Koodinvaihto ja keskustelun moniäänisyys

• Jyrki Kalliokoski

Avainsanat: keskusteluntutkimus, koodinvaihto, monikielisyys

Abstrakti

Codeswitching and polyphony in conversation (englanti)

1995 (99)

Jyrki Kalliokoski (University of Helsinki; fi)

Codeswitching and polyphony in conversation

The traditional approaches to codeswitching (henceforth: CS) can be roughly divided into three: grammatical, socio-linguistic and pragmatic. Grammatical approaches have concentrated on the morpho-syntactic constraints on CS. Sociolinguists have traditionally been interested in the situational aspects of CS and aimed at explaining language alternation by referring to social parameters of the speech situation. The present article discussess CS as a pragmatic, interactional phenomenon. Gumperz's (1982) idea of CS as a contextualization cue is taken as a starting point, but some conventional problems of the linguistic description of CS (CS. vs. borrowing, CS and diglossia, morpho-syntactic constraints on CS) and their suggested solutions are also presented briefly.

CS has often been treated as an exceptional linguistic phenomenon which causes problems for grammatical processing. By shifting the perspective to language use it is possible to view CS as an essential part of human communication instead of seeing it as deviance.

Many of the studies on CS deal with more or less balanced bilingual communities where the speakers are often fluent in both languages. In second language acquisition research, CS has been treated as a sign of lacking L2 competence. It has been classified as interference or as a communicating strategy.

The aim of the article is to enrich the picture of CS as a linguistic phenomenon. It attempts to combine a Gumperzian interactional sociolinguistic analysis of CS with some aspects of the markedness model of CS proposed by Myers-Scotton. On a larger scale, CS is analysed in respect to such notions as footing and participant framework (Goffman), activity type (Levinson 1979), dialogicity and intertextuality (Bakhtin, Voloshinov).

Several functions have been suggested for conversational, not-situationally constrained CS (eg. in Gumperz 1982). The article underlines the multifunctionality of CS (as any language phenomenon), and the impossibility of predetermining the interpretation of an individual CS. A contextual approach to CS is recommended: CS must be studied in real (spoken) interaction. The article is based on analyses of spoken data originating from different types of interactions in different speech communities. The problems of the interpretation of CS are illustrated in detailed analyses of six examples.

The contextualising function of CS can be traced in all kinds of interaction. The first example is a conversation between two native speakers and a non-native speaker of Finnish in Finland; the second and third examples are taken from a conversation between two bilingual Helsinki Finland-Swedes; the fourth example illustrates some functions of style shifting in monolingual narration in an experimental setting; the fifth excerpt is an example of style shifting in L2 (a Vietnamese speaker of Finnish) and the sixth example comes from a temporarily bilingual speech community: it is an excerpt from a conversation between Finnish tourists and their Estonian hosts in Tallinn.

The examples show that not all instances of CS can be automatically classified as communicative strategies, even in cases where the non-native's command of Finnish was bad. CS can be seen as a means of expressing the dialogicality of language, activating several contexts and several genres within a single conversation. CS occurs very often in sequences of (explicit or implicit) speech and thought representation, thus marking a change of footing and a change of the authorship (Goffman 1981) of the words produced by the current speaker. In terms of speech activity, CS can be seen as a signal of a change in the activity type.

These conclusions are based on a pilot study of a small corpus. The only way to gain more information about the pragmatic aspects of CS and its relationship with other linguistic phenomena is by looking at a great number of interactions in different kinds of speech communities and situations. The results of the present study have some implications for the study of second language acquisition and the study and teaching of intercultural communication. As it is not advisable to treat every instance of CS as a sign of the speaker's lacking L2 competence, CS should be seen as a proper linguistic resource, characteristic of both native speech and learner varieties of different levels of competence serving multiple stylistic and interactional purposes in everyday (intercultural) communication.