Functional explanations for Subject to Subject Raising

[Jan Anward]

The figures for S-to-S raising in Spansih make sense if we assume, as
several people have,
e.g. Pawley, A. & Syder, F.H. 1983: Natural selection in syntax: notes on
   adaptive variation and change in vernacular and literary grammar,
   Journal of Pragmatics 7: 551-579,
that spoken language is essentially dialogic, even when produced by a
single speaker, and is subject to a one-clause-at-a-time constraint, which
I think may be grounded in the dialogic system for turn-taking (Sacks,
Schegloff, Jefferson's classical account in Lg 1974).
If spoken language obeys such a constraint, constructions which necessarily
span two clauses would tend to be avoided and alternatives - in case there
are alternatives - which span only one clause at a time would be preferred.
This also predicts that pseudo-raising, which preserves the integrity of
the second clause, should be preferred to regular raising.
A further prediction is that spoken language would not tend to avoid
constructions with simple predicatives after 'seem': He seems sad, and
similar examples. If Spanish has that kind of construction, it would be
interesting to test this prediction.

Jan Anward