Comparative AND Reconstructive Method

[Sean Crist]

[ moderator re-formatted ]

On Wed, 4 Aug 1999 ECOLING at aol.com wrote:

>> The Comparative Method is concerned with the reconstruction of categories,
>> _not_ the phonetic values which might have been the realization of those
>> categories.  This is a very important point.  When we talk about
>> Proto-Indo-European */a/, we don't mean "the phonological category in
>> Proto-Indo-European which had the phonetic realization [a]"; we mean "the
>> hypothetical PIE category which gave rise to the Sanskrit category /a/, the
>> Latin category /a/, etc."

> Agreed in part, in part not.
> We DO and MUST mean not merely something which represents a set of
> correspondences, we DO and MUST mean something WHOSE PHONETIC SUBSTANCE could
> have given rise to the observed attestation.

We assume that the speakers of the languages we reconstruct had phonetics.
What I am saying is this: the Comparative Method (strictly defined) does
not tell us what the phonetic values of the reconstructed categories were.
This is simply a statement about what the Comparative Method is, and what
it can and cannot do.

> Therefore the following is not valid:

>> As far as the Comparative Method is concerned, we could designate that PIE
>> category with an integer, e.g. "Category 27".  */a/ is just a convenient
>> label or nickname for it.

> Re the following, the hypothesized proto-form should be much more than
> merely a guess, it should be a highly-educated hypothesis or estimate,
> resting not on "intuition" but on educated reasoning and experience.

I'll agree this far:  if I were betting on the phonetic value of PIE */a/,
I'd put my money on the pronunciation [a], much in the same way that I'd
bet e.g. that the human skeletons we excavate from 1000 BCE in Denmark
probably belonged to individuals who had white skin.  There is, however,
no direct evidence which allows us to confidently state the skin color of
those individuals.  We assume that they had skin, and we can make educated
guesses about their skin color, but we're not entitled to state their skin
color as a fact.

When it comes to the prehistoric mergers of phonological _categories_,
however, we _do_ have evidence of what happened.  The evidence is the
correspondence sets among related, historically attested languages.

> Whether one calls it "Comparative Method" or not, Calvert Watkins is
> absolutely correct, that the RESULTS of the valid application of the method
> must in general aim at a reconstructed language FROM WHICH one can derive the
> hypothesized descendants,

I agree totally (other than the fussy point of switching "hypothesized"
to "attested", which is probably what you meant). I'd be surprised if
Watkins would disagree with what I'm saying, however; it is not at odds
with your paragraph.

> If some wish to call this "Reconstructive Method", and distinguish it from a
> very narrow sense of the "Comparative Method", then they will logically be
> forced to CEASE applying the term "comparative method" to much of what we
> traditionally have called that.  The "comparative method" includes, ideally,
> the "reconstructive" aspects also.  There is no point in trying to define it
> to exclude those, because they are part and parcel of the best comparative
> practice.

> Making this explicit allows us to IMPROVE AND EXTEND the totality of our
> comparative & reconstructive methods.
> Hiding it under the rug blocks progress, and misrepresents to students what
> we think know and how we know it.

I'm going with the definition of the Comparative Method in the strict
technical sense of the term, as discussed in Hoenigswald 1973.

I guess it depends on what you mean by 'know'.  Conclusions drawn from the
mechanical application of an investigative method are one thing; educated
speculation is another.  When we reconstruct PIE laryngeals, we do so on
the basis of a strict application of the Comparative Method; when we
speculate about the phonetic values of those categories, we're doing just
that: speculating.

The most important point, I think, is this: while the question of the
phonetic values of e.g. the PIE laryngeals is interesting, it _doesn't
matter_ from the standpoint of the Comparative Method.  You don't have to
know the phonetic values of those categories to be able to do the
Comparative Method.

> Lloyd Anderson
> Ecological Linguistics

What is Ecological Linguistics, BTW?

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]