him, keep alive for yourselves.

In fact, it seems likely that between group genetic differences have been driven by cultural differences. The Jews are a famous case, but this dynamic also crops up in surprising places, such as among Christian & Muslim Arabs in Lebanon. It seems likely that high FST values in small-scale societies tracks their cultural and linguistic diversity; as conventional gene flow through movement of mates between populations which can not communicate or worship different gods is likely to be dampened due to mutual unintelligibility and suspicion. By contrast, the expansive spread of the possible Y chromosomal lineage of Genghis Khan within the last 1,000 years is a testament to the power of cultural prestige and conquest over large areas to generate rapid gene flow.

The ethnogenesis of the Hazara in Afghanistan illustrate the complex interplay between religion, language and lineage. The Hazaras show clear evidence on Structure-based analyses of being an East Eurasian and West Eurasian hybrid population. Y chromosomal lineages tied to Genghis Khan the Mongols are common among them, and there are historical legends that they arose from exactly this group. Their practice of Islam & usage of a Persian dialect likely dates to the time that the Mongol Khans of the Ilkhanate accepted the religion of the local majority in the 13th century. The Mongols who refused to accept Islam emigrated to Inner Asia, while those who remained assimilated, accepting the religion and dominant language of the local populations whom they ruled. Today as a physically distinct Shia Muslim Persian speaking group in Afghanistan the Hazara are now endogamous, their biological distinctiveness a function of broader cultural-historical forces.

Of course that is just a specific concrete illustration of historical dynamics, and an atypical one at that. The general processes which Peter Turchin discusses have an extreme specific case scenario in the form of the Mongols, who left an outsized impact on the World Island. A big shift from small-scale societies to those of agricultural civilizations is the need for complex hierarchies to mediate decision making from the apex of the political pyramid. Like Archimede's lever with which he could move the world, the functional integrity of these units allowed the decisions of Genghis Khan to affect tens of millions. It is presumed that for small-scale societies primary face-to-face interaction sufficed to coordinate decision making. Interestingly, there is also compelling data which points to relative egalitarianism of material wealth to complement the flat authority structures. By the time history arises to supplement archaeology, meaning that we have records in the form of cuneiform tablets, societies are clearly already quite hierarchical (literacy probably emerged as a more sophisticated form of accounting, so rather complex economies are already necessary conditions). A reliance on rules, heuristics and institutions which coordinate and channel power tracked the crystallization of a powerful and wealthy rentier class (as well as a possible reallocation of power between the sexes). The idea that the poor will always be with us, and that true status and nobility accrue to those who can consume at leisure, as opposed to those who increase productivity, was the norm in the traditionalist transient.** Whereas in small-scale societies alien tribes were subhuman, in civilization the elites would often dehumanize their own subjects as lower orders of a different nature.

It is this last tendency which Turchin examines in this paper, to contrast it with conflicts which emerge on the "meta-ethnic frontiers." If you have read his earlier work you are familiar with the idea, which basically describes civilizational marchlands. The Muslim Ottomans, Orthodox Cossaks and Buddhist Oirats were all forged in the fires of meta-ethnic frontiers. In opposition to this is the "narcissism of small differences", whereby societies exhibit cleavage along what may otherwise seem to be marginal differences. Civil wars within polities can often by traced back to these issues, or sides aligned based on internal factions. Consider the divisions between Protestants which resulted in the English Civil War. Turchin wishes to assess the extent of ethnocide and genocide in the former vs. latter cases. Why? Because he believes that it is the former cases which are responsible for the rise of large empires and re-ordering of civilization and historical shifts. Evolutionary theory tells us that selection needs extant variation to operate, and it seems that along meta-ethnic frontiers such variation would would be extant in more copious quantities than in civilizational heartlands. In particular, along the boundaries of civilizations. Within individual societies there should be less variation, so selective forces should have less traction. To assess this he reviews the literature to evaluate the magnitude of depopulation wrought upon cities by victorious armies. This is a classic form of "hard selection".

Table 1 shows that conflict on meta-ethnic frontiers does have a stronger effect than civil wars. Why? Turchin posits a simple psychological explanation: those of different civilizational character are dehumanized, so empathy is modulated downward (it is notable that during and before the Albigensian Crusade the Cathars were subject to many conventional demonizing tropes, and effectively de-Christianized in the eyes of the rest of Christendom) .This is clear from the history of Christianity and Islam, in the medieval period the religious norms in both civilizations accepted the enslavement and maltreatment of unbelievers to an extent not acceptable for believers. This is the explanation for why some of the Christian military orders in the Baltic, whose original raison d'etre was to Christianize the native peoples, actually were among the last to allow and encourage baptism of their subjects. Baptism imposed constraints on efficient extraction of marginal product. The analogy to New World chattel slavery here is clear, as some plantation owners viewed proselytization among blacks dimly lest economics be modulated by morality. When the crossbow was invented the Roman Catholic Church attempted to ban its usage between Christian powers, though declared it acceptable as a weapon against Muslims. This sort of behavior, constraining and/or ritualizing high stakes competition and conflict within groups, while accepting a more "no holds barred" attitude toward between group conflicts is known from small-scale societies (though perhaps the contrast would manifest more in the extent of extreme barbarism with which outgroups were treated, rather than any particular norm of humanity for ingroups). Civilization simply operationalized this on a grander scale, and scaffolded human nature and channeled it through particular institutions and identities.

Peter Turchin argues, and presents data, that these frontier regions where primitive, and frankly savage, passions are channeled toward outgroups serve as the loci for new empires and mega-polities. In particular, being an ecologist, he focuses on particular ecologies as the exceptional cauldrons for state-formation: the semi-arid steppe. It is here that Turchin aligns with some of Christopher Beckwith's insights in Empires of the Silk Road. This should not be totally surprising, though we look through the glass darkly nature is fundamentally one, and history is a phenomenon rooted in nature. Beckwith attempts to generate a revisionist history of the world where the rise and fall of nomadic empires are just as salient as the ebb and flow of peasant-based civilizations, where the eruptions from the heartland echo down the centuries. And Turchin, like Beckwith, seems to hold that it is less important or relevant that the movers of history are unlettered nomads, but that they are derived from the marchlands where nomads or part-nomads are prominent on both sides of the frontier, civilized and barbarian. Consider the Cossacks who pushed the frontier of the Russian Empire back against the Tatars from the 16th to the 18th centuries. Loyal to the Russian state, Orthodox Christians, and trailblazers for Slavic culture and society, they certainly were the civilizational antithesis of the Muslim Turks descended from the Golden Horde of the Mongols. But upon closer inspection there were similarities between the Cossacks and Tatars as rough frontiersmen which exposed affinities when set against the lives of Russian peasants, or the more cultured nobility of the Russian heartland. It is known that many Tatars "went Cossack," abandoning the Muslim religion and eventually shifting toward a Slavic self-identity in the wake of defeats (though this is most notable for elite Tatars who converted to Christianity and were assimilated into the Russian noble class). This is what Turchin would label "ethnocide," cultural extermination if not physical. The men who expanded the domain of Russian civilization in the early modern era were useful barbarians (this seems especially evident to the more sophisticated and European-oriented Russian rulers of the 18th century, who relied upon and disdained & feared, the Cossack). The Turks who had crushed the first efflorescence of Slavic civilization which ran from Kiev to Novgorod were less useful barbarians. But to Turchin this distinction is not particularly important, as civilization-destroying barbarians such as the Arabs after Muhammad and the Mongols can set the stage for the emergence of a new civilizational-system (in the case of the Mongols there was of course the brief world-system of the Pax Mongolica).

So what do the data say?

If these data & the result, repackaged in a statistically significant form are any surprise to you, I recommend you read some books. The importance of the semi-arid steppe, the rise of the mounted cavalry and utility of the reflex bow are plain in the record of civilized societies. Macedonia, the Zhou & Chin, Persia the Turks are just a few examples of peoples who are barbarian or semi-barbarian, and came blazing out of the marchlands to establish a new order over civilized peoples. Naturally the horse looms large here. It is a truism that the average peasant was never more than 10 miles from where they were born. Even if the exact value on this expectation is off, the general thrust is surely correct, in the Malthusian world the average sedentarist was quite sedentary. In contrast the mounted nomads were highly mobile, with whole peoples such as the Avars migrating en masse from Mongolia to the Hungarian plain on the order of a decade! More mobile units of males operated on the scale of years, as was shown by the Turks and the Mongols whose zone of control spanned the margins of the Pacific to the Black Sea. The Mongols were simply the apotheosis of the terror and savagery which mobile calvary could inflict upon settled populations. In the classical period the Scythians and their fellow travelers ranged widely in their depredations, causing havoc in their wake. It is often forgotten that the Huns who were menacing Gaul in the 450s were strafing Syria ~400, sweeping down through the Caucasus from the plains to the north of the Black Sea. Just as the institutions of the traditionalist transient allowed for individuals at the apex of power to control and affect massive change at a distance, so the rise of the horse and bow gave the nomad a combination of mobility and lethality which was only neutralized with the spread of firearms.

Turchin dates the emergence of the nomadic warrior toolkit, and therefore the potential to wreak havoc on civilized polities, to the period between 1000-500 BCE. This sounds about right. Though the records are sparse because literate civilization was thin on the ground, this is the period when the Scythians battered the Assyrian Empire, and the Medes and Persians finally sacked Nineveh. In China the rise of this sort of nomadic lifestyle and warfare seems to have taken a bit longer, with the Xiongnu making permanent the archetype of the raw nomad beyond the frontiers of Han civilization in the 3rd century BCE. But a more critical point is that there is the suggestion that the Axial Age is a deterministic reaction on the part of civilized peoples to the hammer-blow which nomad polities dealt them (in the case of Persia you have the barbarians overwhelm, assimilate and re-order the civilizations of the Near East in totality). To Turchin this is an evolutionary process, as selection operates upon cultures and polities to give rise to adaptations to a new fitness landscape. In this case, the mounted archer, a combination of lethality and mobility which the more primitive modalities of the Bronze Age were helpless. The choice was clear, adapt or be swallowed.

To me it is notable that the Assyrians are reputed to have been particularly savage, while the Persians who were their eventual successors were depicted as relatively benevolent. Some of this is selection bias, as the Persians treated the Jews with less brutality than the Assyrians, and much of our character/narrative history of this place and period come from the Hebrew Bible. But there are other independent records of the nature of Assyrian rule, which seemed to be rather coarse and overly generous in its application of intimidation and cruelty to the conquered. From what I can tell it is as if the Assyrians were Yanomami in chariots, exhibiting a brutal inchoate savagery more the norm in small-scale societies. In contrast, the Persian system of rule were imperial, but often indirect. Local traditions were respected, but it was under the Achaemedids that the Zoroastrian religion began to develop, which later developed into a state-religion for the Persians in the manner that Christianity was for Rome and State Confucianism in China. It is likely that the ethical aspect of Judaism as an ethical monotheism came to the fore during the period of influence under the Persian Zoroastrians, whose primary deity is, Ahura Mazda, is an explicit force for good, not an angry and jealous god. In Empires of the Silk Road Christopher Beckwith suggests that the synchronous efflorescence of religio-philosophical systems across the ecumene during the Axial Age was promoted by the expansion of nomads, their ideas, and the facilitating role of their trade networks. Turchin's model would seem to suggest that nomads played a role as well, but more as antagonists for civilized polities (and in some cases the progenitors of new polities and empires), who had to increase in scale and develop institutional and ideological adaptations. The two models are not mutually exclusive. In terms of religion there are many cases of barbarians beyond the limes being influenced and innovating. Most famously with Islam, but in both Scandinavia and the Baltic before the conversion to Christianity the extant records and preserved mythologies are clear enough to show an influence and institutional mimicking of the "Roman religion." In the latter cases the cult of Baldr and the temple at Arkona were dead-ends, as Christianization eliminated those cultural experiments. A more successful universal religious model is the worship of Tengri, the sky god of the Turks and Mongols who was the focus of worship their "shamanistic" phase. The similarities of Tengri to El, one of the ancestors of the Abrahamic God, can not be a coincidence. Sky gods are portable and plainly omnipresent, looking down from on high, in a way that makes them ideal candidates for the God.

It seems that the model presented here is that from savagery comes civilization. This is basically an evolutionary model of human history, an "arms race" of ideas and institutions between polities and civilizations. Sometimes, as in the case of the mounted warrior with bow, the race was triggered by a technological change. The civilizations of the Near East, from Egypt to Mesopotamia, did not change or adapt fast enough. They became fiefs in a Persian world. It can be argued that the Classical Greeks, descendants of the barbaric city-states which sacrificed humans to placate savage gods as they were falling to the Sea Peoples, did formulate appropriate institutional (the cohesive polis and national identity) and technological (the accoutrements and organization of the hoplite phalanx) responses to the threat. In China the two dynasties which set the tone for Imperial China down to 1900, the Zhou and the Chin, emerged out of the borderlands as semi-barbarian polities. The Zhou introduced the peculiar elite Chinese variant of supernaturalism whereby worship centered around the impersonal "Heaven." The Chin state was organized around an efficient and utilitarian plan which may have been repudiated in name, but not totally in practice, by subsequent dynasties. Reorganized from within by useful barbarians China was ready to meet the nomad threat in the form of the Xiongnu.

In The Human Web: A Bird's-Eye View of World History John and William McNeill point out that history seems to have a direction in terms of human complexity. Between 1200 and 800 BCE the Greeks "forgot" how to write in totality, so that the Linear B system of the Mycenaeans has no connection to the Phoenician derived alphabet of the later period. As the ages progressed these sorts of "Dark Ages" when the clock was reset, the slate wiped clean, became less and less frequent. The world of settled humanity, dominated by rentier elites, purporting to justify their domination through transcendent truth, covered the face of the earth. Ray Haung, in China: A Macro History, observes that the interregnums between dynasties exhibits a persistent decline in length. Why? One hypothesis is that the "Chinese system," as embodied in norms and values passed down through its bureaucratic class, became more robust to the "exogenous shocks" of political chaos. Some, such as Robert Wright in Nonzero: The Logic of Human Destiny, would explain this in generally "microeconomic" terms. Game theory applied through the lens of rational actors. Turchin rejects this view insofar as he seems to be suggest that cultural units are being driven to success because of a process of expansion through the elimination of rivals. Again, there is no need to assume these are exclusive alternative choices. Religions such as Christianity and Buddhism clearly spread through individual action and choice (both seem to have been popular first among cosmopolitan urbanites, and counter-elites). But, they also clearly spread by being adopted as ideological cement for polities, the choice made on high at the apex of the political pyramid and extended down by fiat to the population as a whole. This choice may have conferred upon the polity the benefits which accrue from being members of a meta-ethnic civilizational coalition. The benefits to being members of Christendom for the pagan elites to the north of the post-Roman world were clear. James I of England asserted "No bishop, no King," to indicate the necessary connection between ruler and the priests. And so it was the arrival of Christianity seems curiously concomitant with the emergence of kings on pagan Europe, one God, one ruler. Pagan peoples who remained relatively disinclined toward joining the Christian Commonwealth were liable to be subject to ethnocide, as occurred with the Wends and the Old Prussians.

A focus on elites is evident in Turchin's model, and I think in some ways that is a critical piece of information. Group level selection on the scale that he focuses upon, large polities and such, may be a feature of only a small slice of a given polity. The elites, or particularly important military groups, such as the Cossacks. History is written by and for the elites. The gods and languages of the elites, their norms, often percolate down to the masses (though not always, my example above about a Latin speaking Transylvanian in Oxford is obviously extremely elitist, but in terms of international politics they are all who counted!). Greg Clark documented high mortality rates for the military nobility of England in Farewell to Alms, as opposed to the relative fertility of the pacific gentry. This shows how high the stakes for intergroup conflict for elites may have been, as opposed to commoners. Benjamin Franklin reputedly stated that "We must all hang together, or assuredly we shall all hang separately." Kevin Phllips reports in The Cousins' Wars that much of the Virginia planter aristocracy, including George Washington and Thomas Jefferson, were in severe debt to English financial houses. Their individual material stakes illustrate how victory or loss in war can have huge negative or positive outcomes to those at the apex of the power pyramid in agricultural societies. This is where I think the civilizational values which unite and engender a degree of cosmopolitanism among elites within the bounds of that broader meta-ethnic framework serve to dampen the savagery of loss and the gluttony of gain. The world of a defeated king may seem to collapse upon him, but if the foe shares the same civilizational presuppositions the institutions and values remain intact, and some honor and status may be retained by the rules of the game which are enforced by third parties (e.g., in medieval Europe, the Church). By contrast, it is no surprise that when the kingdom of the Visigoths fell to the armies of the Arabs and Berbers the elites either fled, or, more likely converted to Islam to preserve their positions. This was ethnocide. A process which was inverted in 1492, as Granada fell to the armies of Castile and Aragon, and the Muslim elites either had to flee to North Africa, or convert to Christianity. In the former case they lost their wealth and power. In the latter case they lost their identity. These are of course the less savage cases, on occasion elites are simply exterminated by the conquerors so that the snake's head is removed.

If Peter Turchin turns this most recent paper into a book, I have a catchy title in mind: Civilization: a tale of regicide. It has been said that science can not progress until old ideas die with old scientists, and so it may be that civilization can not proceed until old elites die prematurely thanks to the efforts of new ones. The argument is too broad to be sure, but the history of the evolution of power is a biography of the lives of those in power, so this captures much to a first approximation if it is correct.

* English has a strong influence from the Romance languages via French, but it is still recognizably a Germanic language. Similarly, Sikhism emerged as a new world religion or sect which navigated between the disjoint idea spaces which defined Hinduism and Islam, but it is notable that many Hindus claim Sikhism to be a variant of Hinduism, while no Muslims seem inclined to make this assertion.

** The poor were always with the hunter-gatherers as well, because they were all poor, but a wealthy leisured class who could comment on the plight of the poor did not exist so the observation would have been ludicrous.

Related: Historical Dynamics and contingent conditions of religion, Cliodynamics, the rise & fall of empires and asabiya.

Labels: , , ,

Avatar & the death of "Star Trek aliens"   posted by Razib @ 11/12/2009 02:52:00 PM

Since readers of this weblog tend toward nerdishness I'm assuming they're following the buzz around Avatar: The Movie. I only got interested in it last night trying to figure out the references in yesterday's South Park episode, Dances with Smurfs. Check out the tailer below. Obviously actors in regular films aren't going to be replaced by CGI in the next few years, but, looks like we're on the cusp of a the shift when it comes to a human being necessary to portray humanoid aliens. The "uncanny valley" is to some extent an upside in sci-fi, excluding the problems that will generate when it comes to the sticky issue of hybridization.


Wednesday, November 11, 2009

FOXP2 in Nature   posted by Razib @ 11/11/2009 02:08:00 PM

Human-specific transcriptional regulation of CNS development genes by FOXP2:
...It has been proposed that the amino acid composition in the human variant of FOXP2 has undergone accelerated evolution, and this two-amino-acid change occurred around the time of language emergence in humans...However, this remains controversial, and whether the acquisition of these amino acids in human FOXP2 has any functional consequence in human neurons remains untested. Here we demonstrate that these two human-specific amino acids alter FOXP2 function by conferring differential transcriptional regulation in vitro. We extend these observations in vivo to human and chimpanzee brain, and use network analysis to identify novel relationships among the differentially expressed genes. These data provide experimental support for the functional relevance of changes in FOXP2 that occur on the human lineage, highlighting specific pathways with direct consequences for human brain development and disease in the central nervous system (CNS). Because FOXP2 has an important role in speech and language in humans, the identified targets may have a critical function in the development and evolution of language circuitry in humans.

Ed Young has a long entry on this paper, along with context.

Update: Someone on Twitter is suggesting we create a transgenic. Which direction?

Labels: , ,

Tuesday, November 10, 2009

Gladwell hatin'   posted by Razib @ 11/10/2009 07:57:00 PM

Some red meat for readers, Malcolm Gladwell, Memes and Intellectual Honesty:
Gladwell comes across as a child trying to explain why his hand was in the cookie jar. He advances a series of unconvincing, somewhat contradictory explanations, hoping that we will ignore the larger problem. So far as I can tell from Google searching, this strategy has worked; people have noted that Gladwell is talking about memes but no one has called him out for his failure to acknowledge this prior work. This isn't acceptable. Gladwell's behavior is intellectually dishonest. His failure to credit Dawkins or others who have thought about these ideas before him does a disservice to those individuals and to honest intellectual discourse. I don't think Gladwell's behavior constitutes plagiarism, but it certainly would be punished if it occurred in an academic setting. Failure to cite prior work results in a paper being rejected from any legitimate journal. If a student hands in an assignment that fails to cite prior work, the student receives a bad grade, if not outright failure. Gladwell owes his readers and Richard Dawkins an apology for his failure to acknowledge that Gladwell's idea recycles Dawkins's earlier work.


Spengler does it again!   posted by Razib @ 11/10/2009 12:03:00 PM

Just Spengler (David Goldman) being Spengler, From "Zionism is Racism" to "Judaism is Racism":
Judaism has nothing to do with race-there are Jews of every race-but it does have to do with family. Jews are members of Abraham's family. Not only tradition, but a great deal of DNA evidence support this claim. To insist that Jews adopt the criterion of "belief" for membership is to rule that God must act in accordance with a human court's notion of the permissible range of God's behavior. No wonder the Reform Jews and the British Humanist Association support this.

1) Yes, Jews are genetically distinct.

2) But, they are also the product of genetic admixture.

3) And, it seems more likely that that admixture arrived via maternal lineages, that is, gentile female ancestors (the mtDNA results are somewhat confused, but the Y lineages seem to be relatively strongly Middle Eastern in provenance in comparison to total genome content).

In light of the fact that the debate is over the validity of the criterion of maternal descent as to "Who is a Jew," it seems deceptive to appeal to genetics when that field opens up more questions in regards to Jewish tradition than it closes. Of course, this sort of shell-game is normal behavior for Spengler. Someone should really put a "For Entertainment Purposes Only" sticker on his blog.

Labels: ,

Sunday, November 08, 2009

TCHH & curly hair in Europeans   posted by Razib @ 11/08/2009 10:31:00 PM

Common Variants in the Trichohyalin Gene Are Associated with Straight Hair in Europeans:
Hair morphology is highly differentiated between populations and among people of European ancestry. Whereas hair morphology in East Asian populations has been studied extensively, relatively little is known about the genetics of this trait in Europeans. We performed a genome-wide association scan for hair morphology (straight, wavy, curly) in three Australian samples of European descent. All three samples showed evidence of association implicating the Trichohyalin gene (TCHH), which is expressed in the developing inner root sheath of the hair follicle, and explaining ~6% of variance (p = 1.5 X 10-31). These variants are at their highest frequency in Northern Europeans, paralleling the distribution of the straight-hair EDAR variant in Asian populations.

This sort of stuff has obvious applications forensics.


The quest for common variants & cognition   posted by Razib @ 11/08/2009 11:39:00 AM

A genome-wide study of common SNPs and CNVs in cognitive performance in the CANTAB:
Psychiatric disorders such as schizophrenia are commonly accompanied by cognitive impairments that are treatment resistant and crucial to functional outcome. There has been great interest in studying cognitive measures as endophenotypes for psychiatric disorders, with the hope that their genetic basis will be clearer. To investigate this, we performed a genome-wide association study involving 11 cognitive phenotypes from the Cambridge Neuropsychological Test Automated Battery. We showed these measures to be heritable by comparing the correlation in 100 monozygotic and 100 dizygotic twin pairs. The full battery was tested in 750 subjects, and for spatial and verbal recognition memory, we investigated a further 500 individuals to search for smaller genetic effects. We were unable to find any genome-wide significant associations with either SNPs or common copy number variants. Nor could we formally replicate any polymorphism that has been previously associated with cognition, although we found a weak signal of lower than expected P-values for variants in a set of 10 candidate genes. We additionally investigated SNPs in genomic loci that have been shown to harbor rare variants that associate with neuropsychiatric disorders, to see if they showed any suggestion of association when considered as a separate set. Only NRXN1 showed evidence of significant association with cognition. These results suggest that common genetic variation does not strongly influence cognition in healthy subjects and that cognitive measures do not represent a more tractable genetic trait than clinical endpoints such as schizophrenia. We discuss a possible role for rare variation in cognitive genomics.

David Goldstein is one of the authors. I wonder if this influenced his views on the evolution of intelligence.

Labels: , ,

Abortion   posted by Razib @ 11/08/2009 01:22:00 AM

Over at Secular Right I report again that there is no sex difference in attitudes toward abortion in the general population (if anything, women are a bit more pro-life than men). But, to my surprise in the American Congress women do support abortion rights to a greater extent than men. This holds for both Republicans and Democrats.

Quick thoughts:

1) I used one abortion-rights organization's rating for 2007-2008. Would be useful to use pro-life and other abortion-rights organizations, as well as different years, to see if the difference is consistent. The sample size for women isn't very large in Congress.

2) Try to control for other variables.

Those who know more about the selection process for candidates for higher office might have some immediate insight as well.

Addendum: Poking through the GSS I did once find that among college-educated white liberals women placed a greater weight on their support for abortion-rights than men. I suspect that explains the tacit assumption among many liberals that women in particular care about abortion and support its legality to a greater extent than men, in their social set that is quite true. But, that segment of the population (college-educated white liberals) are only ~5% of the total American population.


Saturday, November 07, 2009

Applied Statistics over at ScienceBlogs   posted by Razib @ 11/07/2009 07:19:00 PM

Just a reminder, Andrew Gelman is now blogging at ScienceBlogs under "Applied Statistics".


Monday, November 02, 2009

Coffee or not   posted by Razib @ 11/02/2009 10:56:00 AM

Real vs Placebo Coffee. There's a real effect. Though interestingly those who secretly were given decaf didn't notice it in their self-reports.

Labels: ,

Inequality & wealth   posted by Razib @ 11/02/2009 10:27:00 AM

A review over at ScienceBlogs of a new paper, Intergenerational Wealth Transmission and the Dynamics of Inequality in Small-Scale Societies. I'm going to comment more in the near future, as I think this an give us insight into historical dynamics. An interesting find is that pastoralists and settled agriculturalists exhibit the same levels of heritability of material wealth (as well as the same values on material wealth). Hunter-gatherers and slash & burn agriculturalists seem to be at the other end of the spectrum.


Sunshine and SEC Football   posted by dkane @ 11/02/2009 07:35:00 AM

The cover story for Sports Illustrated two weeks ago described the dominance of the South Eastern Conference (SEC) in US college football.

"More players are invited to the NFL combine each year from the SEC than from any other conference," Ole Miss's [Head Coach] Nutt says when asked about the quality of the athletes who compete in the league. "The most players drafted just about every year going back 10 years come from the SEC." Indeed, dating to the 2000 NFL draft, the conference has had 400 players selected; the next-best league is the ACC, with 364.

It's no mystery to Nutt why an SEC team has won the BCS national championship each of the last three years (Florida in 2006 and 2008, LSU in 2007) and is favored to produce the champ again this season. "I watch [teams in] other conferences all the time and I think, Boy, I'd like to play them," Nutt says.

But, for GNXP readers, this is the most fun comment.

It all starts with recruiting. Nutt says that players from the South, particularly those who reside in Florida, become better college players than kids from other parts of the country, though he can't explain why. "Maybe it's the sunshine," he says. "In any given year an average of 335 young men [from Florida] sign with Division I schools. When I was coaching at Murray State [in Kentucky], I remember going to Florida and seeing, maybe, coaches from Wake Forest down there. But now? You've got Wisconsin, Minnesota, Purdue, Virginia, Virginia Tech. You've got schools from North Carolina. They're all down there, and they're coming for the speed. We signed nine from Florida this year. Nine!"

Yeah! It's the "sunshine" that causes the "quality of the athletes" in the SEC. Caste Football calculates that the percentage of white starters at SEC teams is 25%, lower than any other major conference.

Of course, from a GNXP point of view, sunshine may be actually have played a role, but not in the way that Nutt implies . . .

Sunday, November 01, 2009

William Gunn is looking for another job   posted by Razib @ 11/01/2009 09:49:00 PM

Early (2002) reader of this weblog, William Gunn, is leaving a biotech company in San Diego and is looking for another job. Here's his Linkedin.


How European is New England...not as much as I thought   posted by Razib @ 11/01/2009 01:53:00 PM

One hypothesis I have held is that there is a cultural gap in the United States whereby the West Coast and the Northeast are more "European" than the rest of the nation. So you have ideas crop up like Jesusland. I decided to see if I could compare European nations and American subregions using the WVS, Eurobarometer and GSS. I looked at two issues:

1) Belief in God
2) Nationalism

These two had equivalent questions in the 2000s for the WVS & GSS. I created an index of religiosity whereby:

Atheists & agnostic = 0
Higher Power = 1
Theist = 2

Very Proud of Country = 3
Somewhat Proud of Country = 2
Not Very Proud = 1
Not Proud at All = 0

On these two traits I don't see much of a Blue America + Europe clade....

Note: Limited GSS results to the 2000s.

Labels: ,