"mono-descent is implicit in the comparative method ..."

[X99Lynx at aol.com]

In a message dated 7/4/01 8:26:11 AM, dlwhite at texas.net writes:
<< "Implicit" may be a bit strong, but the comparative method as
traditionally conceived and practiced has never contemplated mutli-descent. >>

An important distinction.  "Implicit" seems simply inaccurate.  As I
understand it, the comparative method IN AND OF ITSELF contemplates neither
multiple nor singular descent of 'languages'.  Rather it deals with forms.
Your position that finite verb forms are never 'borrowed' or 'mixed' relates
- objectively and without theoretical extension - only to forms.

If there were languages that was made up solely of finite verbs, the question
of relatedness of these languages could be narrowly defined.  But that
appears not to be the case.

<<The comparative method does indeed tell me that such a fundamental aspect
of a language as finite verbal morphology cannot be treated as non-existent
merely because its origins are problematic.>>

Fine.  Assume that 'finite verbal morphology' is present and whole-heartedly
'treated as existent' and assumed to be fundamental and therefore the
preemptive indicator of descent.

The real question in the hypothetical is why, given those assumptions, it is
proper to label the 'nominal morphology, derivational or non-finite verbal
morphology,... and categories' shared with a different language family as
non-genetic.

I wrote:
<<My point is that YOU can't "coherently distinguish between influence and
descent."

dlwhite at texas.net writes:

<<Yes I can, in a way that explains why finite verbal morphology is evidenty
not subject to borrowing in the same way as nominal morphology.  You just say
it's circular when I do, yet still refuse to answer the question of why
language mixture or multi-descent, if it is really so common as you say, has
not led to mixed finite verbal morphologies all over the place.>>

Let's assume that 'finite verbal morphology' is never 'borrowed' or 'mixed'
and can only be transmitted by ordinary processes from an ancestor to a
daughter language.  That still does not tell me or anyone why other forms
cannot be descended from a completely different ancestor.  Certainly, if
those forms include 'nominal morphology, derivational or non-finite verbal
morphology,... and categories' shared with a different language family.

As a real example, Niger-Kordofanian is a "well-established language family
where... there was very very little lexical evidence for this family," but
"the nominal class system of the major subfamilies agree in detail as to the
markers for the different classes."  This prompted "Baxter and Manaster
Ramer(1996), following... Schadenberg (1981)" to classify "Niger-Kordofanian
as ... established purely on [nominal] morphological grounds...."
Non-derivative verbal morphology apparently played no part in establishing
the overall family.  (from a hl post from manaster at umich.edu dated Tue, 24
Feb 1998 14:43:02 EST.)

Here, noun morphology is used to establish descent.  Could a suddenly
discovered contra-indicating verbal morphology shared with other languges c
hange the status of that noun morphology from "genetic" to "non-genetic?"
The comparative analysis of those nominal forms would not have changed in the
least.  So it has to be theory outside the comparative method that causes the
shift in status from "genetic" to "non-genetic."  The systematic
correspondence of the FORMS remains intact.  So, again, why should the
presence or absence of shared verbal morphology change the "genetic" status
of other morphology in the language, or negate the possibility of more than
one familial relationship?

<<It seems that your basic problem is failure (or refusal) to discriminate
between descent and origin of morphmemes.>>

How is "language descent" established in any way other than by the origin of
morphemes and other shared forms?  The concept of language descent is
entirely based on the establishing systematically corresponded forms.

I don't think you have supplied a coherent explanation of how you
"discriminate between descent and the origins of morphemes."

But my question is specifically what theory can justify making language
descent a contest between different sets of validly corresponding systematic
forms - multiple morphemes with validly established multiple origins?

When the comparatively proven descent of forms show different origins, why
should one be "genetic" and the others "non-genetic?"

You will need a completely different theoretical basis for the "mono-descent"
of languages.  The comparative method apparently will not support it.

Regards,
Steve Long