"Syntactic diffusion": changes that only make it part way

[Mark Davies]

(Cross-posted to HISTLING)

While carrying out research on diachronic Spanish clitic climbing, I've run
across some data that has me a bit confused.  Without going into all of the
details, what I've found is a syntactic change (a decrease in clitic
climbing), which I believe is motivated by a similar change in a related
construction (a shift towards INFinitive+CLitic in cases like _para hacerlo_
"in:order:to do:it), that moves strongly in one direction (decreased clitic
climbing), becomes the rule in about 80-90% of all cases, stops, and then
begins to reverse (i.e. a recent increase in clitic climbing).  (The data is
quite certain, being based on tens of thousands of examples from
computer-based corpora of 5,500,000 words of pre-1900 Spanish, and 3,500,000
words of post-1900 Spanish)

In the scenario I've constructed, the syntactic change I'm investigating is
active as long as there is continued change in the parallel construction
that I think is the motivation for the shift.  Once that parallel shift is
completed, the change I'm studying continues on for another 100 years or so,
in which time it becomes the rule in about 80-90% of all cases, then stops,
and has perhaps even begun to move in the opposite direction.  In my
scenario, there is another syntactic motivation that would account for the
reversal, if in fact it is taking place.

I'm somewhat familiar with the idea of "lexical diffusion" (as well as the
polemics surrounding such a concept), in which a phonetic shift is "active"
for a certain period of time and then ends, with the change having reached
only part of the relevant lexical items.  In this model the "unaffected"
lexical items may then undergo a later phonetic shift that only applies to
these lexical items that the first change never reached.

I'm wondering if there has been much research on this type of "diffusion" in
syntax, where a shift is "active" for a given time, then slows, stops, and
perhaps reverses when the motivating factor has ceased to exert its
influence.  Any ideas?

P.S.  If someone is interested in the details of the actual shifts in the
diachronic Spanish clitic climbing construction, I'd be happy to provide them.

Thanks in advance,

==================================================================
Mark Davies, Assistant Professor, Spanish Linguistics
Dept. of Foreign Languages, Illinois State University
Normal, IL 61790-4300

Voice:309/438-7975       email:mdavies at ilstu.edu
Fax:309/438-8038          http://www.ilstu.edu/~mdavies/welcome.htm
==================================================================