Talking About This and That: Deictic Pronouns in Dravidian and Indo-aryan Languages

According to the most recent data in India there are at least 452 attested languages, excluding dialects and varieties. All these languages are grouped into four major families: Indo-aryan, Tibeto-Burman, Dravidian and Austro Asiatic. The goal of this dissertation is to account for the behavior of the pronominal systems of six of those languages, representative of 2 of major families, namely Dravidian and Indo-aryan:• Bangla • Hindi • Punjabi • Malayalam • Tamil• Telugu. What is remarkable about these languages (and in general about their language families) is that their pronominal systems are overtly marked by morphemes that realize what I labelled ”deictic features”. As the name suggests, deictic features have a clear relationship with the context of utterance. More specifically, the 3rd person pronouns of many Dravidian and Indo-aryan languages realize an overt morpheme that encodes the position of their referent in the Place of utterance. In other terms, the deictic pronouns that I am analyzing here realize the same features that are realized by demonstratives in languages like English or Italian. However, their behavior, in some specific linguistic contexts, deviates from what is normally observed for demonstratives. The recent typological and theoretical researches on Indo-aryan and Dravidian pronominal systems focus mainly on problems related to reflexivity (blocking effects, long distance binding, etc.), and little attention has been devoted to the relevance of deictic features for a theory of reference. As a matter of fact, even the basic dataset is not available. For this reason, the first phase of my research project focused on the design of a questionnaire that I submitted to native speakers during my fieldwork activity (hosted by the Central University of Hyderabad). The data set that I collected, which constitutes the real backbone of this dissertation, brought to light some previously unnoticed facts, like the behavior of some pronominal elements akin to the English demonstrative pronoun this (a so called “proximal element”) in bound variable reading contexts, that is consistently attested in the two unrelated language families. Some of the most significant consistencies observed in the sample data are successively discussed and, in part, accounted for. However, the language survey is the first on its genre, to my knowledge. As a consequence, many interesting aspects of the behavior of deictic pronouns remained unaccounted for and had to be left aside for future research. The dissertation is structured as follows. In Chapter 1 I provide some basic notions about the referential properties of pronouns and demonstratives to- gether with some facts about the pragmatic uses of demonstratives. In Chapter 2 I discuss some of the most relevant typological facts about the sample languages that are related to reference, such as word order, argument drop, pronominal morphology and distribution of pronouns and anaphors. Chapter 3 contains an exhaustive account of the data that I was able to gather during my fieldwork activity, accompanied by some preliminary generalizations about the classes of data. In Chapter 4, finally, I provide a discussion of some of the most relevant phenomena that resulted from the survey, such as the deictic features mismatches between pronouns and antecedents, and the behavior of proximal pronouns with respect to coreference and binding.