Re: [TIED] Re: IE, AA, Nostratic etc.

Hakan writes:

 

To me, it seems likely that language arose just once, in one single location, because all the peoples of Earth have a common origin, but I agree with you that a scenario where several tribes of people developed language independently of each other is not impossible, and we will probably never know exactly how it happened.

 

The survival of a species consisting of only one small population would be a miracle from a strictly biological point of view. Cheetahs are a species that must have been nearly doomed at some unknown time in the past but managed to take a few steps back from the brink of extinction. As a result, all cheetahs (from tropical Africa to India) are highly inbred, with all the adverse consequences such as vulnerability to the same diseases and environmental threats. By contrast, there is no evidence of abnormally low genetic diversity for humans. I think it's highly likely that the total population of early anatomically modern Homo sapiens was fully sustainable (that is, rather large) at any time; and if so, then given the size of primitive social units, it was simply too large to support just one language community. I don't think language was invented like the wheel. Since we seem to be biologically adapted to using spoken language, this ability must have developed somehow in the normal course of evolution (unless you are prepared to believe in another miracle) out of more primitive forms of communication in a more or less gradual way. In this scenario, a number of languages could evolve parallelly but not quite independently, because interbreeding between different linguistic communities would ensure the propagation of common biological adaptations related to language use. As a result, all early humans would have shared a restricted language typology, but not a common language.

 

People (including many linguists) tend to think of language evolution in terms of family trees, and the natural expectation is that if you move sufficiently back in time you will get to the root of the tree. But where's the proof that family trees as we reconstruct them for the large Post-Mesolithic language families of Eurasia and N Africa (and for a few other well-established families elsewhere) represent a universally applicable model? The patterns we observe in Melanesia, Sub-Saharan Africa, the New World and Australia make me suspect that throughout the early history of language the genetic lineages of human dialects formed a tangled bush with many roots rather than a neat tree. Some speech communities grew large enough to split and produce sister languages, but such "vertical" tendencies were in a dynamic equilibrium with "horizontal" factors causing areal convergence and creolisation. In any synchronic section a hypothetical observer would have observed a small-sized family here, another one there, and a handful of isolates in between.

 

Piotr