Identifying Some Inbuilt Problems in Anaphoric Interpretation

This paper1 explores an exciting and ever-developing sphere of linguistic research, namely anaphora. Since the inception of generative grammar, anaphora has not only become the dominating issue in syntactic investigation, but at the same time it has gained serious attention from the interdisciplinary researchers, specifically related to the field of philosophy, cognitive and computer sciences. Semantic and pragmatic approaches within linguistics are also becoming more and more relevant in explaining this phenomenon. The present paper attempts to identify some inbuilt anaphoric properties within intra- and inter-sentential domain, leading to the problems in anaphoric interpretation and resolution. Looking into the complexities becomes equally rewarding from the theoretical as well as computational points of view. Anaphoric resolution is at present a challenging area for understanding the problems and prospects of the artificially intelligent model approaching the optimistic design of man-machine interface. In the proto-typical situation, anaphor gets a straight- way linkage with its antecedent in a sentence or a clause. But such an ideal situation no longer prevails in some specific contexts. This paper investigates some of these non-prototypes from a broader generalized perspective without being restricted to the Chomskyan GB framework. The examples are taken mainly from Bengali and English.