No abstract

This Note examines colloquial uses of like as a discourse marker with the goal of specifying its discourse functions, including semantic, pragmatic, and sociopragmatic aspects of meaning. The data presented here come from nine interviews (Speakers 1±9), conducted by two female interviewers (A and B). The analysis examines the use of like by both interviewers and interviewees, and examines the eect of dierent speakers' roles and conversational goals on the use of like in these data.The use of like by the interviewers is examined across interviews to determine what factors give rise to variation in rates of like use in dierent interviews. The interactional factor which will be the focus of this analysis is the factor of speaker role, speci®cally, the roles enacted by the interviewers in interviews with nine dierent speakers. Unsurprisingly, the role of interviewer is played out in dierent ways in dierent interviews. Schirin (1996) describes the negotiation and construction of roles as a collaborative process:. . . roles are not viewed as properties of individuals alone: our roles and statuses are bound together by sets of reciprocal expectations and obligations about what to do, and by sets of reciprocal expectations and obligations about what to do, and about how and when to do it . . . Put dierently, who we are is sustained by our ongoing interactions with others, and the way we position ourselves in relation to those others. (Schirin 1996: 196±197) In these data, the interviewer enacted her role dierently in interactions in which the interviewee made only brief responses compared with interviews in which the interviewee spoke more freely. These dierences are apparent in rates of like usage in these data. DATA AND METHODOLOGYThe data for this study were collected as part of a larger project examining the use of discourse markers (DMs) in two speech contexts (an interview and a naturally occurring conversation) by both native and non-native speakers of English. The data involved in the current analysis are tape-recordings of nine

The analysis of letters written by 19th-century African Americans shows constraints on verbal -s marking which parallel those found in the writing of Scotch-Irish immigrants in the same time period and region, specifically a subject type constraint and a proximity to subject constraint. This correlation is highly suggestive for the study of the development of African American Vernacular English (AAVE). This study finds no support for a basis from a creole or from Standard English for AAVE in verbal concord and concludes that some, perhaps many, African Americans used varieties of English with little or no creole influence. Earlier studies have assumed that standard dialects of English constituted the superstrate in colonial and antebellum America; this analysis makes it clear that we must examine the features of the local varieties, black and white, before making any claims about the influences of language contact on a given variety. Further, the consistent patterns of inflections found in this study show that written documents, in particular letters written by semiliterate African Americans, are a good source for further linguistic study of 19th-century language.

No abstract

Different discourse markers (DMs) have been argued to have different levels of pragmatic detachability, based on whether they are lexical and content-oriented or nonlexical and operational in nature (Matras 1998). DMs that are nonlexical and operational in nature are claimed by Matras to be high in the pragmatic-detachability hierarchy and thus can be borrowed more easily in language contact. The present study addresses this claim with data from Pennsylvania German (PG) and analyzes discourse markers of both German and English origin. The findings support the claims about pragmatic detachability as a motivation for borrowing of DMs. The DMs well and so, which are highly pragmatically detachable, have been borrowed into PG and do not have German-origin counterparts. The DMs y'know and but, which rank somewhat lower on the pragmatic-detachability scale, have also been borrowed but are used in variation with their German-origin counterparts. Finally, the German-origin DMs ja and mol, which have no English counterparts, are still in use but are infrequent in these data. These DMs are very low on the pragmatic-detachability scale, indicating that not only borrowing but also loss may be controlled by the principle of pragmatic detachability.