Linguistic classification

[Scott DeLancey]

----------------------------Original message----------------------------
On Wed, 11 Feb 1998, bwald wrote:
 
> Can we get further than:
>
> NOT ENOUGH DATA  (because of time-depth; that's the biggie, isn't it?)
> CAN'T TELL BORROWINGS FROM INHERITANCES
> BASIC VOCABULARY
> CHANCE RESEMBLANCES
> MULTIPLE SOURCES
 
Try one more--not enough research.  Meaning, not enough researchers.
Take my Takelman (= Takelma-Kalapuya) example.  A special link between
these was suggested by Swadesh, 40-odd years ago, based essentially on
lexicostatistics.  A chancy basis in itself, not helped by the fact
that the work seems as though it involved a kind of, uh, personal
slant on what to consider a match.  It was pursued 20 years later by
Shipley, who basically worked over Swadesh's stuff and showed that
some of it is good.  So, that gives some basis for supposing a genetic
relationship between the two.  But in the context of the hypothesis of
Sapirean Penutian as a genetic unit, it doesn't do anything to support
a case for special relationship.  Indeed, I think that one reason
(actually the only plausible reason I can think of) why various otherwise
conservative Americanists (Lyle Campbell, for example) have bought
Takelman is that they reject the likelihood of Penutian a priori, so
the only possible inference from Shipley's results is Takelman.  Now,
if you look at the structures of the languages, even pretty superficially,
you don't see anything that looks like a particularly close relationship,
so I've always been a bit skeptical about this.  Now, at the SSILA meeting
in January, Daythal Kendall and Marie-Lucie Tarpent have pretty
effectively demolished the case for Takelman, showing that virtually
all of Swadesh's and Shipley's evidence falls into one of three baskets:
apparently spurious, forms that are more widely attested in Penutian,
and forms that seem to be loans from Takelma into Southern Kalapuya,
and aren't found in the other Kalapuyan languages.
 
OK so far.  But the languages still look Penutian (at least, Takelma
certainly does), so, is there a better subgrouping?  With Kalapuyan
out of the way, there's no impediment to grouping Takelma with the
Coast languages (Coosan, Siuslaw, Alsea), which has always been its
inspectionally most plausible affiliation.  But that's not going to be
more than a plausible suggestion until a fair amount of serious
comparative work gets done.  A lot of work there for somebody, and
the linguists aren't exactly lining up for the opportunity.
 
And what about Kalapuya?  The problem with Kalapuya--well, there are
serious documentation problems (reliably-recorded texts, but no grammar
or dictionary to start with).  But, most crucially, there's no Kalapuya
expert anywhere.  If I want to start comparing Kalapuyan grammar with
Sahaptian or Klamath, which would be the likely suspects, I can be
reasonably confident about the Sahaptian/Klamath side--but I'd be
much happier if there were someone somewhere that I could run my
understanding of Kalapuya by.  We all know what can happen to the
unwary in this kind of situation.
 
The real problem with Penutian is that it's a huge and difficult
comparative task, and there just aren't many people working on it.
And the totality of the problems that have to be solved is far beyond
any one or few linguists' capacity to handle.
 
And this is the problem with a lot of deep--and not so deep--proposals.
I'm sure Alexis could list for us, off the top of his head, everybody
in the world who is seriously interested in the Altaic problem, and
I doubt that the list would fill your screen.  No wonder we can't
decide what we've got.
 
Scott DeLancey
Department of Linguistics
University of Oregon
Eugene, OR 97403, USA
 
delancey at darkwing.uoregon.edu
http://www.uoregon.edu/~delancey/prohp.html