----- Özgün İleti -----Kimden: Mnewbroo@...Gönderme tarihi: 24 Temmuz 2003 Perşembe 13:07Konu: Re: [historical_linguistics] Digest Number 20
Re recent material
Again, please circulate this to other relevant groups. Thanks.
Some of Mark Hubey's specific views are themselves highly controversial (as he himself notes in other terms). It might also be held that a distinction (albeit not totally watertight) should be clearly drawn here between traditional comparative linguistic methodology and more directly probability-based methods as used for deep-time comparisons ('mass comparison' etc).
There is really no controversy. The methods are all based on probability theory, explicitly or implicitly, and orthogonally, bilaterally or multilaterally.
There can be no other basis for it. It just so happens that I write explicitly whereas the traditional explanations are based on heuristics whose
justification cannot be other than probabilistic..
Yes, it must be so because it is true. See for example the book by Sheila Embleton.But the generalities of what Mark is saying here would rightly be accepted by everyone with a serious background in historical linguistics.
Polat Kaya is right to say that, if there really had been deliberate interference (anagrammatisation etc), probabilistic considerations would not apply in the same way as in the case of normal linguistic change.
That idea in itself is subject to the laws of probability e.g. what is the probability that Turkish was subjected to anagrammatisation when there is no
evidence that any other language was subjected to the same.
Mark Newbrook
-- Mark Hubey hubeyh@... http://www.csam.montclair.edu/~hubey