Dating PIE

[Scott DeLancey]

----------------------------Original message----------------------------
I certainly agree with Larry's points here, especially:
 
> The linguistic evidence alone is not sufficient to give us a
> *terminus post quem*, but recall that there exists a standard
> argument for estimating this.  Many specialists are satisfed that we
> can reconstruct PIE words pertaining to technology, and most
> particularly to wheeled vehicles, including `wheel', `axle' and
> `nave'.  Now, if this is true, then it follows that the PIE-speakers
> must have known wheeled vehicles.  But the archeologists can find no
> evidence for wheeled vehicles before about 6000 BP.  Therefore, the
> arguments runs, 6000 BP is a *terminus post quem* for PIE.
 
Obviously, as Larry notes, there are ways that this argument could
be wrong, but it is a legitimate argument.  Archeological finds are
datable, by objective, replicable methods.  When it's possible to
link a proto-language with some archeological facts, we have a
respectable basis for dating.
     The kind of argument that I get really upset about is the
one that goes:  "Well, family X seems to have about the same
degree of diversification as Germanic, and everybody says
Proto-Germanic is about 2,500 BP, so the time depth of X is
2,500 years".  Leaving aside the theoretical issue of whether
all language families, under all conditions, will diversify at
the same rate, the really outrageous part of this argument is
the notion that we have some objective, quantifiable measure of
diversification that would let us equate, say, Miwokan with Germanic.
An argument of this kind cannot be anything but purely impressionistic,
and frankly, I don't see any reason to trust *anybody's* impressions
in such comparisons.
 
Scott DeLancey
Department of Linguistics
University of Oregon
Eugene, OR 97403, USA
 
delancey at darkwing.uoregon.edu
http://www.uoregon.edu/~delancey/prohp.html