phonological change as lexical diffusion.

[Rankin, Robert L]

----------------------------Original message----------------------------
> 1. One of the contributions to this discussion raises some interesting
problems, in attempting to distinguish between 'proper' neogrammarian change
and 'secondary' change. I think this is a distinction without a difference:
any historian who works with large-scale corpora covering long time-periods
ought to be aware that neogrammarian change is not a 'kind' of change, but
simply the result of completed lexical diffusion.  (truncated)

This has certainly been the position of many text historians and
sociolinguists (including creolists) since Schuchardt.  But while extensive
corpora can provide insight into many aspects of language, they are
unreliable as a guide to sound change type and chronology, because of the
strong tendency for written language to preserve the spellings of individual
words.  Even corpora supposedly based on transcription (e.g., Bloomfield's
Algonquian texts or James Dorsey's Siouan ones) are poor at these tasks as
they have nearly always undergone normalization.  In other words, lexical
diffusion is certainly a good way to describe spelling change, but it leaves
open the question of phonological change, and especially its actuation
phase.

Practicing historical linguists have always found that they can routinely
distinguish among (1) change of articulatory habit, (2) borrowing, including
dialect borrowing or imitation, and (3) change effected in imitation of some
already-existing model.  These are not distinctions without a difference nor
do they depend exclusively on when the time line is intercepted:  they are
simply different sorts of human activity, some cognitive and some not.  The
first of them can often be seen in its clearest form in regular distribution
of allophonic features, something that is hard to show is the result of
lexical diffusion  (tho' one might additionally insist on change in
distinctive features as a part of a definition of real sound change).  The
second, borrowing, is a conscious or quasi-conscious process and may rely on
social factors.  The third matches what we traditionally call analogy and
may be stimulated by the first and/or second.  Historical linguists also
generally insist, quite rightly, on dealing with change in langue, not just
parole.  The boundary between the two became a bit muddy in the early
generative phonologies of the '60's and then again with attempts to make
phonetics more relevant to phonology in more recent times.  But the
langue/parole distinction is still an important part of understanding
language change.

Those specializing in language usage may well wish to ignore what virtually
all practicing historical linguists have traditionally seen as three
different sorts of change.  For them, these may not be interesting
distinctions, and they may wish to use a cover term like "lexical diffusion"
for some or all of them.  So be it, but lack of interest in individual
mechanisms or occasional difficulty in separating their effects does not
make them any less real.  That the three traditionally recognized origins
for change in pronunciation interact comes as no surprise either.  Labov's
recent work (since his 1980 LSA presidential address) does handle this in
depth, but it was discussed prominently not only by Sturtevant but by
Bloomfield in 1933:365ff.  In fact, one seldom has to go beyond Bloomfield
to find most of the points dealt with that many linguists have debated with
much sound and fury since.  But, as Bloomfield made clear, interaction among
mechanisms does not legitimize either confusing or deliberately lumping
them.

Bob Rankin
Linguistics Dept.
University of Kansas