Friday, September 17, 2010

The evolution of Cavalli-Sforza. Part III

Sir Walter Bodmer in 1977. He wanted to shut down research on race and IQ but had no academic credibility in human genetics. Cavalli-Sforza supplied the missing credibility.

The mid-1950s saw Cavalli-Sforza shift from bacteria to human subjects, with his studies of genetic drift in the villages of Italy’s Parma valley. In the mid-1960s, his scope of interest broadened as he used blood-group data to trace the ancestry of human populations, thus showing how they had progressively broken up to form the ones we know today.

What were Cavalli-Sforza’s views on race during this period? He seems to have been silent. In this, he followed the example of other postwar geneticists who were all too aware of the cloud of suspicion that hung over their heads. Cultural anthropologists were mostly the ones who talked about race, and they downplayed its importance.

The silence was broken in the late 1960s by two academics: a physicist, William Shockley, and a psychologist, Arthur Jensen. In 1969, Jensen argued in the Harvard Educational Review that African-American children had lower IQs for genetic reasons and that efforts to close the IQ gap, like Head Start, were doomed to failure.

This issue was already making waves in academia when Cavalli-Sforza came to Stanford University for a trial year in 1968-69. In Stanford’s genetics department, a close friend and colleague, Joshua Lederberg, had written a letter attacking attempts to link race to IQ. One of its signatories was another colleague, Sir Walter Bodmer (Stone & Lurquin 2005, p. 98).

Bodmer had experienced racism as a child (1). He wanted to refute this new racism in a high-profile journal, with a view to shutting down research on race and IQ. Unfortunately, he had no credibility in the field of human genetics, as Cavalli-Sforza would later discover when the two of them began writing The Genetics of Human Populations:

W. Bodmer spent several months in Italy, and a year in the United States for this collaboration. He was much stronger than me in mathematics, but he did not know human genetics and had only worked with the genetics of bacteria and fungi.
(Cavalli-Sforza & Cavalli-Sforza 2008, p. 169)

Bodmer asked Cavalli-Sforza to co-author a manuscript on race and IQ to be published in Scientific American. It is doubtful that Cavalli-Sforza had much to offer on the subject, having never written on it before.

But it would have been harder for him to refuse Bodmer’s request. He had no tenure at Stanford and his only friends there were Bodmer and Lederberg. The latter, in particular, had helped Cavalli-Sforza rebuild his career during the difficult postwar years and had been instrumental in getting him a position at Stanford. There was thus an implicit exchange of services. In return for past and future favors, Cavalli-Sforza lent credibility to an article that might otherwise have attracted less attention or, perhaps, never been published. It certainly allowed Bodmer to write the following ‘expert opinion’:

As geneticists we can state with certainty that there is no a priori reason why genes affecting I.Q., which differ in the gene pools of blacks and whites, should be such that on the average whites have significantly more genes increasing I.Q. than blacks do.
(Bodmer & Cavalli-Sforza 1970, p. 28)

According to Bodmer, such a statistical difference would be unexpected for two reasons. First, it could not result from chance events (e.g., genetic drift, founder effects, etc.) because intelligence is a polygenic trait. The laws of chance would thus prevent the many different genes from having, on balance, more intelligence-boosting variants in one human population than in another. Second, natural selection could not have created the black-white IQ difference because black Americans have been in the United States for only two hundred years. This is far too recent for their IQ to have diverged from that of white Americans, even with strong differences in natural selection.

The first argument is wrong. Stature is a polygenic trait, yet it will differ significantly among random samples taken from a single population. Although many genes are involved in stature, some have a much stronger effect than others, with the result that variation at such gene loci is not drowned out by other genetic variation. The phenotypic variation is therefore noticeable.

The second argument is also wrong. It assumes that the black-white IQ difference is limited to the United States. Yet no one has ever made this assumption, other than Bodmer.

After being published in 1970, the article was incorporated the following year into a textbook by the same two authors, The Genetics of Human Populations. Both publications presented several counter-arguments to the idea that IQ varies with racial background:

1. Although IQ seems to be highly heritable, with estimates ranging from 40 to 80%, it doesn’t follow that the black-white difference in IQ is 40-80% genetic. The difference could be entirely due to environmental causes. Heritability studies are based on twins who share a common social environment. In contrast, black and white Americans inhabit very different social environments.

2. Because of their unusual in utero environment, twins may provide inaccurate estimates of heritability.

3. Black Americans reportedly have higher IQs when the testers are black American. Thus, cultural factors, including the design of the IQ test itself, might account for the black-white IQ difference.

4. Other contributing factors might include maternal malnutrition and/or a deficient home environment.

5. Even if the black-white IQ difference is proven to be mainly genetic, this knowledge has no practical applications in a free and democratic society. In contrast, a putative environmental cause does have practical applications (e.g., improvements to schooling and nutrition, elimination of barriers to economic advancement, breaking down of cultural barriers, etc.). Even if these applications fail to deliver their promised outcomes, the negative impacts will be minor.

As for Bodmer’s natural selection argument, it was quietly dropped from The Genetics of Human Populations.

The two authors nonetheless acknowledged the possibility that the black-white IQ difference could be genetic:

In summary, therefore, we do not exclude the possibility that there could be a genetic component to the mean difference in IQ between black and white Americans, but simply maintain that presently available data are inadequate to resolve this question in either direction. (Cavalli-Sforza & Bodmer 1971, p. 799)


This position was surprisingly moderate and already trending toward reactionary. The early 1970s saw the beginning of efforts to purge academia of ‘racist’ professors. After Jensen’s 1969 article, students and faculty staged large protests outside his U.C. Berkeley office. He was denied reprints by his publisher and not permitted to reply to letters of criticism. Similar harassment was directed at other academics, such as psychologist Richard Herrnstein and sociologist Edward Banfield.

As a concession to this antiracist movement, Cavalli-Sforza and Bodmer added the following caveat:

We are, of course, aware of the dangers of either overt or implicit political control over scientific inquiry. The suppression of Galileo and the success of Lysenko are two notorious examples of the evils of such control. Most scientists, however, do submit to certain controls over research on human beings such as, for example, the right of an individual to be experimented on, and the confidentiality of the information collected by the census bureau. These controls are imposed to protect the individual from possible direct detrimental effects of scientific investigations. The treatment of the Jews in Nazi concentration camps is a testimonial never to be forgotten to the needs for such controls. There can be no doubt that in the present racial climate of the United States, studies on racial differences in IQ, however well intentioned, could easily be misinterpreted as a form of racism and lead to unnecessary accentuation of racial tensions. Since we believe that no good case can, at present, be made for such studies on scientific or practical grounds, it follows naturally that we do not see the point in particularly encouraging the use of government or other funds for their support […] (Cavalli-Sforza & Bodmer 1977, pp. 801-802)

This reasoning is more than a bit disingenuous. The existing controls concerned only how research was to be carried out. Here, the two authors were arguing for controls on the aims of research, the why.

It is also a bit silly to suggest that the Holocaust happened because the victims weren’t given consent forms. Many deportees had in fact signed forms promising that they would be sent to labor camps. Evidently, such documents were not worth the paper they were printed on. They were a lie. But how do you fight a lie in a society that criminalizes the mere fact of saying what you think? Nazi Germany practiced too much control, not too little.

Cavalli-Sforza himself knew about life in a controlled society, particularly during his wartime research with Dr. Prigge.

Professor Prigge was in no way a Nazi, but of course we spoke about the government with much precaution, whereas in Italy the criticisms against fascism were frequent and overt. Among all the people we met in Germany, none had heard about the Shoah or the concentration camps. We learned about their existence, in Italy, only after the war.
(Cavalli-Sforza & Cavalli-Sforza 2008, p. 35)

Of course, bad things can also happen in a free and democratic society. And they can happen just as often. The difference, however, is that their worst effects can be curtailed—by protesting against them, by denouncing those who are responsible, or simply by pointing out their existence. During the last war, the American and Canadian governments interned people of Japanese origin on the west coast. This was an injustice and was denounced as such at the time. But it did not lead to mass murder. Elsewhere, similar internments did.

Ironically, by endorsing controls on research, Cavalli-Sforza may have been acting on fears he had earlier acquired in a less free world. Today Jensen, tomorrow … who knows? Who will be next? Perhaps someone warned him against sitting on the fence. And then there was his wartime past ... Yes, a fearful mind is the devil’s playground.

By co-authoring the 1970 Scientific American article, Cavalli-Sforza helped initiate a process with long-lasting consequences. As an expert on human genetics, and as someone less politicized than other academics, Cavalli-Sforza gave key support at a key time to the soft totalitarianism that would overrun much of academia. The following decades would see increasing control of the marketplace of ideas (2).

Notes

1. "Walter Bodmer was born in Germany, in the city of Frankfurt am Main which Cavalli had coincidentally visited during World War II. […] Sir Walter’s early infancy had been deeply disrupted by events unfolding in Nazi Germany. His father was a Jewish medical doctor (his mother was a Gentile) with aspirations to academia. However, years before the Nazis took power in Germany, during the period known as the Weimar Republic, Bodmer’s father had already been told that his hopes of becoming a university professor were futile, given his “racial” background. In 1938, threatened by the Nazi political regime, he left Germany under the pretext of taking a vacation. He went to England, where he was soon followed by his wife and young son, who was then only two and a half years old."
(Stone & Lurquin 2005, p. 79)

2. Bodmer is perceived as being one of Cavalli-Sforza’s close associates. He was certainly the closest one during the 1970s, when they jointly wrote two textbooks and a number of articles. Bodmer is given nine mentions in the index to Stone and Lurquin’s biography of Cavalli-Sforza, many of which are lengthy. Strangely enough, he receives only three mentions in the autobiography, two of which are single sentences. It is also odd that the autobiography says nothing about the 1970 Scientific American article, which was Cavalli-Sforza’s first high-profile publication.

References

Bodmer, W.F. and L.L. Cavalli-Sforza. (1970). Intelligence and race, Scientific American, 223(4), 19-29.

Cavalli-Sforza, L.L. and W.F. Bodmer. (1971). The Genetics of Human Populations, San Francisco: W.H. Freeman and Co.

Cavalli-Sforza, L.L. and F. Cavalli-Sforza (2008). La génétique des populations : histoire d'une découverte, Paris: Odile Jacob. (translation of Perché la scienza : L’aventura di un ricercatore).

Jensen, A.R. (1969). How much can we boost IQ and scholastic achievement? Harvard Educational Review, 39, 1-123.

Stone, L. and P.F. Lurquin. (2005). A Genetic and Cultural Odyssey. The Life and Work of L.Luca Cavalli-Sforza. New York: Columbia University Press.


9 comments:

Jason Malloy said...

"Sir Walter Bodmer in 1977. He wanted to shut down research on race and IQ but had no academic credibility in human genetics. Cavalli-Sforza supplied the missing credibility... Cavalli-Sforza lent credibility to an article that might otherwise have attracted less attention or, perhaps, never been published. It certainly allowed Bodmer to write the following ‘expert opinion' "

I find this a little stretched, though I accept Bodmer's greater influence. It's one thing to say that C-S was stronger at genetics and Bodmer at statistics, but Bodmer was a well credentialed expert at both; he earned his PhD in genetics under Ronald Fisher. I don't think there is a strong basis for the claim that Bodmer didn't have the credentials to write an 11 page editorial on race and genetics in Scientific American, or that the unknown C-S gave it much additional credibility. They became academic stars mostly as a team in the 1970s.

They were jointly and deliberately making faulty arguments about their field to discredit Jensen's inconveniently accurate summary of population genetics:

First, it could not result from chance events (e.g., genetic drift, founder effects, etc.) because intelligence is a polygenic trait. The laws of chance would thus prevent the many different genes from having, on balance, more intelligence-boosting variants in one human population than in another.

Worse than wrong. Especially given anti-hereditarian arguments by that other geneticist liar pair - Feldman and Lewontin-- around the same time period that the high heritability of intelligence argues that it was not a trait subject to much natural selection. But polygenic traits under weak selection vary in almost direct proportion to population distance and founder events.

P.S. You should add quotation marks or blockquote to footnote 1.

Peter Frost said...

Jason,

According to Cavalli's biography (Stone and Lurquin 2005), the Scientific American article originated in a letter published in Science, written by Joshua Lederberg, and signed by Walter Bodmer (among others). It contains many of the same arguments that appear in the Sci Am article, including the one that race and IQ research should be held to higher standards than other forms of research.

Bodmer was the senior author and, as such, probably wrote most of it (if not all). Cavalli-Sforza says nothing at all about it in his autobiography.

Bodmer was a geneticist at the time, but his work had been with bacteria, not humans. It is hard to imagine what special expertise he could have brought to the race/IQ debate. It's not just that the mode of gene transmission is very different in bacteria. Bacterial gene expression is not influenced by culture, history, or social organization. His only strong hand was in the application of statistics to genetics.

I've added quotation marks to the footnote, thanks.

Ben10 said...

Do you think CS read your blog?

You might check the IP geographical location of your visitors and if you know where CS is, then you'll know.

Anonymous said...

In 'Race' p. 61 John R. Baker cites the the 1928 publication of Pitirim A. Sorokin's Contemporary Sociological Theories as marking the close of the period when both sides of the controversy could be put forward. During the early twenties Walter Lippman was criticizing the US army tests, Lippmann had five articles attacking testing in the New Republic during 1922. The army apha and beta tests were discontinued after WW1.

I think WW2 was unimportant.

Anyway Eysenck was the main influence on Arthur Jensen and Eysenck said the media gave the misleading impression that his views were those of a maverick outside the mainstream scientific consensus. He cited 'The IQ Controversy, the Media and Public Policy' as showing that there was majority support for every single one of the main contentions he had put forward, further asserting that the idea there was any real debate about the matter among the relevant scientists was incorrect.

So it's only the public marketplace that is affected

Anonymous said...

'By co-authoring the 1970 Scientific American article, Cavalli-Sforza helped initiate a process with long-lasting consequences.'

Hmmm Lippmann was tearing into tests as early as 1922 with 5 articles in the the New Republic and the US army stopped testing after WW1. Sorokins 1928 chapter was probably the last time the issue could be put to the public . however according to 'The IQ controversy the media and public policy' scientific balance of opinion is that is that Jensen was right. They just don't say so openly.

Anonymous said...

I don't think the Nazis were terribly important in forming the views of Bodmer and like-minded anti racists. A previous generation had already joined battle for the delegitimization of the concepts of IQ or race and won that conflict well before anyone took the 'Bavarian Fascisti' seriously.

Although Bodmer could point to the Nazis but that was not really the root cause of views like his.

Eysenck was the biggest influence on Jensen and Eysenck's beloved grandmother died in a concentration camp.

Peter Frost said...

Ben10,

Cavalli-Sforza is based at Stanford, but I'm sure he spends most of his time off-campus.

Anon,

Attacking an idea through vigorous debate is one thing. Shutting down the debate is another. In this respect, the early 1970s were a turning point. It became impossible to talk about the race/IQ issue, except as a "myth."

Anon,

Bodmer talked about his father's experiences with Cavalli-Sforza and apparently with others. I don't think this was just idle chit-chat.

"Victimhood" is a subjective reality in the sense that it results not only from your own experiences but also from those of people you identify with, such as your father. I often meet people who consider themselves "victims" even though they've lived very comfortable lives.

Ben10 said...

About CS, what can you do, he must be 80 or more by now. Nobody produce new ideas at this age and this is a time in life where honors and power feel sweeter than real science. You are not into science at 80.
There was a very good book about Tycho Brahe the astronmer, 'Tycho and Kepler' from Kitty Ferguson. I encourage to read it. You realize that the academic world has not changed since 1601. Intrigues, lies, cheating, peer pressure, greed for money, recognition and political power rule.
Brahe was the supporter of the earth-centric system and all his life he battled for academic power and recognition. On his deathbead and despite all the evidences that Kepler brought to him, he refused to change his mind and asked Kepler to continue his work to proove that Earth was at the center of the universe! Once Brahe dead, Kepler just did the opposite and his theory could at last break the chains of the academic peer pressure.

There is an article on CNN about a Zedonk, a cross betweeen a donkey and a zebra. The zedonk might even be fertile. This is glorious proof that CS is right: there is no races. A donkey and a zebra are the same species as are Lions and Tigers, polar bears and grizzly bears, Coyotees and wolves, that can all interbreed.
No?

The hybrid shows limbs and ventral features that are zebra and donkey features dorsaly. According to the article his behavior is more zebra, in agreement with the law that when you cross two species, the older species's charaters prevail. The brain develops from the ectoderm and so it must be harwired like a zebra here, but to me, the donkey fur shows that donkey genes have prevailed in the neural crest (at least the rostral and dorsal neural crest) which also come from thre ectoderm and form much of the face and coat pattern. Any toughts about this apparent discrepancy?

http://www.cnn.com/2010/US/09/23/what.is.zedonk/index.html?hpt=Mid

elenama said...

Bodmer is perceived as being one of Cavalli-Sforza’s close associates. He was certainly the closest one during the 1970s, when they jointly wrote two textbooks and a number of articles. Bodmer is given nine mentions in the index to Stone and Lurquin’s biography of Cavalli-Sforza, many of which are lengthy. Strangely enough, he receives buy cialis only three mentions in the autobiography, two of which are single sentences. It is also odd that the autobiography says nothing about the 1970 Scientific American article, which was Cavalli-Sforza’s first high-profile publication.