20. Human Diversity and Its History
311
degree of lumping is very controversial, but it may suggest interesting patterns.
There does seem to be support for a North-South, rather than East-West, division
of Eurasia. In the North there are two language families: Dene-Caucasian in scattered patches including several in the New World, and Eurasiastic. The pattern is
one in which the Eurasiatic languages underwent a later expansion at the expense
of Dene-Caucasian leaving behind relics (including Chinese!). South of the
Himalayas there are more groups each more restricted reflecting perhaps lower
mobility or the absence of subsistence innovations that allowed the wide expansions seen in the North.
This pattern in world languages supports the southern-branch theory of the
expansion of modern humans. There are two families north of the Himalayas, with
Dene-Caucasian perhaps the more ancient and associated with a pre-agricultural
expansion and Euro-Asiatic superimposed on it. In the tropical regions there are
four families in Asia and Oceania. The Indo-European intrusion into India and the
Sino-Tibetan intrusion into, southeast Asia may be relatively recent events.
3.4 Kalahari Bushmen
An interesting case for the retention of old appearance in the face of gene flow is
provided by Kalahari Bushmen. (They are called San in some recent literature, but
this is a nasty word to use to someone in the central Kalahari, and we avoid it.)
These are people w~o historically made their living by either foraging or keeping
small stock in southern Africa. They resemble, to European eyes at least, east Asians.
They have yellowish rather than black skin, epicanthic folds, shovel-shaped incisors, and many newborns have "Mongoloid spots" at the base of the spine. The
Asian appearance is not just a perception of Europeans. In the !Kung language
there are three kinds of mammals: !a is an edible animal like a warthog or a giraffe,
!oma is an inedible animal like a jackal, hyena, black African, or European, and
zhu is a person. Vietnamese in Botswana were immediately identifed as zhu by
Bushmen. In other words, their perception of their similarity to Asians is the same
as ours (I.e. Europeans').
Immediately to the north of the !Kung there are groups of so-called "Black
Bushmen" across Africa. The most familiar of these are the Berg Dama of Namibia.
These people speak Khoisan languages related to Nama Hottentot but in appearance they are completely like their Bantu-speaking neighbors. Our model suggests
that intermediate appearance should be unstable and that selection should drive
appearance rapidly toward one "type" or the other. The "Black Bushmen" may
have received just enough gene flow from Bantu invaders to cause a switch to the
other adaptive peak with rapid (in evolutionary time) loss of the earlier Bushman
appearance.
Are there critical tests of this model? The model predicts retention of shared
genes affecting external appearance between Bushmen and east Asians, but retention of nothing else. Hence, in agreement with the model, neutral Bushman genes
seem completely African as shown in Fig. 2. The other more interesting prediction
311
degree of lumping is very controversial, but it may suggest interesting patterns.
There does seem to be support for a North-South, rather than East-West, division
of Eurasia. In the North there are two language families: Dene-Caucasian in scattered patches including several in the New World, and Eurasiastic. The pattern is
one in which the Eurasiatic languages underwent a later expansion at the expense
of Dene-Caucasian leaving behind relics (including Chinese!). South of the
Himalayas there are more groups each more restricted reflecting perhaps lower
mobility or the absence of subsistence innovations that allowed the wide expansions seen in the North.
This pattern in world languages supports the southern-branch theory of the
expansion of modern humans. There are two families north of the Himalayas, with
Dene-Caucasian perhaps the more ancient and associated with a pre-agricultural
expansion and Euro-Asiatic superimposed on it. In the tropical regions there are
four families in Asia and Oceania. The Indo-European intrusion into India and the
Sino-Tibetan intrusion into, southeast Asia may be relatively recent events.
3.4 Kalahari Bushmen
An interesting case for the retention of old appearance in the face of gene flow is
provided by Kalahari Bushmen. (They are called San in some recent literature, but
this is a nasty word to use to someone in the central Kalahari, and we avoid it.)
These are people w~o historically made their living by either foraging or keeping
small stock in southern Africa. They resemble, to European eyes at least, east Asians.
They have yellowish rather than black skin, epicanthic folds, shovel-shaped incisors, and many newborns have "Mongoloid spots" at the base of the spine. The
Asian appearance is not just a perception of Europeans. In the !Kung language
there are three kinds of mammals: !a is an edible animal like a warthog or a giraffe,
!oma is an inedible animal like a jackal, hyena, black African, or European, and
zhu is a person. Vietnamese in Botswana were immediately identifed as zhu by
Bushmen. In other words, their perception of their similarity to Asians is the same
as ours (I.e. Europeans').
Immediately to the north of the !Kung there are groups of so-called "Black
Bushmen" across Africa. The most familiar of these are the Berg Dama of Namibia.
These people speak Khoisan languages related to Nama Hottentot but in appearance they are completely like their Bantu-speaking neighbors. Our model suggests
that intermediate appearance should be unstable and that selection should drive
appearance rapidly toward one "type" or the other. The "Black Bushmen" may
have received just enough gene flow from Bantu invaders to cause a switch to the
other adaptive peak with rapid (in evolutionary time) loss of the earlier Bushman
appearance.
Are there critical tests of this model? The model predicts retention of shared
genes affecting external appearance between Bushmen and east Asians, but retention of nothing else. Hence, in agreement with the model, neutral Bushman genes
seem completely African as shown in Fig. 2. The other more interesting prediction