A note on oblique case 125
morphological case features from the theory altogether. This is a conceptually desirable
move, given that these are notions that play no independent role in the syntax; they
are merely postulated in order to mediate between syntax and the range of actual case
morphemes. The core of Pesetsky’s proposal is that what traditionally are taken to be
various sui generis morphological case features can and should be re-analyzed as affixal
copies of particular independently needed syntactic (part-of-speech) categories. Thus,
nominative morphological case is claimed to reduce to an affixal copy of the category D,
accusative case to an affixal copy of the category V, genitive to N, and oblique case(s) to
the category P.
Given this background, we can proceed to consider now how the above Serbian/
Croatian case realization puzzle can resolve under the above view of morphological
case. This in turn would constitute empirical evidence in favor of the specific proposal
of reducing oblique case to (a copy of ) the syntactic category feature P.
Let us start with oblique complements of P. Adopting the above conception of the
nature and source of oblique case, the category P (with possible further features sub-
dividing it) is what gets copied onto the noun phrase complement of P, and constitutes
what conventionally is labeled “oblique” cases (such as instrumental, dative, locative).^4
Given this, one expects that as long as the category P itself is overt (i.e. non-null), it
on its own should suffice to satisfy the requirement for case realization (which in its
initial formulation, given in (9), was a principle artificially stipulating a limitation to
NPs governed by N and V). Accordingly, consider the following revised (generalized)
version of W&Z’s Case Realization Condition (9):
(14) The Generalized Case Realization Requirement:
(revision of (9))
Oblique cases must be overtly realized by some element of the assignment
domain (where “assignment domain” consists of the assigning head and the
assignee—its noun phrase complement).
So far we have suggested an account for P-governed oblique cases being apparently
“exempt” from the case realization requirement, which did not need to make the prob-
lematic claim that P is merely a case-marker rather than a head category that can be
lexically selected and that takes a noun phrase complement. The next question arising
- A possible objection to attributing the various instances of “oblique case” to the category
P can be raised based on the observation that the particular morphological case under dis-
cussion is non-uniform, i.e. instead of being a single case, oblique corresponds to different
(traditional) cases, such as “dative” vs. “instrumental” (and also “locative” and “prepositional”).
As noted by Pesetsky (2013), the expectation under the reduction-to-P proposal is that the
attested distinctions will turn out upon further investigation to correspond to independently
motivated distinctions among subtypes of the category P.