Back to list

Detail of contribution


Didn’t see that coming: an analysis of unexpressed subjects in English discourse

Abstract/Résumé: It is generally assumed that in non-pro-drop languages such as English, an overt noun phrase or pronoun is mandatory before an inflected verb; however, unexpressed subjects do occur in English, as illustrated in (1). 1) And yesterday was the first day she used it. Ø Put a bunch of stuff in it to read, Ø went home last night, and Ø couldn't get it open. (SBCSAE 35:549) Because they are rare in English, unexpressed subjects have been given little attention in linguistic literature. The few quantitative studies that have been done show several factors that promote their occurrence, including priming and IU initial position in the utterance (Travis & Torres Cacoullos, under review), and switch-reference and clause position (Harvie 1988). Studies on various registers, such as diary writing (Haegeman 2011), conversation (Leroux & Jarmasz 2006), and elicited narrative (Clancy 1980) reveal varying rates, contexts, and motivations for subject omission in English. The purpose of the current study is to identify factors that promote unexpressed subjects in conversational English. 724 instances of unexpressed subjects, extracted from the Santa Barbara Corpus of Spoken American English, provide evidence that they occur not only in conjoined clauses (the only context recognized by English grammars; cf.Quirk et al.1985), as in (2), but also in non-conjoined contexts, as in (3). 2) So the guy with the nice shirt..came in and Ø started talking to me, (SBCSAE 06:1015) 3) But he's the one that grabbed me, Ø kissed me, ...Ø took my underwear off, (SBCSAE 44:141) In fact, of the 724 tokens, 52 percent occurred in non-conjoined contexts. Other factors that were found to promote unexpressed subjects included subject continuity and low degrees of distance from the preceding mention of the same referent, a finding that is supported by both cross-linguistic research (cf. Travis 2007; Paredes Silva 1993) and literature on topic continuity and argument structure (cf. Givón 1983; Dubois 2003). There were also strong grammatical person effects. The most commonly omitted subjects were 1sg and animate 3sg. The 3rd person category, which was coded in the data for animate, inanimate, non-referential (dummy) 'it', and existential 'it', showed different patterning, according to subsequent verb types and modality, and included varying degrees of referentiality. That the majority of unexpressed subjects in these data exist contrary to the "rules" of grammar indicate that other interactional factors are at work here. This study ultimately suggests that overt subject omission in English warrants attention as a deliberate strategic device in conversation, where grammatical and discourse factors combine to determine patterns of use.