jeff_l
(Jeff_L)
#1
12 2 Overview
verb-subject sequence, but not to a verb-object sequence. For more on the
role this plays in "case marking," see §2.3, below.
For verbs, the only true local dependency is that of an inflected verb on a
preceding Negative (8.b). This applies in all combinations where no elements
(other than floating clitics) intervene between the two, hence to perfective,
(long) imperfective, and imperative (i.e. prohibitive) stems. Basically the
modification converting Y to Y' adds phonetic content (where it is not
vacuous) in the perfective, reduces phonetic content (erasing a V-length ablaut
formative χ-pcl and an accent ablaut formative χ-pcl) in the imperfective and
prohibitive, and modifies the stem vocalism in all of these stems.
The cases in (8.c) are limited to definite relative clauses, which begin
with a definite demonstrative (in apposition to the head NP, which if overtly
expressed is external to the clause). The modification in ablaut form of the
verb involves erasure of a V-length ablaut formative (χ-pcl), and (with
specific exceptions) a shift in the location of the ablaut accent formative (in
one combination, this formative is entirely erased). The formulation of the
relevant rules requires an extraordinary interpenetration of "syntax" and
"phonology," and constitutes an important empirical argument against highly
modularized theories of grammar.
It is possible to combine two [X Y'...] configurations, as in the
combination [Negative verb noun(subject) ...]. In this combination, the
relevant pairs involving local dependencies are [Negative verb] and [verb
noun(subject)]. As a result, internal bracketing as in [Negative [verb
noun(subject)] ...] or [[Negative verb] noun(subject) ...] is counterproductive
(any such bracketing would appear to separate two forms that enter into a local
dependency). The only points that matter are adjacency (disregarding clitics)
and the grammatical relationship (e.g. Negative + verb, or verb +
noun(subject).
Superficially similar cases are those in (9)
(9) [X Y'...] Dependencies Not Requiring Adjacency
X Y' modification in Y'
Future particle verb Short Imperfective
complementizer verb Short Imperfective
The cases in (9) involve either a Future particle or a particular type of
complementizer, plus an inflected "Short Imperfective" form. The latter cannot
be used clause-initially with its usual subject pronominal inflections. However,
the short imperfective is related to the positive imperative (which has no
morphological connection to the prohibitive). The positive imperative is
clause-initial (hence is preceded by no "X" element), and it lacks the usual
subject pronominals.
