Friday, August 27, 2010

A late convert

Cavalli-Sforza with Kistler Prize (2002). The paths of Glory lead …

Human Biology has interviewed Luigi Luca Cavalli-Sforza, probably the best-known human geneticist (Manni, 2010). His worry? The growing rift between anthropology and biology:

Unfortunately, at least in the United States, anthropology is in decline: The cultural moiety is separating from the rest and loves to declare that it is not science—what is it, instead? The rest of the U.S. anthropology departments (variable according to places, but mostly a mixture of physical anthropology, primatology, archaeology, linguistics) fortunately seem to resist this trend, but anthropology seems to be losing ground. There is a need for a new anthropology capable of keeping unity, and to develop human science curricula that may promote an original, fresh outlook.

Why this retreat from biology? Ask almost any anthropologist. You’ll be told that there is much more genetic variation within human populations than between them. With culture, we see the reverse: the differences are primarily between populations. So how can genetic differences explain cultural differences? Such explanations are inevitably wrong and, needless to say, dangerous.

And you’ll be told that this view is endorsed by all credible geneticists, including Cavalli-Sforza. Just read the same interview:

Don’t you feel that results pointing to intracontinental genetic differences can reinforce racist theories, as some pharmacogenomic studies recently did?

LLCS: The between-population genetic variation observed with 650,000 SNPs on the 52 populations of the HGDP is 11% (Li et al. 2008) with a very small standard error. It becomes 16% for the X chromosome, as is expected if nearly all the genetic variation is due to drift—that is, the role of natural selection is very limited. The ca. 30-year-old estimate by Lewontin (1972) of this quantity (15%) was based on other markers and populations and was a reason to encourage banning the use of the word race in humans. In any case the new value is even more supportive of dropping the word race. What we really need to ban is racism, and this is not a socially easy-to-do program. Charles Darwin, already, was skeptical about the usefulness of the race concept in humans, having noted that human variation is geographically almost continuous for most traits.

Man has been studied more carefully than any other animal, and yet there is the greatest possible diversity amongst capable judges whether he should be classed as a single species or race, or as two (Virey), as three (Jacquinot), as four (Kant), five (Blumenbach), six (Buffon), seven (Hunter), eight (Agassiz), eleven (Pickering), fifteen (Bory St. Vincent), sixteen (Desmoulins), twenty-two (Morton), sixty (Crawfurd), or as sixty-three, according to Burke. This diversity of judgment does not prove that the races ought not to be ranked as species, but it shows that they graduate into each other, and that it is hardly possible to discover clear distinctive characters between them.
(Darwin 1871: 226)

Well, no. Darwin was not “skeptical about the usefulness of the race concept in humans” In the above quote, he was denying that humans had diverged into different species. But he did not deny the reality of human races, as he stated a few pages previously:

There is, however, no doubt that the various races, when carefully compared and measured, differ much from each other,—as in the texture of the hair, the relative proportions of all parts of the body, the capacity of the lungs, the form and capacity of the skull, and even in the convolutions of the brain. But it would be an endless task to specify the numerous points of difference.

[…] If a naturalist, who had never before seen a Negro, Hottentot, Australian, or Mongolian, were to compare them, he would at once perceive that they differed in a multitude of characters, some of slight and some of considerable importance. On enquiry he would find that they were adapted to live under widely different climates, and that they differed somewhat in bodily constitution and mental disposition. If he were then told that hundreds of similar specimens could be brought from the same countries, he would assuredly declare that they were as good species as many to which he had been in the habit of affixing specific names.

But why bother quoting Darwin? It’s more fun to quote Cavalli-Sforza!

The differences that exist between the major racial groups are such that races could be called subspecies if we adopted for man a criterion suggested by Mayr (1963) for systematic zoology. Mayr’s criterion is that two or more groups become subspecies when 75 percent or more of all the individuals constituting the groups can be unequivocally classified as belonging to a particular group. As a matter of fact, when human races are defined fairly broadly, we could achieve a much lower error of classification than 25 percent, implying, according to Mayr, the existence of human subspecies. (Cavalli-Sforza & Bodmer, 1977)

Well, that was Cavalli-Sforza back in 1977. Perhaps he was still unaware of Richard Lewontin’s finding that genes vary much more within human populations than between them. Was not this “a reason to encourage banning the use of the word race in humans”?

Except that Lewontin had published that reason in 1972. Five years later, Cavalli-Sforza was apparently still unimpressed. And he continued to use the race concept much later, as in this journal article from the late 1980s:

The first split in the phylo-genetic tree separates Africans from non-Africans, and the second separates two major clusters, one corresponding to Caucasoids, East Asians, Arctic populations, and American natives, and the other to Southeast Asians, (mainland and insular), Pacific islanders, and New Guineans and Australians (Cavalli-Sforza et al., 1988).

Cavalli-Sforza did not convert to race denialism until the mid-1990s, with his opus The History and Geography of Human Genes.

What motivated his conversion? Had Lewontin’s 1972 paper been further bolstered by new data and arguments? Not really. In fact, the 1980s and 1990s saw growing evidence that the same genetic overlap that Lewontin found between human populations also occurs between many species that are nonetheless distinct in anatomy, physiology, and behavior (see previous post).

As one anthropologist told me: “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.”

References

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

Cavalli-Sforza, L.L., A. Piazza, P. Menozzi, and J. Mountain. (1988). Reconstruction of human evolution: Bringing together genetic, archaeological, and linguistic data. Proc. Natl. Acad. Sci. USA, 85, 6002-6006.

Cavalli-Sforza, L.L. and W.F. Bodmer. (1977). The Genetics of Human Populations, San Francisco: W.H. Freeman and Co.

Darwin, C. R. (1871). The Descent of Man, and Selection in Relation to Sex. London: John Murray.

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

Manni, F. (2010). Interview with Luigi Luca Cavalli-Sforza: Past research and directions for future investigations in human population genetics, Human Biology, 82, 245–266.

19 comments:

Anonymous said...

I think C-S has been publicly opposed to racialist research at least from the 70s. He was involved in campaigns against Arthur Jensen in those days.

Peter Frost said...

Anon,

In his 1977 textbook "The Genetics of Human Populations", Cavalli-Sforza argued against the existence of black/white differences in IQ, but he did so in a civil and non-dogmatic manner. He also kept an open mind, saying that this debate was legitimate and that further data were needed.

To my knowledge, and unlike Richard Lewontin, Cavalli-Sforza never supported efforts to silence 'racist' professors.

Finally, he was not a race denialist in the 1970s and 1980s. He accepted the existence of human races. In this, he was already different from many of his contemporaries.

Ben 10 said...

If Mayr's definition of subspecies doest not apply for human, then SF should say why.
If he prefers a genetic definition, then fine, give the quantitative formula, apply it to human and if humans do not belong to different subspecies, that's fine again. But if the same formula says that a Lion is actually a Tiger, or a Donkey a Zebra, and for this reason SF decides to use different formulas for different animal groups, he should say why.
Classification should be quantitative, testable and modelisable. Any disgression of a model is acdeptable but should be justified.

So far, the only definition I see is here:
http://www.goodrumj.com/RFaqHTML.html
And it says that humans major groups qualifies as subspecies.

Tod said...

If SEXUAL SELECTION AND HUMAN GEOGRAPHIC VARIATION is correct then most people are wrong about the origin of the variation called 'race', for Europe and Africa at least.

If sexual selection rather than adaptation is responsible for Europeans' and Black Africans' special characteristics does that then alter the validity of infra-specific categories for them ?

Anonymous said...


If sexual selection rather than adaptation is responsible for Europeans' and Black Africans' special characteristics does that then alter the validity of infra-specific categories for them ?


Why is it only one or the other?

Selection against dyslexia has operated for longer in groups that have had writing, and some groups have had longer histories of living in large scale civilizations. Eg, Aboriginal Australians have only had any exposure to it over the last 100 years or less.

Of course, I guess one could always claim that all of these are susceptible to sexual selection.

Anonymous said...

> as is expected if nearly all the genetic variation is due to drift

Gosh, what a shock to the system, 30 years or whatever after the (nearly-)neutral theory was first propounded.

Ben10 said...

"If sexual selection rather than adaptation.."

Point #1: Adaptation is not politically incorect because it is driven by the environnement, the sun, the weather, the food etc.
Sexual selection on the other hand is very politically incorrect because it is driven by behavior, and a behavior similar to racism. It could be argued that racialism is the modern evolution of sexual selection. So, if racialism, or worse, racism, is evil, then the people who practiced sexual selection, our remote fathers, were evil. And the product of this selection, us, is a product of racism and is evil too.

Point #2: those who consider 'sexual selection' as the most politically incorrect or 'evil' are also those who often champion the ideas of biodiversity and try to preserve animal species for this reason. For example, I never heard of a program to intentionally breed Siberian Tigers with Sumatra Tigers in order to homogenize the Tigers subspecies. That would go against the politically correct preservation of biodiversity and of all Tigers subspecies.

However, Sexual Selection vs adaptation vs drift is a technical problem related to evolutionary drive. In Taxonomy, I don't think that it matters either Tigers subspecies differentiated under sexual selection, drift or adaptation or all of them together. Tigers are still classified into different subspecies as Siberian, Bengal or Sumatra Tigers.

Point #3: are human groups diffrent subspecies or not? Can we settle the question once and for all ?

We had a discussion about that earlyer. 'Race' is not a scientifically defined term. some say that 'Race' equals subspecies, some say that it equals a group of population. However, I quote wikipedia:

'In zoology, the International Code of Zoological Nomenclature (4th edition, 1999) accepts only one rank below that of species, namely the rank of subspecies', there is no infraspecies and obviously no 'race'.
However, there doesn't seem to be more objective criterias than:'Members of one subspecies differ morphologically or by different coding sequences of a peptide from members of other subspecies of the species'.
The matting criterias are very moot. Tigers breed with Lions and some of the hybrids have been fertile. But taxonomist put Lions and Tigers in different species, not even subspecies.

There is a need for a quantitative scale based on measurable differences. Otherwise we are talking Art, not Science. I prefer this, you prefer that. The link I gave above seems to indicate that human groups could indeed represent sub-species. But it this scale officialy recognized ?
Can we settle this problem?

Insightful said...

This is the problem with the word race. In biology, the word race is synonymous with species. Everyone on the planet is the same species however, Homo sapien sapiens. It would be more appropriate to possibly say breed, but most people would probably be offended by being compared to dogs.

The truth is, the genetic difference between races are small and is the reason why we can all breed with each other. Note that humans and chimps cannot mix although we are like 98% similar. That should tell you something right there.

And these morphological differences that we do have aren't as cut and dry as people want to believe. It is not like you cross the border out of Europe and all of a sudden people are brown skin. These are gradual changes as you move across the world. People are obviously black in Africa, as you move north there is a gradual lightening in skin tone. Epicanthic eyefolds are not a solely East Asian trait. They are also found among certain sudanese tribes and the Khoisan

A lot of the science though coming out suggests that these traits do not take nearly as long to develop as once thought. Afterall, it has only taken a few generations for European immigrants to come to the US, move to south Florida and now the grandchildren of those immigrants can often be very well tanned in skin tone. Other traits will take longer to develop, but possibly as short as a few thousand years I doubt white Americans will resemble white Europeans anymore...

Peter Frost said...

Ben10,

Terms like 'population', 'race', and 'subspecies' are just words. It's like the debate over whether we should call ourselves Homo sapiens sapiens or just Homo sapiens. It's ultimately a judgement call and there no clear reason why we should choose one or the other.

Tod and Anon,

Sexual selection is a form of adaptation. There are many populations and even species that have diverged from each other primarily through sexual selection.

Insightful,

Many species can produce fertile hybrids. Hybrid sterility is no longer a criteron of speciation.

In any case, you seem to be assuming that hybrid sterility correlates with functional difference. If two populations can hybridize, they must be really similar. If they cannot hybridize, they must be really different.

Wrong on both counts. Wrong, wrong, and wrong.

Hybrid sterility is caused by accumulation of minor mutations at many points on the genome. Often, these mutations have little selective value in and of themselves. Large, functional differences are due to selection acting on a small fraction of the genome, typically regulator genes.

"And these morphological differences that we do have aren't as cut and dry as people want to believe"

What people? Nobody I know. What sort of people do you hang around with?

"it has only taken a few generations for European immigrants to come to the US, move to south Florida and now the grandchildren of those immigrants can often be very well tanned in skin tone."

Uh, tanning is due to sun exposure. I would be more tanned if I move to Florida. You're confusing genotype and phenotype.

Anonymous said...

"The between-population genetic variation observed with 650,000 SNPs on the 52 populations of the HGDP is 11% (Li et al. 2008) with a very small standard error. It becomes 16% for the X chromosome"

As my interest is most focused on female development in populations (which seems so often overlooked-- with the exception of this blog, thank you!), I was curious as to the significance of the genetic variation among women. I notice that the percentage variation is higher for the X chromosome than for the Y. Is it by an amount that is statistically significant? Does it signify that genetic variance is greater between females of different populations than between males?

Just curious as a layperson who has begun digging into this.

Any feedback would be appreciated.

-- Pamela

Insightful said...

Pamela, the reason why the female line has more genetic variation than male line is because females have often been the ones to move around and introduce new 'blood' into tribes, whereas males have more often stayed put. The famous saying, "I want to marry you and take you home" wasn't said for nothing. There is meaning to that.

Add to that the fact that the more powerful men in tribes had more than one wife while other men may not have had any and you create a situation where the mitochondrial dna of humans is a lot deeper than the male Y chromosome.

So scientific Adam (the founding father of all of us) is a lot more recent than scientific Eve. In fact, by analyzing the Y chromosome DNA from males in all regions of the world, geneticist Spencer Wells has concluded that all humans alive today are patrilineally descended from a single man who lived in Africa around 60-70,000 years ago. On the other hand Eve lived 200,000 years ago or 130-140,000 years earlier than Adam. That is a lot more time to accrue variation..

Ben10 said...

@ Insightfull
'The truth is, the genetic difference between races are small and ' 'race' is not defined in Taxonomy.
'Note that humans and chimps cannot mix although we are like 98%similar' Note that Tigers and Lions can mix together althought they are recognized distinctive species. Some of their offspring have been fertile. Interbreeding also occur easily between subspecies of bear (grizzly/polar), canidae (grey wolf/coyotee) etc.
'It is not like you cross the border out of Europe and all of a sudden people are brown skin. These are gradual changes '
FINE, but it doesn't matter. A scale can be arbitrary as long as it matches reasonably with previous classification. Temperature scales are also arbitrary but are noneless quantitative/objective mesurement of a defined scientific concept.
@Peter
"Terms like 'population', 'race', and 'subspecies' are just words"
I am disapointed. Some biologist 'feel' it's cold while others 'feel' it's warm, but we still don't know the exact temperature, right?. I don't say it's your fault. As shown here:
http://en.wikipedia.org/wiki/Species_concept#Definitions_of_species
There are no less than 14 definitions of species!!
Something that puzzled me is the famous case of the Coelacanth/Latimeria. Coelacanth is the name of the fossil thought to be extinct at the end of the cretaceous ~65 millions years ago, while Latimeria is the name of the living 'modern' form. Are they the same species ?
Based on morphology a taxonomist can claim that yes, they are the same species and Latimeria is a living fossil. A geneticist could argue otherwise. There is no reason to imagine that Coelacanth could have a mutation rate much lower from similar groups. Therefore, between the early Coelacanth that appeared 350 millions years ago and the modern Latimaria, there is very likely a great amount of nucleotide divergence. Possibly enough to prevent any inbreeding between the old form and the modern form. The geneticist will give you tons of divergence of SNP and based on this high nucleotide divergence and incapacity to breed he will decide Coelacanth and Latimeria are different species.
Everything makes MORE sense if you consider gene activity regulated inside an interconnected network instead of sequence divergence. This idea proposed by Stuart Kauffman is powerfull. Kauffman showed that, depending on the interconnectivity of the networks (such as number of connection per genes and type of connection, like activation, repression, co-activation etc) the connectivity of the network will be resistant to change, not resitant to mutations, but these mutations don't change the overall connections and activity of the genes. Thermodynamic analogues of temperature and therefore entropy can be derived form these gene networks. A coelacanth could be represented by a very cold, almost frozen network, even with very divergent DNA sequence. Every species could be represented by such a gene network and the definition of a species is not dependant only on the sequence divergence or the capacity to breed.

Tod said...

Off topic

Does Zahavi's handicap principle have any validity in relation to humans ?

Anonymous said...

"Selection against dyslexia has operated for longer in groups that have had writing, and some groups have had longer histories of living in large scale civilizations. "

Proof of this?

Most people, prior to relatively modern times, were largely or entirely illiterate.

Peter Frost said...

Anon, Insightful,

Within any one population, there is generally more genetic variation among women than among men. There are two explanations:

1. The male gene pool is drawn from a relatively smaller number of polygynous men, i.e., more men than women don't leave progeny, so their genetic variation is lost.

2. Many human societies are patrifocal, i.e., brides move to where their future husband lives, not vice versa. In any one community, women tend to be thus drawn from a larger geographic area and are genetically more diverse.

See:

Hammer, M.F., Mendez, F.L., Cox, M.P., Woerner, A.E., & Wall, J.D. (2008). Sex-biased evolutionary forces shape genomic patterns of human diversity. PLoS Genet, 4(9), e1000202. doi:10.1371/journal.pgen.1000202

Keinan, A., Mullikin, J.C., Patterson, N., & Reich, D. (2008). Accelerated genetic drift on chromosome X during the human dispersal out of Africa. Nature Genetics, early view December 2008; doi:10.1038/ng.303

"I am disapointed. Some biologist 'feel' it's cold while others 'feel' it's warm, but we still don't know the exact temperature, right?"

Ben 10,

The problem is that organisms can differ from each other on a continuous scale. So the degree of difference does not bunch up into nice clusters that we can call 'race', 'species', 'subspecies', 'genus', etc. These terms are discontinuous constructs that we try to impose on a continuous reality.

This is a general problem that is not limited to evolutionary theory. When are people 'young'? When are they 'middle-aged'? When are they 'old'? The Quebec government has recently decided that people are 'young' until they are 35 years old. Not so long ago, a 35 year old was considered middle-aged.

I agree with your other points. Real, functional genetic differences should not be measured by differences in nucleotide sequences, since most of these sequences are of low selective value and many are "junk DNA". The Coelacanth is essentially the same animal as it was hundreds of millions of years ago, but a lot has probably changed in its junk DNA.

Anon,

"Most people, prior to relatively modern times, were largely or entirely illiterate."

This is a half-truth. Estimates of illiteracy are usually based on one's ability to write one's own name. Such 'illiterate' people, however, could usually read short sequences of block writing, such as one finds on storefront signs or graffiti. As is so often the case, literacy was not an either-or question.

Anonymous said...

"
This is a half-truth. Estimates of illiteracy are usually based on one's ability to write one's own name. Such 'illiterate' people, however, could usually read short sequences of block writing, such as one finds on storefront signs or graffiti. As is so often the case, literacy was not an either-or question. "

Then a better comparison should be degrees of literacy. I've yet to see a source on that statement, and in face of what you and I have said, seem doubtful in face how much less literate people were in the past.

n/a said...

Pamela,

"I was curious as to the significance of the genetic variation among women. I notice that the percentage variation is higher for the X chromosome than for the Y."

The 11% number is for autosomal (not Y) DNA. I would guess the higher between-population variation on the X relative to the autosome is due to the smaller effective population size for and higher levels of selection on the X.

"Does it signify that genetic variance is greater between females of different populations than between males?"

No. The Y is the male-determining sex chromosome, but the X is not a "female" chromosome. It's present and expressed in men. If anything, since men have only one X chromosome, we might expect greater variance between males of different populations (on the basis of greater X chromosomal between population variation). Also, the X chromosome spends more time in women, but is subject to greater selection in men.

Chris Crawford said...

Might I suggest a simple solution to this problem: replace the notion of race or subspecies with the notion of intraspecies deviation. The big difference here is that "race" and "subspecies" are both boolean notions: they're black and white. Either you are or you are not a member of this or that race or subspecies. I propose replacing this with a numerical evaluation of how far an individual or group of individuals deviates from the average for the species. Even better would be to build an N-dimensional array of N genetic markers, with N being as large as you please, and then carry out multi-dimensional analysis (MDA) to identify whatever groupings you are interested in. My guess is that you'll find so many different groupings, some quite tight, that you'll have to abandon the simple notion of race and replace it with a vastly more complex notion -- which is vastly closer to reality.

Anonymous said...

The whole whine against "race denial" is BS.

Is not that anyone denies the unsurprising different genetic frequencies that exist in such a pattern that allow us to draw rough some sort of mosaic map with more or less fuzzy borders. That what self-labeled race "realists" trumpet as the evidence of "race".

The problem is that from such state of things does not follow everything that everyone ever thought about "race" is real, or that such different genetic frequencies are a more parsimonious explanation for any difference we see between those populations. The real picture of human genetic diversity does not even match significantly with the more common notions of race, such as "one drop rule" blackness, and even many, if not most attempts at taxonomy by scholars.

So, whenever something really generalizes to a whole population, it's more precise to refer to that population, rather than vague "races" with fuzzy lines of uncertain placement. It's probably that more likely real science will deal not even with generalized population, but with specific traits, without the BS of typological thinking.