Linguistic classification

[bwald]

----------------------------Original message----------------------------
I think  it would be interesting to try out Scott and Alexis's suggestions
on problematic areas of lanuage classification.  The prerequisites are that
whoever has a suggestion or criticism to make can relatively CONCISELY
explain what the issue is, and why there's disagreement.  Is the
disagreement over some CRUCIAL piece/s of data, or is it about the AMOUNT
of data, or what?  Furthermore, while literature can be suggested for those
who are more interested, it cannot be necessary, because most people won't
read it.  They just don't have the time.  And if understanding somebody's
argument depends on reading literature, most people will just delete the
message without reading it (or store it for "future" reference, if they can
ever find it again.)  NO EXTRA READING!
 
I would be interested in seeing how a number of shared issues would drop
out of the various problematic areas, so that we might all get more
interesteg in the details of whatever areas are discussed -- as "case
studies" in a larger set of problems which will confront historical
linguists in virtually any area of classification they would choose to work
in (if they push it far enough).
 
For example, Scott suggests:
 
> It's time to drop
> the notion of Takelman as a genetic unit; Takelma clearly belongs
> with the Coast languages, and Kalapuya, if we really knew much about
> it, is probably most closely related to the Plateau languages.
 
On what basis has Takelman been considered a genetic unit?
Why is this any more problematic than Romance or Germanic as a genetic unit?
What is the status of the Coast languages as a genetic unit?
What is the relation of Takelman to Kalapuya?
What is the relation of Coast to Plateau?
What is the point?  Is it equivalent to deciding whether Latin is "closer"
to Celtic or Germanic, or is it about whether Haitian Creole is a Romance
language, or whether English is a Bantu language, or what?
 
I am not interested in the classification of Takelman and Kalapuya per se.
But I am interested in why problems of the type implied (if not satirized)
by Scott happen.
 
We have a common set of references in IE at least.  Are there
classificatory issues in other language families which are not the same as
in IE?  Why (not)?
 
I already asked obliquely with regard to Alexis.  What's with Altaic?  Why
is it such a problem?
 
Do we need thousands of pages to explain why IE or Germanic works as a
unit?  I think we can boil such accounts down to something quite succinct
(if I'm wrong we have another lively discussion).  Why isn't it the same
with Altaic?  After all, it doesn't seem to me that IE was a particularly
"easy" language family to reconstruct?  What makes Altaic so much harder?
Is the problem in the phonological reconstruction, in the grammar -- what
in particular, how?
 
Are there any generalizations about where the attempts to unite groups of
languages tend to become problematic?  What are the arguments we can
dismiss (typological ones? areal ones?)  Which ones are generally
bothersome because we don't know quite how to deal with them?
 
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
 
What else?