Re: Race, intelligence, and racism

Lane Singer (
13 Feb 1995 07:54:32 GMT

In <> (Tom Lathrop) writes:

>I am afraid I am going to have to let your others posts under the
>Genetic Evolution thread go unanswered, as I simply don't have the
>time, and most of your points I felt I had already answered earlier.
>One point however just seems too egregious to let slide...
>>>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

>>Yes, because as I illustrated in a previous post, when confronted with
>>genetic data on almost 2,000 populations, he began pooling them on
>>a geographical basis. And, he used the names to identify them that
>>were present in the sources for much of the genetic data upon which
>>"The History and Geography of Human Genes" was based. By the
>>way, he ended up with 42 populations. Are you defining them as
>>42 different races?
>In other words, you are saying that the reason that Cavalli-Sforza's
>genetic trees again and again have a branch labeled "Caucasian" or
>"European" is that the people he got his data from used those labels to
>describe the sample populations. That the structure of his genetic
>trees are determined merely by the names "present in the sources", and
>other than that they have no content. That he is simply passing on to
>us the labels that were given to him, and that these labels don't
>actually mean anything. Can you really believe this?

-- From one of the unanswered posts:
The authors state that
they are using the names given the various populations by
scientists who gathered and published some of the genetic data
used as the foundation of this work, in 1977 and 1984. (p.73)

On page 20 they state that although "the individual can be the unit
of evolutionary study, this requires testing every individual for a
large number of genes, and the amount of information generated
soon becomes prohibitive." As it is, they started with data on
almost 2,000 populations and were forced, for logistical reasons,
to combine them: "Our main criterion in pooling populations for
generating higher categories was geographic, but it was clear
that, especially for populations from the developing world, the
geographic criterion had to be supplemented with general
anthropological information of some kind because populations
of widely different origins occasionally live only short distances
from one another."

The focal point for each of the 42 populations was geographical.
You didn't answer my question: are you saying that each of these
42 populations is a race?

Again, what =is= your definition of "Race", and how do you divide
people up into discrete races?

>And if so, how do you account for the principle component map on page

The PC maps result in a reflection of the populations as they exist
geographically, rather than as a representation of genetic distance
(this is touched upon in Chapter 8). The names used for areas
in the map are thus geographic in nature.

>This map, which in Cavalli-Sforza's words "epitomizes the
>conclusions reached by the tree analysis", places the 42 populations in
>a square graph space in which the vertical and horizontal axes are
>measures of genetic similarity between the groups. Oddly enough, of
>the 42 groups, *all* of the groups traditionally thought of as
>"Caucasian" are tightly clustered in the far upper right hand corner of
>the map (where we find the groups labeled Italian, Iranian, Danish,
>English, Greek, Basque, Sardinian, and Near Eastern), or are fairly
>close to this cluster (Indian, Lapp, and Berber). Way, way down in the
>lower right corner of the map are *all* the African groups (San,
>E.African, Mbuti, Bantu, Nilo-Saharan, and W.African). The separation
>between these two clusters is large and visually striking, with the
>only group even close to an intermediary position being
>(unsurprisingly) the Berbers, which nevertheless clearly belongs to the
>Caucasian cluster. With the exception of the groups labeled Chukchi
>and S.Dravidian (which are in the top middle of the map) all of the
>groups we normally think of as Asian are on the left side of the map,
>and again there is a striking separation, both from the Caucasian and
>African clusters and between the north and south Asian groups.

As stated above, PC maps tend to reflect geography more than anything
else. The Tables 2.3.1A and 2.3.1B (in the same section, on
pages 75 and 76) are much more revealing than the map.
They list all of the 42 populations and the genetic distance
between each one of them and the other 41. Some very interesting
data is shown here:

- There is greater genetic disparity between the San (Khosian) and
the Mbuti, both subSaharan African, than exists between the San and
the following populations:
- English, Danish, Iranian, Near Eastern, Mongol, Malaysian,
Basque, Greek, Italian and Chukchi (Asian)

- The Nilo-Saharan have more in common with the Mongols and the Basque
than with the Berbers.

- The Lapp, English, Danish, Greek and Italians all have more in common
with East Africans than they do with the Thai.

Also, you haven't touched upon the DNA mapping, which clearly shows
African presence (genetically) throughout the human race.

[more insulting rhetoric deleted]