> Mark P. Line scripsit:
>
>> Yep. The key lies in the concept of uniformitarianism.
>
> And a dangerous concept it is too. Lyell (who was by training a lawyer)
> established the concept of uniformitarianism of natural law (a necessary
> assumption for any scientific discipline: there are no causes for
> historical events that are hidden from us today) and invalidly extended
> the argument to uniformitarianism of state and of rate
Aye. Fortunately, my scientific worldview doesn't admit of natural law in
the first place, so I can afford to use a much more pragmatic version of
uniformitarianism. Obviously the principle can only be applied over a
piece of time during which your invariants can reasonly be held to be
invariant.
>> I don't
>> know of any comparative lexical work that uses cladistic methodology,
>> but
>> it wouldn't surprise me if somebody had thought to try it.
>
> Take a look at Ringe and Warnow's stuff, which was briefly discussed
> on the list in May 2003:
http://www.cis.upenn.edu/~histling/ et seqq.
> A particularly interesting result, which certainly no one expected,
> is that Germanic probably started out as a satem language that was
> mugged by centum speakers later on, due to a migration into centum
> territory.