25-10-03 15:05, elmeras2000 wrote:
> Only
> those splits that led to language varieties that were recorded in
> such a way that they have entered the basis of later scholarship, or
> even - the best cases - were allowed to live and become separate
> languages of the present-day world, can be seen as part of the
> breakups we care about. This shows that the much in the business of
> prehistoric dialectology is based on whims of chance and therefore
> extremely suspect.
Palaentologists have a convenient name -- taphonomy -- for the
subdiscipline concerned with the processes thanks to which living things
become part of the fossil record, and with how the enormously heavy odds
against an animal or plant becoming a museum specimen affect the
information that our reconstructed phylogenies are based on. Taphonomic
considerations _should_ be important in linguistics too, but many
scholars apparently prefer to ignore the unknown. I've seen maps of,
say, early Iron Age Europe showing (pre-)Proto-Germanic here,
Proto-Celtic there, then Proto-Baltic, Proto-Slavic etc. as large
coloured areas that border on one another and leave no place for less
familiar groups. It's more than likely, however, that the continent was
a crazy-quilt in which most of the patches would represent extinct
languages that have left no documented descendants.
Piotr