Re: Genetic Evolution (was: HGHG)

Tom Lathrop (
Sun, 29 Jan 1995 19:06:17 GMT

This post contains discussion on some points extracted from a longer
article I posted around a week and a half ago, under the title "The
History and Geography of Human Genes". I think I have made a rather
strong case that recent genetic studies of the human race, in
particular those summarized in the recent book The History and
Geography of Human Genes, by Luca Cavalli-Sforza (and two others), do
*not* render the common idea of "race" meaningless, despite the
insistence of many on the Left that this is so. Readers who agree with
me on this should read this article carefully and use it as a resource,
especially to counter the argument that racial differences are
"superficial" and therefore unimportant.

In article <3fosb4$>,
Lane Singer <> wrote:
>In <> (Tom Lathrop) writes:
>>I never denied that there were flows of genetic diversity over
>>different populations. But - our definition of "race", from a
>>scientific perspective, bore the requirement that one population
>>contain at least one genetic characteristic that was absent in all
>>other populations. And this =doesn't= occur among humans, anywhere.

>>Hold it, hold it! Exactly *where* did you get that definition of
>>"race"? Given this definition there certainly are no distinct human
>>races, but this definition seems pretty far from what the
>>man-in-the-street thinks of as "race", and it seems like a pretty
>>useless definition in any case. Given this definition, if there *were*
>>two distinct races of humans or some other species, and even a slight
>>amount of mixing between them occurred, then instantly there would no
>>longer be two races! Your definition of race is far too unstable to be
>>interesting scientifically. Of course, I can understand why it would
>>be *very* interesting politically... :-)

>The sources on this are many, but since we're talking about this
>one book, I'll use a quote contained therein:
>HGHG, p.4
>"For a long time anthropologists tried to reconstruct evolutionary
>relationships and history ->on the basis of a single characteristic
>or gene<-. A favorite ->for over 100 years<- was the cephalic index
>(the percentage of skull breadth to length) introduced shortly before
>the middle of the last century. However, with a single trait, two
>populations of different origin could well turn out to be more or
>less identical. Anthropometric traits of this kind aslo have
>another very serious drawback: there is no guarantee that the
>character is completely ->under the control of biological inheritance<-
>and the the variations observed could be due to ->short term
>response to environmental changes<-. This was shown by Boas (1940)
>at the beginning of the centruy, ->but this lesson was, and still
>is, usually forgotten<-."
>Any politics involved in the study of human genetic evolution was
>imposed by your side, by the racists, who scrambled then, and do
>still now, to prop up their unscientific position.

Lane, you claimed that "our definition of 'race'" required that
populations have unique traits. When I asked you to justify this, you
talk about the methods people were using 100 years ago to try to trace
the origins of different populations. Non-sequitur Lane! *My*
definition of race does not require such unique traits. I'd like you
to show me a modern definition that does!

And even if you can, that doesn't take care of my other objections.
One, that such a definition is different from, and much more
restrictive than what most people think of as "race". So when you run
around telling people "science has proven there's no such thing as
race" you're seriously misleading them, because you are not talking
about what they think you're talking about. And two, your definition
of race is unstable, in that if there were two separate races, and they
mixed even slightly, then they would instantly stop being separate
races. What good is a definition like that, unless you are absolutely
intent on running around shouting "there's no such thing as race"?

>>And, the groups that we have divided the world into were based,
>>initially, upon the theory of multiple, separate evolutions for
>>man, which led us to imagine a =much= greater genetic disparity
>>than exists.

>>No, you have it backwards. The multiple, separate evolution theory was
>>a (failed) attempt to explain observed racial differences.

>And the fissions that are so keenly obsessed over by racists are
>are also "a (failed) attempt to explain observed racial differences."

Lane, you are in denial. The primary concern of Cavalli-Sforza's book
is mapping those "fissions", and he repeatedly affirms that the deepest
of them lies between Africans and non-Africans. I've quoted him
several times to that effect, but you simply refuse to acknowledge it.
The genetic trees in his book consistently have a branch labeled
"European" or "Caucasian", but you take no notice. Why?

>>>And yes, North Africans group with the Caucasians (as shown in the
>>>*second* color map, representing Africa, and explicitly stated at the
>>>beginning of section 3.4 in the text).

>>Modern north Africans maybe, but not ancient north Africans. This
>>map in question represents the world today, or as of 500 years ago
>>(he clearly states that he attempted to sample populations that
>>were extant 500 years ago in their present locale).

>>I don't think I believe this. Did you get this from the book, or do
>>you have another source?

>HGHG, p.4
>"Our primary interest is in understanding this evolutionary process.
>The first task is to describe the existing variation, using a variety
>of techniques that lend themselves to this work and allow us to test
>the relevant evolutionary models. We restrict our interest to aboriginal
>populations, which we define as those already living in the area of
>study in A.D. 1492."
>And on p.15
>"Because intercontinental travel began around A.D. 1500, we limited
>our study to aboriginal populations, those that were in place before
>that date; intercontinental transportation has subverted the earlier
>patterns of migrational flow."

Yes yes, I knew that the map in question represented the world 500
years ago. What I was asking for was the source of your assertion that
"ancient north Africans" were not Caucasian. Was *that* in the book,
or did you get it from somewhere else?

Why can't you answer my questions as I ask them?

>>>I happen to believe that black Africans and their descendants, as a
>>>group, are less intelligent than whites or Asians. I may be right or
>>>wrong in this, but it is certainly possible, just as it is possible
>>>that they have, as a group, darker skin and curlier hair.

>>HGHG, p156.
>>"Because genetic divergence was subject more to random than
>>selective[s] only moderate, IF ANY, influence of climatic
>>factors at the level of the NUCLEAR GENES investigated, but a greater
>>influence on genetic factors involved in the adaptation of body
>>build and bodily surface characteristics, which notoriously respond

>>Note the ellipsis everyone; something important was omited.

>Oh really? Then why didn't you include it here, so that you could
>back up your implication of dishonesty on my part with fact? Or
>is it really not pertinent to anything I have said? here it is:
>"Because genetic divergence was subject more to random than
>selective forces,--> much of the gradient of the human gene pool
>goes from west to east. The first principal component therefore
>extends in this direction and explains 35% of the total human
>variation,<-- showing only moderate, if any, influence of climatic
>factors at the level of the nuclear genes investigated, but a
>greater influence on genetic factors involved in the adaptation
>of bodily surface characteristics which notoriously respond to
>climate. A dichotemy is thus observed between genetic data, and
>observations based on the physical constitution..."
>Doesn't seem to affect the gist of what I'm trying to convey. In
>fact, it points out that the post-split genetic differences,
>most of which were the result of random genetic drift, and not
>of any kind of selection, compose only 35% of the 15% variance
>in genetic makeup between any two humans. That is, the 6% number
>that is often mentioned as the maximum difference to be found
>between two distant populations.
>As they say "moderate, if any, influence of climatic factors" on
>anything other than "bodily surface characteristics which notoriously
>respond to climate." So what does this say about your whole
>"winterlander" hypothesis, other than that they had a Vitamin D
>deficiency in their diet?

Interestingly, even after making a big show of including the entire
quote, you still leave out the part which indicates that Cavalli-Sforza
was talking about *Asia*, not the entire world. And why, after all
that indignation over my pointing out the ellipse in your quote, did
you go and delete the three paragraphs where I tried to show why the
part you left out was important? Let me reinsert them...

>>You have been using this quote a lot in other posts, so I'd better
>>address this. What Cavalli-Sfortza is saying is that "body build and
>>bodily surface characteristics" are not necessarily good indicators of
>>genetic relatedness. And he is absolutely right. I was well aware,
>>even in high school, that the dark skin of Africans, southern Indians,
>>and Australians did not imply that these people formed a group. This
>>is the "dichotomy" Cavalli-Sforza is talking about.
>>So what *does* Cavalli-Sforza consider a good indicator of
>>relatedness? Well, in the example you quoted from, in the part you
>>omited, he points out that in Asia "much of the gradient of the human
>>gene pool goes from west to east", that the "first principal component"
>>extends in this direction, and that the north-south climatic variation
>>in Asia has little effect on this. In measuring relatedness, it's
>>these "principle components" that matter, not appearence.
>>But Lane, an east-west genetic gradient is exactly what I expect to see
>>in Asia, as part of the long continuous transition from Caucasoids in
>>Europe to Mongoloids in east Asia. Once again, I have no problem
>>whatsoever with Cavalli-Sforza's conclusions. You are the one
>>obsessing on physical appearance, not me.

Lane, I expect you'll go on misusing this quote no matter what I do,
but maybe I can explain it for everyone else.

One -- *any* set of genes can evolve quickly if there is sufficient
selective presure.

Two -- most of the human genome does not evolve quickly, and has been
stable for millions of years (note that we share 98 percent of our
genes with Chimpanzees). This is because most of the genome codes for
proteins and enzymes and such that are well optimized, and are not
normally under much selective pressure. If you look at the genome as a
whole, most of the change comes from random genetic drift. When two
populations separate, their genomes begin to drift apart in a fairly
regular way. Because this drift, in genes that are not under great
selective pressure (most of the genome), is somewhat predictable (in a
statistical way), we can use it to learn something about how a given
population is related to other populations, and how long ago they might
have separated. This is what Cavalli-Sforza is trying to do in his

Three -- some sets of genes, such as those that code for skin color, or
body build, or intelligence (maybe!), *are* put under a lot of
selective presure when people move from one environment to another.
Traits like this are in a sense superficial (which is not necessarily
the same as "unimportant"), not because they can change quickly (any
set of genes can do that under the proper circumstances), but because
they *have* changed quickly, under the influence of non-random external
forces, and that makes them useless for the sort of statistical
analysis that Cavalli-Sforza is engaged in. What this means in
practice is that you cannot claim that Africans, south Indians, and
Australians form a proper group simply because they are all dark
skinned. *This* is what Cavalli-Sforza is pointing out in this quote.
He is noting that in Asia even though the climatic gradient goes from
north to south the genetic gradient (i.e., the gradient of those genes
which drift in a predictable way, and are thus useful for determining
how groups are related) goes from west to east.

Now Lane has seized upon this quote and used it to insist that since
certain "bodily surface characteristics... notoriously respond to
climate", it therefore follows that race, which as far as he is
concerned is entirely a matter of "surface characteristics", has
nothing to do with how human populations are related to each other.
Note that he entirely ignores the possibility that human intelligence
might be one of the traits that responds to environmental stress. This
is because he is intent on using this quote to argue that it isn't
*possible* for people from different parts of the world to differ
significantly in intelligence, the way they manifestly differ in other
(supposedly more superficial) traits. To a less biased observer the
weakness of this argument is clear; we simply don't know enough about
either the human genome or the genetics of intelligence to draw such a
conclusion, however much some people (for example the clueless reviewer
of HGHG in Time Magazine) would like to.

But Lane ignores something else just as important, which is that the
map of human relatedness that Cavalli-Sforza *does* come up with, after
all his careful statistical analysis, looks a great deal like the
racial classifications physical anthropologists came up with years
ago! His strongest conclusion is that the deepest genetic division in
the human race lies between Africans and non-Africans. His genetic
trees typically divide first into African and non-African branches, and
then the non-African branch divides into a South-Asian/Australian
branch and a North-Asian/European branch. "Europeans" or "Caucasoids"
generally have a branch of their own, and the main area of confusion
seems to be Asia, with the relationship between north and south Asia

These genetic trees and graphs, and the relationships they represent,
form one of the major conclusions of the book. Lane has simply blown
them off, because they are not what he wants to see. But how is it
possible, if surface features are indeed "superficial", that physical
anthropologists who had little else to work with should have arrived at
conclusions so close to those Cavalli-Sforza arrived at by analyzing

I think it is because they never relied on just *one* trait, but tried
to look at as many as possible. If you group by just dark skin alone,
your results are unreasonable. If you group by dark skin, tightly
curled hair, a broad nose and thick lips, then you come up with
Africans and some of the people of Oceania. Make an educated guess
that these groups are too far apart to be related (which turns out to
be correct), and you've replicated one of Cavalli-Sforza's groupings
(although not his conclusion that Africans form one of the two main
branches of humanity).

Note BTW that it isn't necessarily true that *all* racial features
respond quickly to climate. Skin color and body build certainly do,
but what about hair form, or the epicanthic eye fold and characteristic
facial bone structure of Asians? At least some racial features
probably did arise after the races separated, and *do* trace separate

Anyway, I've spent more time on this quote than I intended to, and I
want to remind Lane that I had *two* main points in my post that I
wanted him to respond to. He took a weak stab at one elsewhere (my
query as to how long *he* thought it would take for a 15 point IQ
difference to arise between two separated populations), but he has
entirely ignored the other, which is that while genetic variation
between races may account for only 6 percent of total human genetic
variation, this number is highly dependent on how much variation
existed in the ancestral human population prior to racial
diversification, and tells us little about how important the
differences between different racial groups might be. He also has yet
to explain Cavalli-Sforza's use of the word "hierarchy" to describe how
his "population clusters" are structured, and what this does to the
argument that the notion of race must be meaningless because the number
of racial groups you can come up with is arbitrary. In fact, it seems
that *most* of the points I made in my post have gone unanswered.

Maybe I should just repost it. >:->

Tom Lathrop | Politics: A strife of interests masquerading | as a contest of principles. -- Ambrose Bierce