From: johnvertical@...
Message: 70566
Date: 2012-12-11
--- In cybalist@yahoogroups.com, Bhrihskwobhloukstroy wrote:
( )
> Everything, in this model, seems possible (to me). If there's
> anything that a priori must be excluded, please have the patience
> and courtesy to point it out (once again and more precisely than
> before)
The biggest issue I see here is not any particular detail, but the suggestion that there is a division between "possible" theories and "a priori excludable" one. With sufficient special pleading - on archeology, geology, astronomy and physics if need be - a scenario wherein PIE was originally spoken arbitrarily wide and arbitrarily far in the past can be proposed (or, for that matter, some Last Thursdayistic conspiracy theory where PIE was still the language spoken in 1901), without its plausibility ever dropping to complete zero.
The suggestion that all such theories have to be "taken into consideration" is of course valid, but only in proportion to their plausibility. The most outlandish scenarios would deserve not even an explicit mention.
> I meant that phonological changes from Proto-West-Germanic to
> Middle English, from Proto-(Balto-)Slavic (the stage of the
> Prehistory of Slavic at about the first centuries AD) to Mediaeval
> Polish, from Latin to Old French and so on (all taking place inside
> the time extension from 1 to 1500 AD) are more numerous and complex
> than the ones from Late Western Indo-European to Proto
> West-Germanic, from North-Central Indo-European to
> Proto-(Balto-)Slavic and from Proto-Italic (or its variety in
> Latium) to Latin (all presumably between 2000 and 1 BCE), which in
> turn are much more numerous than from Proto-Indo-European to Late
> Western Indo-European, North-Central Indo-European, and Proto-Italic
I agree with this observation, with one caveat: this applies to the number of *known* phonological changes. Applicable data generally gets scarcer, the farther in the past we look, and with it, so does the extent to which we can uncover changes affecting it.
A look at the subtypes of sound changes seems to confirm this. What we can reconstruct between PWG and ME (or anything in this period) amounts mostly to conditional changes, with not many shifts in the overall phonological system. In PIE to PGmc times we can identify the large-scale transformations that have affected wide parts of the reconstructed vocabulary, but rather fewer "small" changes.
I consider it rather likely that minor conditional changes are what causes numerous irregular correspondences found within the proposed lexicon of any older protolanguage - but with scarce data it becomes easy to propose conflicting hypotheses, so determining exactly what these were is a harder question. (I also believe that a rigorous, detailed analysis of all available data can still lead to plenty of further discoveries, but only time will tell.)
_j.