ABSTRACTS

Introduction: ‘Bilingual Conversation’ ten years after

Peter Auer - Universität Hamburg

Abstract not available.
Please refer to the volume presentation.

The ‘why’ and ‘how’ questions in the analysis
of conversational code-switching

Li Wei - University of Newcastle upon Tyne

One perennial issue in the study of code-switching is how to address the relationship between individual speakers’ interactional behaviour and the social contexts within which the interaction occurs. Two competing approaches, the broadly ethnographic approach and the conversation analytic approach, seem to offer very different solutions, often at the expense of each other. This article deals with the issue as it emerges in the methodological debate between conversation analysts and others, including the ethnomethodologically oriented ethnographers, over how to analyse and interpret the meaning of code-switching. Using conversational data collected from a Cantonese/English bilingual community, it is argued that one’s initial move should be to pay close attention to ‘how’ questions of bilingual interaction, addressing the interactional details as either the local production or the local enactment of contexts, but accepting in the meantime the utility of raising the ‘why’ questions once how questions have been dealt with.

1. Introduction. 2. Key claims of the CA approach to conversational code-switching. 3. Cantonese/English conversational code-switching: applying the CA approach. 4. Back to ‘Why’: integrating CA with Ethnography of Communication.

The conversational dimension in code-switching
between Italian and dialect in Sicily

Giovanna Alfonzetti - Università di Catania

Abstract not available at this point.
Please refer to the volume presentation for information.

Bilingual conversation strategies in Gibraltar

Melissa Moyer - Universitat Autònoma de Barcelona

The present chapter analyzes bilingual conversational activities in Gibraltar in relation to work carried out by Auer (1995) on the sequential organization of conversational code-switching, as well as by Muysken (1995a, 1995b) on proposals for a typology of sentential code-switching structures. Special attention is dedicated to the forms and meanings of English/Spanish code-switching in situated verbal interactions. The data considered were collected by Moyer from various social situations in Gibraltar between 1987 and 1992. Gibraltar is an ideal language laboratory for undertaking research on conversational code-switching. In this small British territory English and Spanish are used on a daily basis in almost every verbal interaction. The population has a balanced linguistic competence. The high level of proficiency in two languages explains the variety and richness of communicative strategies available to bilingual members of this community.

Bilingual conversations from Gibraltar are analyzed and explained in relation to language choices at three levels of conversation organization. These choices are strategies available for communicating meaning available to speakers with a high level of proficiency in both English and Spanish. The first choice involves the selection of a main or dominant language for the entire conversation. The main language serves to frame the code-switching choices made at the other two levels of conversation organization. At an intermediate level participants can momentarily switch their language for a limited number of turns. Language negotiation between participants takes place at this level. Meanings related to inter-turn language choices are recuperated by a sequential analysis along the lines proposed by Auer (1995). There are several intra-sentential switch types within a turn that a speaker fluent in English and Spanish can choose. Such structural switches function as contextualization cues for the creation of the situated meaning. These proposals are applied to several bilingual conversation extracts in an effort to show how contextualization cues combine with additional textual resources such as topic, alignment of speakers, and humor to represent the ambivalent and multifaceted identity of Gibraltarians.

The first section of the chapter offers background information about the community of Gibraltar together with some basic insights on the norms, values and attitudes of language use. Section two on conversational competence illustrates bilingual conversational competence in a simulated bilingual conversation where specific linguistic strategies are used in the creation of situated meaning. The next part, discusses a proposal for analyzing bilingual conversations in terms of languages choices which are made at three distinct levels (as outlined above) of a conversational event. Part four on bilingual conversation strategies presents three extracts which are discussed in relation to the three level model of language choice proposed. The last paragraph summarizes the main ideas about bilingual conversation set out in the present chapter.

References

Auer, P., 1995, The pragmatics of code-switching: A sequential approach. In: L. Milroy & P. Muysken (eds.), One speaker, two languages. Cambridge: Cambridge University Press, 115-136.

Muysken, P., 1995a, Code-switching and grammatical theory. In L. Milroy & P. Muysken (eds.), One speaker, two languages. Cambridge: Cambridge University Press, 177-198.

Muysken, P., 1995b, ‘Ottersum revisited: Style-shifting and code-switching’, Paper presented to the Dutch Sociolinguistics Conference, mimeo.

From ‘switching code’ to ‘codeswitching’:
Toward a reconceptualization of communicative codes

Celso Alvarez-Cáccamo - Universidade da Corunha

Code-switching research finds itself at a crossroads: On the one hand, ample research has shown that the alternate use of recognizably distinct speech varieties in discourse may have accountable meanings and effects. In this line, speech varieties have been associated with ‘codes’, and thus variety alternation is tantamount to ‘code-switching’.

On the other hand, some research has shown the impossibility or inappropriateness of assigning specific meanings to some types of variety alternation, which has thus implicitly started to question whether meaningless code-switching can be called code-switching at all (Alvarez-Cáccamo 1990; Stroud 1992; Swigart 1992). That is, if codes do not contrast, can we maintain that they are indeed distinct code s? Given the different natures of ‘unmarked’ and ‘marked’ code-switching, are we witnessing two distinct phenomena? Or is something missing in the way ‘code-switching’ is currently conceptualized?

This work is an attempt to shed light on the issue, by tracing back the origin and development of the notion of code-switching from its earliest formulations as connected to information theory (Fano 1950; Jakobson, Fant & Halle 1952; Jakobson 1961), bilingual contact studies (e.g. Haugen 19 50a, 1950b; Weinreich 1953; Vogt 1954; Diebold 1961; Gumperz 1962), and structural phonology (Fries & Pike 1949), to current conversational and anthropological work on code-switching (Auer 1984; Heller 1988; Eastman 1992). Briefly, contrary to Jakobson’s pioneering formulations on the communicative mechanism of ‘switching code’, the increasing merging of the notion s of communicative code and linguistic variety has been paralleled by the increasing lexicalization of the term codeswitching: in a sense, ‘codeswitching’ now subsumes and globalizes a number of possibly unrelated phenomena while excluding others which are clear candidates for being considered switches in communicative codes.

The arguments in this chapter are based both on theoretically-oriented works and on the discussion of selected cases from previous research, including data from my own work. The connecting thread in this work is the need to return to a communicative view of codes as systems of transduction between two sets of signals: on the one end, communicative intentions, and on the other end, linguistic-discursive forms amenable to interpretation. In this vein, it is suggested that the maintenance of the distinction between linguistic variety (in its broadest sense) and communicative code is crucial for explaining conversational conduct. Consequently, a parallel distinction between variety-alternation and code-switching is proposed. While linguistic varieties are held together by structural co-occurrence constraints, codes are held together by situational coherence constraints. Apparently ‘meaningless’ switches between varieties may therefore be simply explained as the outcome of communicative coherence constraints. Conversely, it is possible that code-switching be effected in monolingual discourse through the deployment of (lexical or non-lexical) re-contextualizing conversational tokens which signal shifts in activity type, alignments, footing, or frames.

At the broader sociological level, this approach points to the need to reassess the validity of outsiders’ accounts of the articulation between ‘code-switching’ and social identity. Formulations about ‘unmarked code-switching’ (Scotton 1988) as an index of an ‘ambiguous’ or ‘dual’ group identification may be a contradiction of terms. In this sense, it becomes necessar y to recapture Gumperz’s (1964) communicative view of supposedly code-switched discourse as a socio-functional, culturally determined ‘style’.

References

Alvarez Cáccamo, Celso. 1990. Rethinking conversational code-switching: Codes, speech varieties, and contextualization. In: Proceedings of the Sixteenth Annual Meeting of the Berkeley Linguistics Society, Feb ruary 16-19, 1990. Berkeley: Berkeley Linguistics Society, 3-16.

Auer, Peter. 1984. Bilingual conversation. Amsterdam/Philadelphia: John Benjamins.

Diebold, A. Richard. 1961. Incipient bilingualism. Language 37, 97-112.

Eastman, Carol M., ed. 1992. Codeswitching. Clevedon/Philadelphia/Adelaide: Multilingual Matters.

Fano, R. M. 1950. The information theory point of view in speech communication. Journal of the Acoustical Society of America 22.6, 691-696.

Fries, Charles C., and Kenneth L. Pike. 1949. Coexisting phonemic systems. Language 25.1, 29-50.

Gumperz, John J. 1962. Types of linguistic communities. Anthropological Linguistics 4.1, 28-40.

Gumperz, John J. 1964. Hindi-Punjabi code-switching in Delhi. In: H. Lunt, ed. Proceedings of the Ninth International Congress of Linguistics, Cambridge, Massachusetts, 1962. The Hague: Mouton, 1115-1124.

Haugen, Einar. 1950a. The analysis of linguistic borrowing. Language 26.2., 210-231.

Haugen, Einar. 1950b. Problems of bilingualism. Lingua 2.3., 271-290.

Heller, Monica, ed. 1988. Codeswitching: Anthropological and sociolinguistic perspectives. Berlin/New York/Amsterdam: Mouton de Gruyter.

Jakobson, Roman. 1961. Linguistics and communication theory. In Roman Jakobson, ed. On the structure of language and its mathematical aspects. Proceedings of the XIIth Symposium of Applied Mathematics [New York, 14- 15 April 1960]. Providence (R.I): American Mathematical Society, 245-252.

Jakobson, Roman, C. Gunnar M. Fant, and Morris Halle. 1952. Preliminaries to speech analysis: The distinctive features and their correlates. Cambridge (Mass.): The M.I.T. Press.

Scotton, Carol Myers. 1988. Code switching as indexical of social negotiations. In Heller 1988, 151-186.

Stroud, Christopher. 1992. The problem of intention and meaning in code-switching. Text 12, 127-155.

Swigart, Leigh. 1992. Two codes or one? The insiders’ view and the description of codeswitching in Dakar. In Eastman 1992, 83-102.

Vogt, Hans. 1954. Language contacts. Word 10.2-3, 365-374.

Weinreich, Uriel. 1953. Language in Contact. The Hague: Mouton.

Code-switching and the notion of code in linguistics:
Proposals for a model of a double-focussing speaker

Rita Franceschini - Universität Basel

The longstanding and multiplex discussion on how to draw a line between code-switching, code-mixing, borrowing, transfer, etc. indicates that the problem is a heuristic one, which links up to linguistic theory. Usually, the notion of a “code” or a “language” is presupposed in the literature on code-switching. However, seen from the perspective of the language user (and the intuitions of the members of a speech community), this presupposition may be questioned, as e.g. data from the African context suggest (cf. Meeuwis & Blommaert, in this volume). A conversation analytic approach is able to shift the perspective in fundamental ways: it is able to show that speakers separate linguistic systems in ways which do not agree with the categories of dominantly monolingual linguistics.

In this chapter, I argue that the way in which bilinguals converse (typically by code-switching) is not a “third” - exceptional - way (between the monolingual ways in the two languages), but the central one. The notion of a “code” will be grounded in the speakers’ multilingual practice. Seen as a simple continuum, the consistently monolingual ways of speaking would be the extremes, while code-switching (as well as other phenomena of language contact) might be located in the central area (which, in itself, would have to be conceptualized as multi-dimensional).

The view on code-switching I present has emerged from (and will be developed on the basis of) observations and data analyses in German-speaking Switzerland (where Italian, Italian dialect, Swiss German dialect and French are in contact). It has numerous points of convergence with work by R. Le Page and A. Tabouret-Keller on Creole languages and on Alsatian language contact.

The article uses two data sets. First, I focus on some prototypical instances of intra-sentential code-switching (Preziosa Di Quinzio 1992, Canegrati 1995) in order to demonstrate that a clear separation of “languages” is difficult. In the second part, I discuss phenomena of unexpected code-switching into Italian by Germans in conversation (inter-sentential switching); they are similar to those discussed by Rampton in the British context (in this volume). On the basis of these data, I address the question of how code-switching can be learned and can spread across ethnic boundaries, with a variable identity value. (This includes the question of how written forms of code-switching develop out of wide-spread oral switching.) Finally, code-switching as a sedimentation of multilingual practice may be found in word formation, which opens up a possibility to capture the diachronic dimension of switching.

I do not introduce new terminology for old phenomena in this paper. Instead, I discuss the “margins” of code-switching from the perspective of a highly multilingual situation. In addition, some links to theoretical linguistics and the investigation of language contact are made.

References

Canegrati, G. 1996. Code-switching a Toronto e a Monza: teorie e corpora a confronto. Unpublished thesis, University of Bergamo.

Preziosa Di Quinzio, I. 1992. “Teoreticamente la firma fa indietro”. Frammistione di italiano e schwyzertüsch nella conversazione di figli di emigrati. Unpublished M. Phil. thesis, University of Zürich.

A monolectal view of code-switching:
Layered code-switching among Zaireans in Belgium

Michael Meeuwis & Jan Blommaert - IprA Research Center, Antwerp

1. Introduction. This chapter focuses on the conversational alternation of two codes which are in themselves products of code-switching (CS). The material we use consists of transcribed conversations among Zaireans residing in Belgium.

Whereas most existing approaches to CS emphasize the presence of two languages in one stretch of speech, and thus identify ‘codes’ with languages, we will use a ‘monolectal’ approach, which is more in tune with speakers’ metalinguistic perceptions. In this approach, the CS variety is in itself seen as one single code, and within this code, semantic and pragmatic effects are based on code-internal variation (e.g. stylistic variation).

In section 2, a brief sketch of Zairean multilingualism is presented, involving an ethnographic assessment of the structure of speaker’s repertoires in terms of the accessibility and availability of codes. Section 3 elaborates the monolectal view of CS. In the fourth section, the discussion of the repertoire is recapitulated in terms of a monolectal perspective, and applied to instances of layered CS among Zaireans. It is shown that two monolectal CS variants are combined in one speech event so as to make up a hierarchical pattern of ‘CS within CS’.

2. Zairean multilingualism: a brief sociology of language. In addition to its official language, French, and its 220 or so vernacular languages, Zaire has four languages of wider distribution: Lingala, Kikongo, Swahili, and Tshiluba. These so-called ‘national languages’ each occupy a specific area of distribution and primarily function as languages of interethnic contact, although their number of native speakers is increasing. In spite of the geographical distribution, a number of social and political factors has caused that more than one of these languages are nowadays heard alongside each other in the same area. The enhanced accessibility of codes entailed by these processes of language spread has opened up the possibilities of CS. This is particularly salient in diaspora situations such as that of Zairean emigres in Belgium.

3. A monolectal view and its implications. Given the sociolinguistic background of Zairean émigrés, in particular with regard to their knowledge and use of French alongside Zairean languages, CS with French appears to be a naturalized speech commodity. Therefore, distinctions made from the linguist’s point of view (in the case of CS, the distinction of two or more ‘languages’) don’t appear to have the same relevance from the members’ vantage point. We present evidence for the claim that common-sense associations between languages and sociolinguistic indexical values do not hold. The data rather point in the direction of a speaker’s metalinguistic perception of CS speech as one language, putting greater relevance on intralectal variation such as style, accent, dialect, etc. In particular, the ‘quality’ of CS material seems to be of overriding importance in assessing speaker’s intended effects in conversation.

4. Layered code-switching. In the context of an enhanced accessibility of codes, monolectal CS variants can themselves become the objects of conversational switching. Interactional material involving CS between a Lingala/French monolectal CS variety and a Swahili/French monolectal CS variety is analyzed as to their societal anchoring and functions. It is shown that, from an ethnographic viewpoint, Lingala/French is the preferred code for multilingual situations involving speakers of more than one Zairean national language. Swahili/French CS only appears in situations where all speakers know Swahili. This hierarchy is a diaspora phenomenon, and relates to processes of identity construction among Zairean émigrés.

5. Conclusions. A brief summary of the fields is presented, and theoretical implications, especially regarding the status of a monolectal approach vis-à-vis a language-based approach, are discussed.

Discourse connectives in bilingual conversation:
The case of an emerging Italian-French mixed-code

Cecilia Ösch-Serra - Université de Neuchâtel

The study I am presenting analyses an interactive pattern which appears very frequently in the speech of Italian migrants in French speaking Switzerland. The speakers construct an original argumentative system on the basis of two monolingual systems, which is not identical to either of them.

In such a system there are three adversative connectives rather than two in monolingual Italian: ma (‘but’), però (‘but, however, yet’) or one in monolingual French, mais (‘but’). The system uses the adversative connectives in order to organize argumentative patterns with their own ru les for application. The patterns are variable, ranging from the most simple: the connection with a single mais or a single ma, to the most complex: the connection with the three connectives ma + mais + però. Moreover, each pattern pre-supposes the existence of the others and is in a relation o f intensity with them. A tendency is therefore revealed which sees the specialisation of the single ma as a speech marker and that of the single mais as an argumentative connective.

Conversational analysis of these occurrences reveals, however, that various factors -- such as the discursive function, the importance of the argument they introduce or the base language used at this point in speech -- may play an important role in their respective choice. Despite their difference, the isolated or joint use of connectives reveal common conversational functions. The organization of this argumentative structure, its frequency and its regularity provide additional arguments supporting the emergence of a mixed-code, which has no equivalent in the source languages.

On the transition from code-switching to a mixed code

Yael Maschler - The Hebrew University of Jerusalem

This paper is concerned with two types of language alternation phenomena in bilingual conversation and with transitions and interactions between them. On the one hand, we find the case of code-switching -- using two languages for ad hoc, interpretive purposes, as a typically bilingual contextualization cue (Gumperz 1982). On the other hand, we find the case of a mixed code-- using two languages such that a third, new code emerges, in which elements from the two languages are incorporated into a structurally definable pattern. First, I offer a diachronic perspective on language alternation phenomena by examining the emergence of a new, mixed code from ad hoc cases of code-switching. I then examine the interaction of this mixed code with ad hoc motivations for code-switching. An underlying theme throughout the study concerns criteria for deciding when a particular case of language alternation constitutes a code-switch and when it can be said to constitute part of a new-mixed code.

References

Gumperz, John J. 1982. Discourse Strategies. Cambridge: Cambridge University Press.

We, they and identity: Sequential vs. identity-related
explanation in code-switching

Mark Sebba - Lancaster University & Tony Wootton - University of York

This paper examines the role of code-switching in establishing and maintaining speakers’ identities in conversation, re-examining and re-interpreting Gumperz’s distinction between “we code” and “they code.” Applying Conversat ion Analysis techniques to a corpus of conversational data collected from adolescents bilingual in “London Jamaican” (Creole) and London English, the authors find evidence for both sequential and identity-related code-switching. However, they ar gue that it is necessary to go beyond fixed associations of “we” and “they” with particular codes to see that these associations are themselves negotiated and constructed in the interaction, drawing on cultural resources located both i nside and outside it. The notions of we and they are part of more local, and changeable, social identities which are made salient from time to time within a conversation.

Language crossing and the redefinition of reality

Ben Rampton - Thames Valley University London

Drawing on ethnographic research on multilingual adolescent friendship groups in the UK (Rampton 1995), this chapter begins by describing ‘language crossing’. Even though they don’t have any automatic right to them, youngsters often make spontaneous use of ethnic outgroup languages, and they manage to do so by crossing in ‘liminoid’ moments and events, when assumptions about “the world of daily life known in common with others and with others taken for granted” are interrupted (Turner 1982; G arfinkel 1984). In fact, crossing constitutes a complex symbolic dialogue about inherited ethnicities, and although it never finally abandones them, it continually denaturalises absolutist ideas about ethnicity being fixed at birth and in the early years of socialisation (cf. Gilroy 1987).

With this characterisation in place, the chapter then considers crossing’s implications for code-switching research more generally. It suggests that

(a) by focussing overwhelmingly on bilingual in-groups, research has tended to neglect the emergence of new plural ethnicities, built in an acceptance of old ones (Hall 1988);

(b) to gain any purchase on the exploration and/or renegotiation of reality - here, a reality of race stratification and division - full recognition needs to be given to Gumperz’ notion of metaphorical code-switching, though this needs some further clarification, perhaps most profitably by being drawn into close association with Bakhtin’s notion of double-voicing (1984:181-204);

(c) for the same reason, it would also be helpful if code-switching research relaxed its commitment to discovering coherence and systematicity in code-switching, and attended more closely to incongruity and contradiction. In the process, a clearer view would emerge of the (not infrequent) local occasions when code alternation no longer functions adequately as a contextualisation cue and instead becomes part of the ‘main action’, an object of explicit political dispute (cf. Goffman 1974:Ch7; Hewitt 1986:169, 181).

Conversation analysis has made a huge contribution to the study of code-switching. It is inclined, however, more to a celebration than critique of common sense (Thomson 1984:115-118), and so there are limits to its value in the analysis of interethnic processes in multilingual settings. In such settings, c oncentration on ordinary conversation can also be unduly restrictive, since it is often in the stylised performances studied in ethnography that participants engage in contestation and change (Bauman & Briggs 1990).

The last part of the chapter explores crossing’s role as a concept mediating between code-switching and second language learning and teaching. Crossing invites a certain amount of rethinking of the basic vocabulary of second language acquisition rese arch, and its identity as a ritualised speech activity suggests a number of problems in communicative language teaching methodology.

References

Bakhtin, M. 1984. Problems in Dostoevsky’s Poetics. Minneapolis: University of Minnesota Press.

Bauman, R., & C. Briggs. 1990. Poetics and performance as critical perspectives on language and social life. Annual Review of Anthropology 19, 59-88.

Garfinkel, H. 1984. Studies in Ethnomethodology. Oxford: Polity Press.

Gilroy, P. 1987. There Ain’t no Black in the Union Jack. London: Hutchinson.

Goffman, E. 1974. Frame Analysis. Harmondsworth: Penguin.

Hall, S. 1988. New ethnicities. ICA Documents 7, 27-31.

Hewitt, R. 1986. White Talk, Black Talk. Cambridge: Cambridge University Press.

Rampton, B. 1995. Crossing: Language and Ethnicity among Adolescents. London: Longman.

Thompson, J. 1984. Studies in the Theory of Ideology. Oxford: Polity Press.

Turner, V. 1982. Liminal to liminoid in play, flow and ritual. In: From Ritual to Theatre: The Human Seriousness of Play. New York: PAJ, 20-60.

Children’s acquisition of code-switching for power-wielding

Normann Jørgensen - University of Copenhagen

This chapter discusses bilingual children’s acquisition of code-switching as one of several means of administering power relations in face-to-face communication. Firstly I describe how bilingual children at an early stage can relate to differences in personal power, and behave linguistically according to these differences. The means to exercise power, and a brief outline of how the children develop power relations is my next step. Code-switching will be in focus in this context. Finally I compare excerpts from group conversations between the same children, but five years apart.

The material used in my paper comes from a longitudinal study of Turkish children’s bilingual development in the Danish public school system. The descriptive basis of the linguistic data is an ethnographic outline of the population concerned, and the situations in which the linguistic material is collected. The language use of the bilingual children is understood in light of the power relations represented in their immediate surroundings when they are at their school, and in society at large.

Nevertheless, our understanding of the children’s acquisition of turn-by-turn power wielding is also a question of small-scale analyses of their linguistic contributions to the conversations they are engaged in. The problems of interpreting the individual children’s development in the light of macro-power relations is one of the problems discussed in the chapter.

For instance, it can be shown that between grade 1 and grade 4, the percentage of Danish-matrix utterances used by the bilingual (Turkish/Danish) children in bilingual groups grows from almost zero to about one fourth. Some of the Danish-matrix utterances are clear indications of power oriented language choice which can be explained in terms of macro power relations in Danish society. But other utterances, and more of them the older the students become, seem to be just the opposite. In this chapter I speculate whether this may point to a development of a ‘neither-nor’ or a ‘both-and’ linguistic identity of the bilingual students, i.e. their “we” code is neither Danish nor Turkish, but both Danish und Turkish are “their” languages. The real “we” code may then be the pointedly bilingual languages use which employs code switching as one of its defining characteristics.

Perspectives on cultural variability of discourse
and some implications for code-switching

Christopher Stroud - Stockholms Universitet

This chapter explores the culturally specific social uses of conversational code-switching between Tok Pisin and the local vernacular, Taiap, found in a Papua New Guinean speech community. It does so by situating the study of conversational code-switching within an orientation to language that emphasizes the dialogic and situated co-construction of meaning and a concern for cultural variability.

The data for the study is taken from a set of case studies of a single community, the East Sepik village of Gapun in Papua New Guinea. More specifically, the vivid and volatile genre of the kros is taken as a focus for this analysis. Although superficially a disruptive speech event in that its purpose is to publicly proclaim anger, conflict and critique, distribute blame and elicit sympathy, it is in point of fact a highly significant semiotic event in Gapuners’ striving to establish new consensuses through the realignment and recontextualization of social relationships.

The analysis of code-switching is contextualized in an ethnographic account of the ways in which Gapun villagers conceive of social action, agency and knowledge, and discussed against the background of the implications these conceptions carry for villagers’ indigenous notions of personhood and intentionality, as well as for their ways of construing meanings in events, including linguistic events. Specifically, the chapter focusses on Gapuners’ indigenous sociocultural ideologies of social action and how they are articulated in language use.

In the context of the kros, strategic use of code-switched utterances occurs with rhetorical questions, irony, repetition and other “double-voiced” utterances (Bakhtin), that is, words that orientate towards the words of another ‘voice’ without necessarily subjugating them to the “ideologically controlling monologic voice”. Code-switching is viewed in this context as one type of the “double-voiced word”, thereby permitting ambiguity about whose voice is being heard.

By way of conclusion, the analysis provided here speaks to the need for ethnographically sensitive approaches to conversational code-switching.

Última alteração: 25-06-2003