Language and Anthropology in the Americas

[manaster at umich.edu]

----------------------------Original message----------------------------
I dont think that the concept 'long-range' is meaningful.
Afro-Asiatic and Uralic are no less diverse internally
than Altaic, but the former are universally accpeted
(except by Doerfer) while the latter is considered by
some to be in the 'long-range' category.  As for Nostratic,
without AA (as Starostin, Greenberg and some others would
like to see it), it ends up being not all that internally
diverse either.  And as for Amerind, first, I don't
think that there is any basis any more for critiquing
it (or any theory) for lack of regular correspondences.
A few years ago people could plead ignorance of teh history
of language classification and thus maintain the dogma
that relations are estblished on the basis of regular
correspondences and nothing more or less.  I would have
thought that such work as mine and Sidwell's on the
early history of language classification and mine
on the Sapir era would have helped put an end to this.
If not, then perhaps Lyle Campbell's recent publications
on N. American classification or Goddard's 1979 paper
on the Comecrudan language family would be enough.
I think Ives and Lyle are as well-regarded splitters
as anyone, but it is clear that they do not take
regular correspondences to be necessary and Ives
at least has always argued that they are not enough
either (because he says you need morphology).
 
Note well that I am defending Greenberg's Amerind
as such, but to say that it is no good for lack
of regular correspondences tabulated in a neat
little chart is definitely not acceptable in 1998
as it may once have been. Heck, I am pretty sure
I used to say this--before I learned better.
 
AMR
 
On Sun, 10 May 1998, Miguel Carrasquer Vidal wrote:
 
>
> Greenberg's Amerind is the "moral equivalent" of Nostratic (in the
> sense that it's definitely "long-range"), although given the fact
> that Nostratic is based on more or less regular sound
> correspondences, it isn't equivalent in any formal sense, nor *can*
> it be: the Nostratic theory is based on the further comparison of
> "medium-range" (IE, Uralic, AA) language family reconstructions, and
> in the Americas, with few exceptions, we don't have those
> "medium-range" families, and we don't have the reconstructions.
>
>
> =======================
> Miguel Carrasquer Vidal
> mcv at wxs.nl
> Amsterdam
>