Trivial Truths and Genetic "Patterns"

[Rich Alderson]

On 30 Jun 2001, Steve Long wrote:

> In a message dated [= received at AOL --rma] 6/30/2001 5:52:51 AM,
> alderson+mail at panix.com writes:

>> So that, I think, is the pithy answer to your continued asking of why any
>> language can't have more than one ancestor under the comparative method:
>> The method does not address single languages, but groups of languages, and
>> as was pointed out by Larry Trask, if we can't build a protolanguage with
>> the comparative method, the languages in question are *JUST* *NOT*
>> *RELATED*.

> Then let my rephrase my question.  Can more than one proto-language be
> reconstructed from a "group" of languages?

No.  That's what we've been telling you all along:  The comparative method
absolutely assumes a single parent for any group of related languages, and
cannot assume anything else.

> And, if so, then will those additional proto-languages show up in
> reconstruction if one assumes before hand that there was only one
> proto-language?

The question is meaningless.

> If on the other hand the answer to the first question is no, then why not?

The quote you cite from Lehmann makes that clear:  The reconstruction process
assumes a single ancestral form for each set of forms taken from the multiple
languages under discussion.

> This goes back again to the terms I quoted from Winfred Lehmann: "In using
> the comparative method, we contrast forms of two or more related languages to
> determine the precise relationship between these forms.  We indicate this
> relationship most simply by reconstructing the forms from which they
> developed."  (Hist Ling 3d ed, 1992 pb) p 142.

The total set of reconstructed forms, based on all the forms attested in the
group of languages under examination, is considered to be a single language.
Period.  By definition.  Period.

> The reason was to relate it way back to the hypothetical that Dr White
> brought forward where the "nominal morphology, derivational or non-finite
> verbal morphology,... and categories" might be shared with one language,
> "finite verbal morphology" with another.  Dr White indicated that verb
> morphology might have some preemptive claim over those other forms.

That is between Dr. White and you.  I can sympathize with his position, but am
not prepared to defend it.  However, be careful here, because this example is
of *ONE* *SINGLE* *LANGUAGE*, and you already know that the comparative method
is about *groups* of languages (and the individual forms taken therefrom).

> But my point was that the presence of those diverging systematic forms would
> suggest more than one "proto-language" could be reconstructed at least with
> regard to the language that showed both forms.

*Only* if these diverging forms can be found elsewhere within the same group of
languages being examined at the time.  If there are no correlates, there is no
reconstruction possible, and therefore no proto-language.

I repeat in full a definition I have given before in brief, from Haas' _The
Prehistory of Languages_ (p. 32):

	A protolanguage, then, is reconstructed out of the evidence that
	is acquired by the  careful comparison of the daughter languages
	and,  in  the  beginning  of  the work,  what  is  reconstructed
	reflects what  can be discovered  by working backwards  in those
	cases where all  or most of the daughter  languages point ot the
	same  conclusion.  This  provides the  initial  framework.  Once
	this is established, the principle of analogy can be drawn upon,
	and  by  its  use  instances  in which  there  are  aberrations,
	statistically  speaking, can often  also be  plausibly accounted
	for.   Deductive  as  well   as  inductive  hypotheses  must  be
	constructed and checked.  Then when all the comparisons that can
	reasonably   be  made  have   been  made,   and  when   all  the
	reconstructions that can reasonably  be made have been made, the
	result is  a PROTOTYPICAL MODEL  OF THE DAUGHTER  LANGUAGES, or,
	what we normally call a protolanguage.

(The emphasis by all caps is in small caps in the original; it is what I often
cite as the definition of "protolanguage".)

								Rich Alderson
								linguist at large