Are Pronouns Always Zero in Zero Pronominal Languages? : The Case in Japanese

Language, Information and Computation(PACLIC 11),1996, 209-218 Are pronouns always zero in zero pronominal languages?: The case in Japanese Satoshi Uehara Wellesley College suehara@wellesley.edu Abstract This paper presents the results of a comparative text analysis of English anaphoric pronouns & their Japanese counterparts. It first demonstrates that although anaphoric pronouns in English are often compared to zero pronominals in Japanese, not all, but only some, of the English anaphoric pronouns have their counterparts in zero pronominals in the same context in Japanese, and then argues that discourse and perspective factors, which are both structurally prominent in Japanese, play a crucial role in anaphor resolution in the language. 1. Introduction So-called zero pronominal (i.e., unexpressed pronoun) languages like Chinese,Japanese and Korean are known to be highly context-dependent languages, and pronouns (or any clausal arguments recoverable from the context, for that matter) in these languages are said to be freely dropped (e.g. Sells 1989, Hudson 1994, for Japanese). This seems to suggest the idea that overt pronouns in English, more or less, correspond to zero pronominals in these languages. In the Centering approach to pronoun resolution in discourse, for example, Kameyama (1986), after setting up the rule for pronouns in English, states that the basic rule in Japanese "can be obtained by changing the word pronoun to zero pronominal" in the English rule although she proposes an additional rule for Japanese discourse. Some questions arise, however, at least on the simple, one-to- one matching between the anaphoric mechanisms of the two languages. First, if anaphoric pronouns in English always correspond to zero pronominals in the same context in Japanese, what is the function of the so-called pronouns in Japanese (e.g. kare 'he', kanojo 'she')? Similarly, how do those discourse functions expressed by the discourse particles like wa 'topic' in Japanese interact with zero pronominals of the language since such discourse functions take the form of [noun + particle], where the overt (pro)nominal form is required? Also, in terms of methodology, how do we know that the two relevant clausal arguments in English and Japanese are in the same discourse context? These questions are particularly important because most previous analyses of this issue have based their conclusions on 209 constructed discourse fragments and informants' out-of-context interpretations and acceptability judgments of them. The current research attempts to answer these questions, by using the method of text analysis for the collection of natural discourse data. To see the degree of correspondence between pronouns in English and zero pronominals in Japanese, I chose 0. Henry's famous story, The Last Leaf, and its Japanese translation by Yasuo Ohkubo, to see what those pronouns in the English original are translated into in the corresponding Japanese translation. Before starting the discussion, I would like to comment on the reasons for choosing the two texts for our data collection. The last Leaf by 0. Henry was chosen for the English original because the story has been popular among Japanese people and there have been several translations available in Japanese. Also the …

Content truncated to 500 words