Thursday, January 21, 2010

Cavalli-Sforza - The price of collaboration?

In a previous post, I noted certain discrepancies between Luca Cavalli-Sforza’s current stand on the race concept and his earlier one. This reversal seems to have occurred between his 1976 book Genetics, Evolution, and Man and his 1994 opus The History and Geography of Human Genes (whose cover map is curiously at odds with his statement that human genetic variation does not cluster into racial groups).

Such a change of heart is all the more puzzling because the case against the race concept had already been made in 1972, when Richard Lewontin showed that genetic differences within human races greatly exceed genetic differences between human races. That was—and still is—the main argument for race denialism. If this argument failed to convince Cavalli-Sforza in 1976, what happened to make it more convincing in 1994?

This is all the more puzzling because several authors since 1972 have challenged Lewontin’s argument. Mitton (1977) and (1978) showed that within-race variation exceeds between-race variation only if one gene is examined at a time. The pattern reverses if several genes are examined at the same time. Another flaw in Lewontin's argument is that he used genetic data from genes that code for enzymes, blood groups, and various building blocks of human tissue. Yet these ‘structural’ genes appear to have been marginal to human evolutionary change. As Stephen J. Gould (1977, p. 406) noted:

The most important event in evolutionary biology during the past decade has been the development of electrophoretic techniques for the routine measurement of genetic variation in natural populations. Yet this imposing edifice of new data and interpretation rests upon the shaky foundation of its concentration on structural genes alone. (faute de mieux, to be sure; it is notoriously difficult to measure differences in genes that vary only in the timing and amount of their products in ontogeny, while genes that code for stable proteins are easily assessed).

Indeed, if we look at differences in structural genes we see a high degree of overlap not only between human populations but also between morphologically distinct species (see previous post).

So why did Cavalli-Sforza change his mind? A cynical answer was provided to me by one anthropologist: “I don't think our perception of the general patterns of genetic variation changed much from '76 to '94, but the intellectual climate that geneticists operate in sure did.”

Some light has been shed on this question by
Sesardic (2010). Aside from being an excellent rebuttal of race denialism, this paper also quotes an unpublished manuscript by one of Cavalli-Sforza’s collaborators A.W.F. Edwards. The manuscript describes the following episode:

When in the 1960s I started working on the problem of reconstructing the course of human evolution from data on the frequencies of blood-group genes my colleague Luca Cavalli-Sforza and I sometimes unconsciously used the word ‘race’ interchangeably with ‘population’ in our publications. In one popular account, I wrote naturally of ‘the present races of man’. Quite recently I quoted the passage in an Italian publication, so it needed translating. Sensitive to the modern misgivings over the use of the word ‘race’, Cavalli-Sforza suggested I change it to ‘population’. At first I was reluctant to do so on the grounds that quotations should be accurate and not altered to meet contemporary sensibilities. But he pointed out that, as the original author, I was the only person who could possibly object.


Bodmer, W.F. & L.L. Cavalli-Sforza. (1976). Genetics, Evolution, and Man. WH Freeman and Company, San Francisco. pp 563-572.

Cavalli-Sforza, L.L., Menozzi, P. & Piazza, A. (1994). The History and Geography of Human Genes. Princeton: Princeton University Press.

Gould, S.J. (1977). Ontogeny and Phylogeny, Belknap Press: Cambridge (Mass.).

Lewontin, R.C. (1972). The apportionment of human diversity. Evolutionary Biology, 6, 381-398.

Mitton, J.B. (1977). Genetic differentiation of races of man as judged by single-locus and multilocus analyses, American Naturalist, 111, 203-212.

Mitton, J.B. (1978). Measurement of differentiation: reply to Lewontin, Powell, and Taylor, American Naturalist, 112, 1142-1144.

Sesardic, N. (2010). Race: a social destruction of a biological concept, Biol. Philos. early view.


Eugene said...

Incredibly, I've also been recently fascinated by subject of race and what it connotes. This is an explosive subject and the general public will instantly shy away from discussing it and its implications.

Very interesting is the question of why race-deniers exist in the first place. A huge group of people, perhaps the vast majority, believe that "we are all the same," but what are their motives for pushing this viewpoint? What is their agenda? I haven't been able to answer these questions. I can only guess that race-deniers vs. race realists, similar to gender-deniers vs. gender realists, are in fact "born not made," and no amount of persuasion will work.

Anonymous said...

One possible reason for Cavalli-Sforza venturing to endorse the biological concept of race in '76 is that, although a politically incorrect concept, racial genetic differences were not thought to have significant effects.

By the early Seventies it was really believed that the social environment was the cause of differences in personality and IQ.

It gradually became ever clearer that genes do more than produce skin deep variations in physique. In that light saying that races do in fact cluster together in their genetic make up is now understood to be a statement about non trivial human differences.

Tod said...

(I should have made it clear in the above 'Anonymous' comment.)

Those like Cavalli-Sforza who worked in the field always knew the hereditarians were right. The anti-hereditarian cultural establishment were the ones who were really taken in - for a while.

Ben10 said...

As much as the human genome has evolved during the last 50 000 years, it still cannot be that much. We are still within 1% genetic distance with the chimp and possibly within 0.1% with an erectus or a Neanderthal and I bet that in 100 000 years, we will still be in that range.
It's actually unfortunate that none of the early humans are still alive, so that Lewontin could have included them in his experiments and conclude with ambarassment that his techniques didn't have the accuracy to resolve an early erectus from a late neanderthal or a modern sapiens. But anyway, mesuring genetic change to define subspecies is partly irrelevant within homo Sapiens, where Sapiens is now more important than homo. An infime genetic change, like a basepair mutation without even the slightest impact on any protein structure, could still have a tremandous impact on cognitive capabilities. Therefore we have reached the realm of infinitesimal genetic causes with immense cognitive effects and mesuring the causes rather than the effects doesn't make more sense than pretending the LHC at CERN and a case made of cow poop are at the same complexity level because they are at 0.1%genetic distance from each other.

Peter Frost said...


The motive began as a high-minded one, i.e., a belief in universal brotherhood and a desire to build a better world without ethnic conflict. Unfortunately, many groups and individuals are now using race denialism as an instrument to serve other ends.


You must be very young ;) Actually, people were already arguing about race and IQ in the early 1970s. In those days, Jensen and Eysenck were the bêtes noirs of this debate. It began with a paper that Jensen published in the Harvard Educational Review in 1969. Phil Rushton got involved later (early 1980s?).

I met Cavalli-Sforza once. He struck me as a rather shy, unassuming man who just wants to be left alone.


In the past, geneticists estimated that chimp and human genomes were 1-2% different. This figure is now known to be an underestimate (it ignores higher-level variation due to nucleotide insertion-deletion). In any case, it's a meaningless figure, since nucleotide sequences vary considerably in their functional significance.

Humans and chimps are very similar if we look at 'structural' genes that code for the proteins of body tissue. The differences seem to be in the regulatory genes that control the development of these different tissues.

Anonymous said...

Don't stop posting such stories. I love to read blogs like this. BTW add more pics :)

Ben10 said...

Peter said "...nucleotide sequences vary considerably in their functional significance".
Indeed, 'functional significance' has another name: it's 'Phenotype' and that's exactly why naturalists who study phenotypes have the most sensitive way of defining species where molecular biologits miss most of it. For example, given the proteins and some DNA samples of the Gallapagos finches that Darwin studied, Lewontin most likely, might not have been able to recognize different species while Darwin easily did.
But why did SC changed his mind? you said it yourself Peter. Don't be too hard on him, a grant is a grant and life is hard.
I have experienced that every time you want to discuss a point in an 'established truth', even said in the softest terms and with the most precautions, you'll get the harshest answers and you'll pay a disproportionate price for it.
There is no way around guts and courage and intelligence alone cannot replace it, but it's exhausting.
To Eugene, instead of asking 'why' race deniers exist, you should ask 'who' are the race deniers, and you'll get your answer.

Tod said...

you should ask 'who' are the race deniers, and you'll get your answer

I disagree, there is a certain moralising attitude of mind that goes with denial and it was present way back when the field of anthropology and sociology were defined. It all started with Lord Kames' Sketches of the History of Man (1776).

'Strictures on Lord Kaim's discourse on the original diversity of mankind' was written by Samuel Stanhope Smith who thought the 'state of society' an important factor.
"An example he provides involves "the blacks in the southern states." Smith noted that field slaves had darker skin pigmentation and other "African" features than did domestic slaves, and hypothesized that exposure to white, European culture through their "civilized" masters had changed their anatomy as well".

Whether or not those who deny that human genetic variation clusters into racial groups really believe what they say, one thing is certain: race deniers are totally convinced they're morally in the right.

Cavalli-Sforza could know it was true while thinking it was a bad thing for people to believe. Still he hasn't falsified the data so isn't it just a matter of abandoning outdated terminology?

Eugene said...

I've recently tried to discuss the issue of race denialism vs. race realism on the forum of Richard Dawkins (the famous evolutionist and author of "The God Delusion" and other notable science/atheism books).

His forum,, is billed as "the free-thinking oasis" when it comes to issues of science and evolution.

Nevertheless, this particular topic proved to be "too much" for the moderators to handle. Any discussion of concepts such as genetic distances between races, race-related traits and behaviors, and the recent study about the effect of skin color on the human amygdala response (a very interesting study, I recommend it if you haven't read it yet: got banned, with a stern warning from the moderators that any further discussion of race would be instantly blocked.

This is a bizarre development for a community which claims to be science-oriented. The great irony is that a lot of left-leaning, liberal people who pay lip service to evolution and science, are in fact loathe to discuss the *implications* of evolution and science, which are sometimes politically incorrect and unpalatable, and refuse to acknowledge them.

Xenophon Hendrix said...

Eugene, most persons--even those who claim to be free thinking--have thought taboos that they either can't face or pretend that they can't face. This includes scientists and those who claim to be ruled by science. It is sad, but that's the way it is.

Off hand, I can think of only a handful of public figures who are or were willing to openly consider just about anything. Most of those are or were science fiction writers.

Peter Frost said...


You're right! My blog needs more pics. I'll try to fit one into every other post from now on.

Ben10 and Tod,

I would have no problem if Cavalli-Sforza simply said that we should use the term 'population' instead of 'race' because the latter term is often misunderstood or misused. Unfortunately, he has gone well beyond that sort of clarification. In fact, he is now widely cited by race denialists, even though his actual findings run counter to their position.


Did Richard Dawkins himself express an opinion on this topic?


In the early 1980s it was still normal to believe that all subjects could -- and should -- be resolved by free debate. I'm frankly worried by the soft totalitarianism that is now developing. And it's developing because leading intellectuals, like Cavalli-Sforza and Dawkins, have decided to go along with it.

Ben10 said...

I won't discuss the race deniers of the past, but in our present times, the reasons for denying races have nothing to do with biology. Allow me to do so, in my opinion, it now goes down to this: there is one big advantage to promote the idea of a single human race displaying small fluctuations around an homogenous racial type. It is that ethnic intermixing is almost justified, if not dutyified. A 'globalized' Man of mixed ethnicity fits better the Global World Market than a Man deeply rooted in his culture, religion and land.
Even if those who promote such a global world do actually believe that races exist, it is in their best interest to promote the opposite idea.
But, thanks to a multitude of internet blogs, every curious person can connect the dots in a variety of domains.
Like the political dots: for example, Jacques Attali, former adviser of the left wing president Mitterand, is now adviser of the right wing (understand 'liberal') prezydent Sarkozy. His advice is simple: introduce 60 millions new immigrants in europe in the next 30 years to sustain the european economy. Since humans are all the same race, Attali can call a 'racist' anyone who criticize his views. Another example, the director of the FMI, Dominique Strauss Kahn, suposedly a socialist but in effect a liberal economist, is in perfect agreement with Attali. The 'socialist' stance of DSK is just here to run for the 'french' presidency, and he has very good chances to get it with the help of the heavily controled 'french' televysion. Obviously, the people of this political elite abhore the slightest hint of ethnic awareness because it is at the root of nationalism. To fight, they recuscitate the Abominable Beast of WWII. To seduce, they use humanitarian and political freedom rhetoric and of course they tend to favorise the scientific views that support them.
There is therefore no surprise that in a putative televisual debate, fairly composed of 8 supporters of Lewontin (The Goods, including Tiger Wood, always looking for women loaded with more VitamineD), 1 'neutral' (let's say a researcher waiting for his funding grant decision, still pending) and a researcher supporting the opposite view (the bad Bad), the debate goes always the same way. First a rappel of the humanitarian and democratic values, then an impartial examination of the 'scientific' facts, in total disagreement with the 'irrational views' of the dissident researcher, and then the first traps come. Usually the IQ issue among ethnicity, or the issues touching the feminist lobby will be mentioned and that will be enough to destroy the dissident, but let's say he astuciously avoids the traps, then from the race issue comes the unavoidable eugenism issue, and therefore the Abominable Beast is recalled and that's the end for the kamikaze researcher. That's the Godwin's law:'s_law

Eugene said...

Peter -

To my knowledge Richard Dawkins hasn't explicitly addressed this issue himself, but based on what I've read and watched (I used to follow his work closely) his views are generally egalitarian and optimistic. On several occasions he mentioned that "we have moved beyond our evolutionary past."

To me, advocating evolution as passionately as Dawkins does, while in the same breath considering it as no longer relevant today, is a bit of a travesty. I actually call this "the liberal paradox" -- paying lip service to the mere fact of evolution while at the same time dismissing it as unimportant and no longer relevant (by espousing feminist or utopian views, for example). A lot of liberal, left-leaning people fall into that trap.

By the way, although Dawkins' primary interest is debating creationists (he's focused on the religion/atheism debate), here, too, the race question enters the picture. Intriguingly, experience shows religiosity affects some races more than others. For example, I've met plenty of white and Asian atheists, but never any African-American or Hispanic atheists. Why is this? Dawkins could have delved into this question in at least one of his 15 books devoted to religion, and yet he hasn't, which is also a sign of how he feels on the subject.

My impression is there are two camps of evolutionists--the idealists and the realists, and my fear is that Dawkins, for all his scientific stature, belongs to the former group.

Ben10 said...

Eugene, I've read the "selfish gene" without being convinced. I am an evolutionist in the sense that we can discuss and study the mechanisms of evolution but it is irrelevant to speculate its divine purpose, if there is one.
You cannot put the Presence or the Absence of God as a piece of the mechanisms of evolution.
To me however, Creationists's argument that Darwin theory can be formulated as "Survival of the survivyest" is correct, and this applies to the selfish gene as well.
I now find the concepts of Information Theory deeper and more usefull to understand Evolution. A serious creationist like Behe is using information theory to restate old beliefs in modern words, that there are levels of 'irreductible complexity' which forbids 'step-by darwinian-step' evolution. Even if this was true he should not have concluded that an 'intelligent cause' is at the origin of the biological complexity (to fill these irreductible gaps), but that 'a source of information has filled the apparently irreductible gaps'. The source doesn't have to be intelligent. These considerations anyway are very far from the concept of a selfish gene or the survival of the best survivor, but they still don't eliminate, nor create, the need for God.
At least I have to agree with Dawkins that "we should move beyond our evolutionary past" until the issue of races becomes somehow the future. I understand this sort of idealism. But in practice, we are not there yet and right now, the races-are-irrelevant or the no-God attitude are not usefull for anything except serving the interests of the hyperclass who is, indeed, very selfish.

Peter Frost said...


It's the same problem here in Canada. Both the 'Right' and the 'Left' have embraced the same globalist business agenda. Very often, the same business interests will donate money to both of the main parties. Partisan differences are increasingly differences of personality, image, etc.


"Dawkins could have delved into this question in at least one of his 15 books devoted to religion, and yet he hasn't, which is also a sign of how he feels on the subject."

If memory serves me right, he did write an article against race denialism (unless I'm confusing him with another person). The title, I think, was: "Races don't exist. Really?" This was six years ago. I'll see if I can track it down.

Tod said...

Cavalli-Sforza and Dawkins became well known and successful, then they changed their tune.

Easier for a camel to go through the eye of a needle...

James said...

"Mitton (1977) and (1978) showed that within-race variation exceeds between-race variation only if one gene is examined at a time."


Do you know how he was measuring variation? (my library doesn't subscribe to this journal, so I can't access it) If it were a measure based on heterozygosity (e.g. Fst/Gst), then surely within-race variation would have become an even higher proportion.

Many thanks,


Anonymous said...

Nice fill someone in on and this post helped me alot in my college assignement. Gratefulness you for your information.