>>12021897>>12021897A discussion whether or not various human populations cluster closely together by selecting some haplogroups or calculating overall genetic distance. How latter is actually calculated remains however obviously obscure in this thread.
The classic model used in the first half of the 20th century divided the human population into three to five races. Negroids, Caucasoids, Mongoloids, Americans (if not grouped together with Mongoloids) and Australoids (if not grouped together with negroids). This model is in general a somewhat good approximation for some racial boundaries. For example, the border regions where two races meet are characterized by a rather fast and very visible change in phenotypes. Going from caucasoid Delhi to mixed Nepal into purely mongoloid Tibet is such an example.
Modern genetics obviously has found out that there is no clear and sharp boundary of races in terms of single haplogroups or genes. This however doesn't mean that there is no difference at all. It also doesn't mean that various races don't cluster together which they obviously do.
Based on phenotypes, it is possible to define as many as 20 races. Race here is determined by a selection/set of concurring or clustering phenotypes that will allow someone to properly identify a member of such a race with the exclusion of all other races.
For example, one can't introduce the idea of a nordic race because the phenotypes that characterize the nordic race are found throughout Europe without an actually sharp boundary. However, one can easily draw a line between the European and the Siberian race allowing the proper identification of any member with a nearly non-existing false-positive chance. You can't mistake a hunter living in the northern Urals for a Russian for example.
The overall genetic distance between humans is actually not that small. The rather significant genetic distance actually observed in humans is probably the result of numerous emigrations.
