Next: What licenses the constructions?
Up: Manner of Motion Verbs
Previous: Construction Grammar
As has become apparent through the preceding
discussion, any solution to the problem of modeling the resultative
construction and the use of manner of motion verbs must meet several
criteria. These can be summarised as follows:
- Syntax: What licenses the construction must
be clearly identified, since their generation must be accounted for
as well as their interpretation.
- Syntactic constraints: The restriction of
the syntactic form of the ``unaccusative resultative construction''
(Levin and Rappaport Hovav 1995) to only certain classes of
unergative verbs, specifically the verbs of manner of motion and
verbs of sound emission, must be accounted for.
- Semantics: The appropriate semantics of
each of the variations of the resultative construction must be
captured.
- Conventionality and Pragmatic felicity: The
semantic restrictedness of the phenomena must be captured (i.e. an
explanation must be provided for why certain result phrases cannot
combine with certain verb phrases to produce a felicitous
resultative, and why semantically similar items cannot always be
felicitously substituted for one another in the resultative
constructions).
We will discuss each of these in more detail in the subsequent
sections, pointing out where the previous proposals fall short of the
criteria. In what follows, we isolate two main categories of
resultatives: what Levin and Rappaport Hovav (1995) call the
unaccusative resultative construction, in which an intransitive verb
is directly followed by a resultative phrase (see
Section 4.2), and the unergative resultative
construction, in which a resultative phrase appears after, and is
predicated of, a post-verbal NP (I will continue to use the
terminology for these constructions for ease of reference to the
syntactic structures, although I do not take on the notion implicit in
these terms that unaccusativity should play a part in the analysis).
The basic syntactic structures are therefore as shown in resalt1,
where NP 
is the noun
phrase of which the resultative phrase (ResP, either a PP or an AP) is
predicated. Each of these construction categories is further
subdivided in order to identify more specific properties of the
sentences which instantiate them (cf. differences identified in
Jackendoff 1990).
NP V ResP
V is a true unaccusative (e.g. freeze)
V is one of the manner of motion verbs or a verb of
sound emission
NP V NP ResP
NP is a fake reflexive,
coindexed with the subject NP, and V
is an unergative intransitive.
NP is the object of a
transitive verb
NP is neither subcategorised by
V nor a fake reflexive
Next: What licenses the constructions?
Up: Manner of Motion Verbs
Previous: Construction Grammar