Principled Comparative Method

[Sean Crist]

On Mon, 19 Jul 1999 ECOLING at aol.com wrote:

> So if we had case (d)
> and we also had case (d'),
> one example of IE /t/ corresponding to Uralic /s/,
> we might have more evidence for saying that /d/~/z/ was real.

> Or if we had case (d) before a following /u/,
> and we also had case (d'')
> ONE example of IE /d/ corresponding to Uralic /z^/ before /i/,
> (interpret the ^ as a hachek = wedge, upside down)
> we would again have more evidence for saying that /d/~/z/ was real,
> in this case because we would assume that /z/ became /z^/ before /i/.

> And if lucky we might have case (d') above with following /u/,
> and also case (d'''),
> one example of IE /t/ corresponding to Uralic /s^/ before /i/.

> With all four of these examples, we would have a plausible system,
> and would in fact have four examples reflecting the same sound changes,
> perhaps actually only a single sound change (namely
> "affricate apical stops before high vowels, with the result being
> the fricative of the same voicing and "reflecting" the vowel quality)
> but we need not have more than one example of any particular correspondence
> on the surface.

	As a matter of terminology, the kind of argumentation you propose
here falls outside the Comparative Method.  It's true that we do sometimes
fall back on this kind of external guesswork, but it's the sort of special
pleading you make to bolster some assumption which you need for your
argument.

	In any case, I'd be really careful in assuming that phonological
rules always apply across entire classes of categories.  It's true that
things often do work this way, but they don't always.
	Around 12 years ago, I was interviewing a speaker of the South
Midlands dialect area of American English.  In this dialect, /I/ > /i/ for
some speakers, and /U/ > /u/ for some speakers.  I assumed that if the
speaker had one rule, she'd have the other as well; I figured that the
general rule was "high lax vowels become tense".  To my great surprise,
she had the second rule but not the first.  There was no denying what I
was hearing; my assumptions about the expected symmetry within the system
were just plain wrong.
	I've fallen into this same kind of trap plenty of other times.
There are so many cases of beautiful symmetry and parallelism in
phonological systems that it's an ongoing challenge to remember that
things don't always work out so neatly.  Even in the Japanese case that
you give, there are recent loan words (e.g. tiishatsu "T-shirt") to mess
up the nice symmetry of the system.

> One of them I am quite certain is to develop both articulatory and
> acoustic "spaces", relative "distances" between different articulations
> and different acoustic effects, so that when attempting to judge
> likelihood of cognacy of pairs of words, we can judge similarity
> by degrees, not by yes/no dichotomies.  (These will partly depend
> on the general typology of the sound systems of the languages
> concerned, they will not be completely universal, but they will
> also not be completely idiosyncratic.)

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."

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.  Along with that label comes a built-in guess
about what the prehistoric phonetic value for that category might have
been, but this is _just_ a guess; the actual phonetic value is beyond the
reach of the Comparative Method.

To look at it another way, imagine that we have two languages, A and B.
Suppose that in every single case where A has dental [d], B has alveolar
[d], and vice versa.  From the standpoint of the Comparative Method, A and
B are the same language, because there have been no mergers in categories.
Strictly speaking, the phonetic realization of the categories is of no
consequence to the Comparative Method.

It is true that we do sometimes point out that if the phonetic values for
such-and-such categories were such-and-such, then such-and-such a rule is
phonetically plausible.  But this is guesswork on top of guesswork, and
it's really shaky ground to build an argument on.

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