308
H.C. Harpending and E. Eller
3 A New Direction
The emerging picture from new and plentiful data about neutral markers is one of
a smooth relationship between geographic and genetic distances with some perturbations due to recent history and to language. These findings suggest that there has
been a lot of gene flow between neighbors and that isolation by distance, rather
than history, dominates the distribution of neutral genes in human populations. We
believe it is time to extend our domain of inquiry to marker systems where history
is better preserved. In their pioneering study of human race differences, Nei and
Roychoudhury (1974) hint at a new direction. They say that their finding of rather
small race differences "does not apply to those genes which control morphological
characters such as pigmentation and facial structure." Darwin (1871) thought that
our race differences were driven by sexual selection, a view defended by Diamond
(1992) who shows that environmental selection accounts of race differences in skin
color and other traits do not bear close scrutiny. If skin color were a response to
climate, Tasmanians should have turned white. If this view of differences in appearance is correct, even in part, then physical appearance, like language, could
carry a signature of history much deeper than the signature in neutral genes.
When gene flow occurs between populations gene frequencies blend in a linear
way, so that over time differences between the populations are erased as history
yields to local migration-drift equilibrium. But both language and sexually selected
appearance may respond very differently to contact and admixture. We think it is
time for biological anthropology to turn its attention again to language and to external appearance, but we realize that both these topics are, putting it mildly, unfashionable. We will discuss a simple model showing how traits like this should
preserve deep history and discuss several examples.
3.1 Language and Rh among Basques
The Basques of the Pyrenees speak a language that is perhaps distantly related to
languages of the Caucasus mountains but is otherwise unique in the world. A common theory is that they are a relic of an earlier occupation of Europe that was
overrun by later invaders, perhaps the Indo-Europeans. Their genes are mostly
similar to those of their neighbors except that they have a frequency of Rhesus
negative that is greater than one-half. Following Ruhlen (1994) imagine that there
has been some small level of gene flow, one percent per generation, into the Basque
popUlation from their new neighbors for five thousand years, say 250 generations.
Then of the neutral genes in the Basque population today only .99 250 - eight percent
are descended from ancient Basque genes. The neutral parts of the genome have
essentially been replaced.
But consider the language spoken by these immigrants. Since they arrive in
small numbers they and their children learn Basque and, save for occasional loan
words, have little effect on the language. The Basque language has persisted over
H.C. Harpending and E. Eller
3 A New Direction
The emerging picture from new and plentiful data about neutral markers is one of
a smooth relationship between geographic and genetic distances with some perturbations due to recent history and to language. These findings suggest that there has
been a lot of gene flow between neighbors and that isolation by distance, rather
than history, dominates the distribution of neutral genes in human populations. We
believe it is time to extend our domain of inquiry to marker systems where history
is better preserved. In their pioneering study of human race differences, Nei and
Roychoudhury (1974) hint at a new direction. They say that their finding of rather
small race differences "does not apply to those genes which control morphological
characters such as pigmentation and facial structure." Darwin (1871) thought that
our race differences were driven by sexual selection, a view defended by Diamond
(1992) who shows that environmental selection accounts of race differences in skin
color and other traits do not bear close scrutiny. If skin color were a response to
climate, Tasmanians should have turned white. If this view of differences in appearance is correct, even in part, then physical appearance, like language, could
carry a signature of history much deeper than the signature in neutral genes.
When gene flow occurs between populations gene frequencies blend in a linear
way, so that over time differences between the populations are erased as history
yields to local migration-drift equilibrium. But both language and sexually selected
appearance may respond very differently to contact and admixture. We think it is
time for biological anthropology to turn its attention again to language and to external appearance, but we realize that both these topics are, putting it mildly, unfashionable. We will discuss a simple model showing how traits like this should
preserve deep history and discuss several examples.
3.1 Language and Rh among Basques
The Basques of the Pyrenees speak a language that is perhaps distantly related to
languages of the Caucasus mountains but is otherwise unique in the world. A common theory is that they are a relic of an earlier occupation of Europe that was
overrun by later invaders, perhaps the Indo-Europeans. Their genes are mostly
similar to those of their neighbors except that they have a frequency of Rhesus
negative that is greater than one-half. Following Ruhlen (1994) imagine that there
has been some small level of gene flow, one percent per generation, into the Basque
popUlation from their new neighbors for five thousand years, say 250 generations.
Then of the neutral genes in the Basque population today only .99 250 - eight percent
are descended from ancient Basque genes. The neutral parts of the genome have
essentially been replaced.
But consider the language spoken by these immigrants. Since they arrive in
small numbers they and their children learn Basque and, save for occasional loan
words, have little effect on the language. The Basque language has persisted over