Tense variation in narrative

TENSE VARIATION IN NARRATIVE Deborah Schiffrin University ofPennsylvania The narrative is a naturally bound unit ofdiscourse in which both formal and functional aspects of grammatical variation can be examined in a controlled and systematic way. This paper is a quantitative analysis of the past and the historical-present tenses as alternative ways of referring to past events in narrative. It shows how the organization of narrative delimits the area in which the historical present can occur, and how various structural and functional constraints restrict (or favor) switching between the two tenses. It also shows that the historical present evaluates narrative events because it is a use of the present tense, and that switching out of the historical present separates narrative events from each other.* The oral narrative is an ideal site for the quantitative study of variation in discourse ; because it is a naturally bound unit of discourse with a regular internal structure, both formal and functional aspects of variation can be examined in a controlled and systematic way. However, despite much recent work on how narratives are constructed and performed (e.g. Bennett 1977, Knapp 1976, Labov 1972, Polanyi 1979), there has been little use of quantitative methods in the analysis of grammatical variation within narrative. Thus we have, on the one hand, a well-developed methodology for analysing linguistic variation, and, on the other hand, a rich body of qualitative descriptions of narrative, but few studies which utilize the advantages of both approaches. This paper examines tense variation in narrative. The historical-present tense—the use of the present tense to refer to past events—alternates with the past tense in narrative. In the following fragment of narrative, events are understood as having occurred prior to the moment of speaking, regardless of the tense of the verb:1 * I am indebted to Anthony Kroch and William Labov for helpful comments on and criticisms of earlier drafts of this paper. Anne Bower, John Fought, Phyllis Nudelman, Ellen Prince, and Louis Scavo have also provided useful suggestions. An earlier and shorter version, using slightly different data, was presented at the 1978 LSA Annual Meeting, Boston. ' Narratives used in this study were primarily from sociolinguistic interviews conducted by members of a research project studying language change and variation in the Philadelphia speech community (NSF-75-00245, Principal Investigator: William Labov), and to a lesser extent from data collected by members of the Linguistics 560 class in 1976 and 1977. I thank everyone for allowing me access to their narratives. Special thanks go to Anne Bower, whose own interest in narratives no doubt contributed to her success in eliciting them—and, ofcourse, to William Labov, whose systematic collection of narratives made the job of looking for the HP much easier. Note that Wolfson 1976, 1978 has argued that narratives told during sociolinguistic interviews are not likely to contain the HP—that it is more likely when speaker and hearer share norms of interpretation and evaluation. As Anthony Kroch has suggested (p.c.), increased use of the HP in interview narratives can thus be seen as an indication ofbetter fieldwork methods, with interviews becoming more like natural conversational situations. For descriptions of sociolinguistic fieldwork methods in general, and of methods used in the Philadelphia project, see Labov et al. 1979. In all examples of excerpts from narratives, only clauses which represent narrative events (clauses in the complicating action section) are lettered. Each line of transcript with a letter preceding it contains one complicating action clause, and the letter indicates only the clause on that 45 46LANGUAGE, VOLUME 57, NUMBER 1 (1981) (1) a. Then all of a sudden everybody gets involved. b.and they made a mess. c.So uh ... this lady says ... uh this uh Bert, Oh, my son'll make them. He's an electrician.' d.So he makes them, e.and he charges all the neighbors twenty dollars a set, and there I paid three dollars. f.So I called her a crook. g.And I called her son a crook. h. So, they were really mad at me. Here clauses b and f-h contain past-tense verbs (P), while clauses a and c-e contain historical-present verbs...