| This
monograph offers an account of the assignment of antecedents
to some referentially dependent elements, such as lexical
anaphors and pronouns, and uses Hindi to demonstrate its
empirical consequences. It is in this specific sense that
the theory proposed here is a theory of some referentially
dependent elements of Hindi. In one way, the proposed
theory, chiefly concerned as it is with the assignment
of antecedent issue, does not distinguish between anaphors
and pronouns for this specific purpose; they both need
antecedents to be maximally usable in the (semantic) pragmatic
component in order that they are fully interpreted. In
another, it does distinguish between them, following the
main stream Principles and Parameters grammars; anaphor
has the feature "anaphoric" and lacks the feature
"pronominal" whereas pronoun lacks the feature
"anaphoric" and has the feature "pronominal".
The proposed theory therefore aims to make statements
about anaphors and pronouns that must directly reflect
the similarity between them as referentially dependent
elements requiring antecedents, and also the difference
between them, namely that one has features precisely the
converse of the other. The theory must also dispense with
a separate theory for the assignment of antecedents to
PRO, and must demonstrate that antecedent assignment for
PRO is a direct consequence of its constitutive features,
namely, both the anaphoric and the pronominal features.
Being a theory of referentially dependent elements, the
binding theory excludes any statement about R-expressions,
which are referentially independent. |
| The
attempt to explore an alternative to c-command to capture
the hierarchy among the arguments and the skepticism regarding
the delineation of the binding domain (needed by the binding
theory alone) using the notion of government lead to the
formulation of the binding theory without employing these
concepts, and the related one of grammatical function.
It may be noted in passing that although nothing in the
above prevents the account of the occurrence of PRO in
terms of government, for the rather obvious reason that
occurrence of, and antecedent assignment to, PRO are not
the same issue or related issues, the occurrence of PRO
has indeed been accounted for in this study without reference
to government. It has the desirable consequence of eliminating
a clearly unsatisfactory resolution of the so-called paradox
involving PRO. Thus the description of the relevant facts
of Hindi has totally dispensed with the notion of government
without any loss of either coverage or generality or elegance. |
| As
already mentioned, the earlier version of this monograph
was the doctoral dissertation "Theta Role in Syntax::
A Theory of Some Dependent Elements in Hindi" (Jain
1990). The current version retains the theoretical ideas
and the model and modifies the work only stylistically
(which includes modification in organization of the
same). The nineties have witnessed what can be called
the second phase of the Principles and Parameters Approach
in works such as "A Minimialist Programme for Linguistic
"Theory" (MPLT) (Chomsky, 1993), "Bare
Phrase Structure" (Chomsoky, 1994), and "The
Minimalist program (MP) (Chomsky, 1995). The Minimalist
model, although in a still evolving stage, is significantly
different from the LGB or KOL models. Notions such as
government, D- and S- structure, and indices all of
which are directly relevant to the theory of binding
have been dispensed with. Given these, we need to offer
an explanation for retaining the theory in essential
respects that was first spelt out about ten years ago.
As already mentioned Jain (1990) made no use of government.
It made no direct reference to where the binding theory
must apply; it made reference to notions such as agent,
theme, proposition, etc., from which it follows that
it has to apply at a level where such notions are available,
which is the LF level in terms of MP. In the interpretive
component, LF, the binding domain may be best conceptualized
in terms of proposition, which is a notion of the same
type as agents, theme, etc. rather than NP and S. Jain
(1990) does contain indices, but they are used merely
for the sake of readability, and not for any theoretical
purpose. This work contains a theory of binding which
can clearly be viewed as an interpretive theory of assignment
of antecedents to referentially dependent entities.
As regards identification of the antecedent is concerned,
Jain's 1990 work, like MP, makes use of hierarchy, but
captures hierarchy in a different way. In short, as
far as the relevant areas of grammar are concerned,
theoretical advances in the Minimalist enterprise do
not necessitate, in our view, and for our present purpose,
any substantive changes in the theory contained in Jain
(1990). If anything, this work can be seen as a modest
one that among many important others anticipated for
its own reasons certain grammatical thinking that finds
articulation in the second phase of the Principles and
Parameters Approach. |
| Abhilasha Jain obtained her PhD in Linguistics at IIT Kanpur (1991) and is presently teaching English at S.N.Sen Post-Graduate College at Kanpur . |
| B.N.Patnaik is a professor at IIT Kanpur, where he teaches generative linguistics and English. |
| The date of the manuscript: 2001 |
|
|
|
