Sociolinguistics Soziolinguistik An International Handbook of the Science of Language and Society Ein internationales Handbuch zur Wissenscha.ft von Sprache und Gesellschaft 2nd completely revised and extended edition 2., vollsHindig neu bearbeitete und erweiterte Auflage

Edited by / Herausgegeben von Ulrich Ammon · Norbert Dittmar Klaus 1. Mattheier · Peter Trudgill

Volume 1 / 1. Teilband

Walter de Gruyter · Berlin · New York 8. Interactional Sociolinguistics/lnteraktionale Soziolinguistik tranSCfJ tants. C text, ho 1. Introduction: What is interactional example, Gumperz demonstrates that sys­ thing 0 sociolinguistics? tematically different ways of using language but rat 2. Theoretical underpinnings and tnethods to create and interpret meaning contribute view, sp 3. Key terms and concepts 4. Representative studies to employment discrimination against Lon­ guistic f .5. Literature (selected) don residents who were born and raised in (or "di: Pakistan, India or the West Indies. By ana­ cording lyzing interaction in employment interviews, systema 1. Introduction: What is interactional Gumperz follows Erickson's (1975) focus n1eanin sociolinguistics? on the linguistic basis of discrimination in one spe "gatekeeping encounters": speech events in ation 0 Interactional sociolinguistics (IS) is a type which a member of a minority community and sp' of qualitative analysis that sits at the inter­ requires the approval of a majority-commu­ (Ericks( section of linguistics, anthropology, and nity member to advance professionally or sociology. Rooted in the theoretical and me­ receive a societal benefit. It is the job of the 3. Ke thodological approaches developed by John interviewer to assess the interviewee's (}umperz, work in this area typically entails qualifications for the position or the benefit, ForGu case-study microanalysis of the language of but the assessment is made based on lin­ pretatic real interaction in the context of social rela­ guistic performance. Thus Erickson and an intel tionships. It is a tenet of IS that language Shultz (1982) argue that differences in con­ notion can only be studied in th1s context, and ventions for signalling listenership between tive pro further that the language of interaction is Americans of European and of African ongoin! constitutive of social relationships. The goal descent result in European-American school ing. Th of IS, then, is no less than accounting for the counselors "talking down" to African- Ame­ develop communication of meaning through lan­ rican students. In a similar vein, Tannen progres guage use. The word "use" is crucial. IS sees (1981, 1984) analyzes the role of what she vention ITleaning not as inherent in words, but rather calls conversational style in the perception pret wh as jointly created by speakers and listeners and stereotyping of New York Jewish speak­ what is in the act of using language to accomplish ers as "pushy" and aggressive. Maltz and anymoJ interactive goals. Borker (1982) and Tannen (1990, 1993) use linguist Gumperz' model of cross-<;ultural communi­ speaker 2, Theoretical underpinnings and cation to elucidate male/female cOffilnuni­ over, th necessa methods cation. In all these studies, close attention to miscommunication based on ways of speak­ verbal ( According to his own account (Gumperz ing serves the purpose of explaining social meanin 1992a, 1999), Gumperz' early research was phenomena. But analysis of miscoffitnuni­ cern, fc based on field work done in India, Norway, cation is also a heuristic device serving the progres and Central Europe during the 1950's and broader goal of IS to explain how meaning is guage r 1960's. The methodological approach and cOlnmunicated in verbal interaction. Mis­ cept of theoretical framework that became the communication makes visible elements of cursor foundation for IS was developed in the communicative practice that go unnoticed speaker 1970's, culminating in his 1982 book Dis­ when communication proceeds without per­ op and course Strategies. In subsequent essays (e.g. ceptible trouble. dual pI Gumperz 1992b, 1996) he elaborated and Whereas the grammatical sentence is the tion is i explained this approach and updated his ter­ unit of analysis in formal linguistics, for IS expecta minology in keeping with developments in the basic unit of analysis is a smaller entity: speech ( the field, but the underlying tenets and con­ an utterance alternately called a "breath ation pJ cepts remain constant. A collection ofessays group," "intonation unit," "idea unit," or or that edited by Gumperz (1982b) highlights a fun­ "inform-ation unit," characterized by an iden­ emotio] damental aspect of IS that is also found in tifiable intonational contour and typically mining the work of many of Gumperz' students and bounded by falling or otherwise phrase-final course. colleagues: the argument that differences in intonation, a pause, and/or a discourse going p linguistic phenomena partially account fOf, marker such as "but" or "and." The focus of resides or aggravate, social inequality, discrimi­ analysis is usually a spoken interchange every a~ nation, and cross-cultural stereotyping. For (audio- or video-taped and subsequently ends cn olinguistics 8. Interactional Sociolinguistics 77 k transcribed) involving two or more interac­ textually-sensitive presuppositions. The tants. Context is key to meaning in IS. Con­ signposts that signal meaning, create coher­ text, however, is conceptualized not as some­ ence (that is, show the relationship among that sys­ thing outside discourse that impinges on it ideas), and guide interpretation are contex­ ~ language but rather as created by discourse. In this tualization cues, defined by Gumperz (1999, ~ontribute view, speakers create context by means oflin­ 461) as "any verbal sign which when pro­ linst Lon­ guistic cues. Furthermore, verbal interaction cessed in co-occurrence with symbolic t raised in (or "discursive practice") proceeds not ac­ grammatical and lexical signs serves to con­ s. By ana­ cording to rules but rather by application of struct the contextual ground for situated in­ nterviews, systematic strategies or principles. Crucially, terpretation, and thereby affects how consti­ 175) focus meaning is not the single-handed creation of tuent messages are understood." In this ination in one speaker but the jointly- constructed cre­ framework, aspects of language which were events in ation of both interactants; indeed, listening previously devalued - dismissed as "margi­ )mmunity and speaking are inextricably intertwined nal" - become the focus of analysis; in other y-commu­ (Erickson 1986). words, they become "core." Key among con­ onally or textualization cues, as specified by Gumperz job of the 3. Key terms and concepts 1992b, 231, are: 1) prosody (including inton­ erviewee's ation, stress, and shifts in pitch and loud­ le benefit, For Gumperz, all meaning is "situated inter­ ness); 2) paralinguistic cues (including rate d on lin­ pretation." The conception of meaning as of speech, rhythm, pausing, hesitation, tone eson and an interpretive process is fundamental to his of voice, latching, and overlapping); 3) the ~s in con­ notion of conversational inference: the ac­ code- or style-switching that was a major ) between tive process by which a listener develops an contribution of Gumperz' early work (Blom , African ongoing interpretation of a speaker's mean­ and Gumperz 1972) - that is, the shifting be­ an school ing. This interpretive process is continually tween two languages, dialects or speech an- Ame­ developing and re-evaluated as discourse styles; and 4) choice of particular lexical , Tannen progresses. Shared contextualization con­ forms or culturally recognizable formulaic what she ventions enable a listener not only to inter­ expressions. erception pret what has been said but also to predict A central concept that both underlies the sh speak­ what is likely to be said. In other words, at theoretical and methodological basis of IS [altz and any moment in an interaction, a listener uses and also illustrates its interdisciplinary na­ 1993) use linguistic signals as guideposts to where the ture is framing. Anthropologist Gregory ommuni­ speaker is heading in the discourse. More­ Bateson [1955]1972 introduced the notion of ommuni­ over, the ability to make such predictions is frame to capture a fundamental aspect of ention to necessary to participate successfully in a meaning in interaction: No utterance or of speak­ verbal exchan8e. This developing thread of symbolic act can be understood except by ng social meaning that a listener must be able to dis­ reference to a "metamessage" that identifies onlmuni­ cern, follow, and predict is called thematic the character of the interaction. Bateson ex­ rving the progression. (Gumperz 1999 credits lan­ plains that while watching monkeys playing eaning is guage philosopher H. P. Grice's [1989] con­ in their cage at the zoo, he wondered how a )11. Mis­ cept of conversational cooperation as a pre­ monkey at play was able to understand a nents of cursor to his own understanding of how bite - an aggressive action - as a friendly ,nnoticed speakers and hearers work together to devel­ move. Bateson concluded that a biting mon­ lout per­ op and interpret thematic progression.) The key sends not only a n1essage - the bite - but dual process of interpretation and predic­ also, at the same time, a metamessage "this ce is the tion is also made possible by co-occurrence is play." One might say that the metamess­ ~S, for IS expectations. For example, members of a age is like an instruction guide to how the ~r entity: speech community know that a listing inton­ message is to be interpreted; that is the sense "breath ation presages more items to come in a list, in which it frames the message. A theory of luit," or or that a raised voice indicates anger - an framing as key to human interaction was aniden­ emotion that will guide the listener in deter­ elaborated by sociologist Erving Goffman typically mining how to understand the ensuing dis­ (1974). Both Bateson and Goffman were ase-final course. Conversational inference is an on­ concerned with the creation of nleaning in liscourse going process of interpreting meaning which hU1l1an interaction in general, not exclus­ focus of resides not only in the words spoken but in ively or even primarily in language, though ~rchange every aspect of how they are spoken. It dep­ in his last book Goffman (1981) began to equently ends crucially on culturally-shared and con- focus his analysis of franling on the parti­ 78 1. The Subject Matter of Sociolinguistics 8. In

cularities of talk. Insofar as speech is an es­ 3 R: So perhaps you could explain to Mr. and sential element of human interaction the C *apart from *that reason, *why such concept of framing is as crucial to ul;der­ else you want to apply for *electrical and standing meaning in language as itt is to work. in it other types of meaning. Parallel to the no­ 4 A: I think I like ... this job in my-, as a Gun tion of frame is Gumperz' speech activity: profession. real the goal-oriented character of an interac­ 5 C: And *why do you think you'll *like cour tion, by reference to which - and only by ref­ it? ers iJ erence to which - an utterance can be inter­ 6 A: Why? port preted. The direct influence on Gumperz is 7 C: Could you explain to me why? Sout more specifically Dell Hymes' (1974) pion­ 8 A: Why do I like it? I think it is more tomt eering work in the ethnography ofcommuni­ job *prospect. of th cation, especially his notion of a speech in tt event, the larger context in which a speech Throughout this interchange, the inter­ and activity takes place. Finally, Gumperz (1999, viewer and course instructor rely on cul­ tion~ 456) credits ethnomethodologist Harold turally-shared expectations for what kind perz Garfinkel (1967) with turning attention to of information should be exchanged in para the organization of everyday behavior and an interview, and signal to the applicant part to the "practical reasoning" by which inter­ what type of response they are looking for selec actants interpret each other's moves. Given through selective accenting of lexical iterns. ofwl these overarching theoretical conceptions Thus in (1) the emphasis on "trade" is an in­ mea1 and diverse influences, Gumperz pioneered vitation to the applicant to bolster his madt the sociolinguistic microanalysis of verbal chances for admission by enumerating and "gra' interaction as a way of discovering and explaining his qualifications, experience wore documenting how meaning is exchanged, and interest as they relate specifically to Asia created, and interpreted through language. electrical work. In lines 3, 5 and 7, the inter­ Thes viewers repeat and escalate their efforts suite to invite self-promotion on the part of the risin 4. Representative studies applicant, still relying on verbal accenting ques to call attention to the relevant words "Wo 4.1. Gumperz' analysis of interethnic ("apart from that, why else"). But the South tona communication in the UK Asian applicant is not alert to the British and Gumperz laid the cornerstone for his IS ap­ English speakers' contextualization cues ­ "Thi proach in field work conducted in Great Bri­ their use of pitch and amplitude to draw at­ playf tian (primarily London) examining intereth­ tention to key words in their questions and selve nic communication among speakers of statements. Furthermore, he does not share of dJ different varieties of English in workplace their expectations for interviews. In keeping viall settings, especially in gatekeeping en­ with his own sociocultural norms, he re­ supe counters such as employment interviews or gards self-promotion as inappropriate in how conversations between teachers and pros­ this context, since it risks the impression of suIte pective students in training courses. For excessive assertiveness. The applicant even... Asia example, in a selection interview for appli­ tually volunteers that he had completely whic cants to a British government-funded skills wired his own house in a way that passed in­ com~ training program Gumperz (1999, 466) spection, but even when this fact emerges, that shows that an applicant of South Asian ori­ he de-emphasizes rather than highlighting simp gin fails to recognize the implicit invitation (let alone exaggerating) the skill, knowledge in t1 to elaborate on the answers he has already and interest that this required. Gumperz' atiol provided on his written application. In the microanalysis of this tape-recorded and ilene transcription, R is the Interviewer, A the ap­ transcribed interaction is bolstered and spea plicant, and C the course instructor who is deepened by his knowledge of the grammar Sout also present at the interview. (An asterisk and interactional norms of both South of E marks accented words.) Asian and British speakers to demonstrate both that the outcome of this gatekeeping en­ pitel R: And you've put here, that you want counter was "interactively produced." and to apply for that course because Another classic study demonstrates the spok there are more jobs ... the *trade. notion of contextualization cues; their cen­ izatil 2 A: Yeah (low). tral role in affecting interactional outcomes; aept] guistics 8. Interactional Sociolinguistics 79

to Mr. and the inextricability of real-world issues and employment discrimination. It also il­ , *why such as job discrimination, on the one hand, lustrates the method by which key segments ~ctrical and understanding how meaning is created of an audiotaped interaction are replayed to in interaction on the other. In this study, participants in order to elicit their com­ Y-, as a Gumperz (1982a) uses the tape-recording of ments, which become another type of data real interaction to investigate charges and to be analyzed. 11 *like counter-charges among servers and custom­ ers in the employee cafeteria of a British air­ 4.2. Sociolinguistic subcultures in port, which had recently hired servers of American conversation ? South Asian background. The cafeteria cus­ An IS approach to communication among s more tomers complained of rudeness on the part speakers of divergent varieties of English of the South Asian servers, who complained has been applied to the conversations of in turn that they were victims of prejudice American-born conversationalists of differ­ inter­ and discrimination. By audiotaping interac­ ent ethnic and regional backgrounds. ~n cul­ tions between servers and customers, Gum­ Tannen (1984) examines a two-and-a-half t kind perz identified a small contrast in the use of hour conversation that took place at a ~ed in paralinguistic features that accounted in Thanksgiving dinner among six friends. On Jlicant part for both complaints. Customers who the basis of microanalysis of the audiotaped ng for selected a meat course were given the option and transcribed conversation, and bolstered items. ofwhether or not they wanted gravy on their by the subsequent comments made by par­ an in- meat. Both British and South Asian servers ticipants upon listening to segments of the er his made this offer by uttering a single word, tape recording, Tannen describes two con­ 19 and "gravy." But the British servers said this trasting conversational styles. Three speak­ ~rience word with rising intonation while the South ers of East European Jewish background lIly to Asian servers said it with falling intonation. who grew up in New York City use a high­ inter­ These contrastitlg intonational patterns re­ involvement style by which a speaker's prime ~fforts sulted in very different interpretations: The responsibility is to demonstrate enthusiasm, of the rising-intonation contour was heard as a interest, and conversational involvement. enting question, and consequently a polite offer: Some of the linguistic features that charac­ words "Would you like gravy?" while the falling-in­ terize this style include faster rate of speech; South tonation contour was heard as a statement, shorter or imperceptible inter-turn pauses ~ritish and consequently redundant and rude: (since silence is avoided as evidence of lack ;ues - "This is gravy." When the audiotape was of rapport); cooperative overlap and partici­ awat­ played for the cafeteria employees them­ patory listenership; expressive phonology lS and selves, they saw it as a self-evident example and extreme shifts in pitch and amplitude; share of discrimination because they were so ob­ more use of personal topics and narratives ~eping viously saying "the same thing." When their of personal experience; abrupt topic shifts; Ie re­ supervisor and English teacher explained and persistence in raising a new topic if it is lte in how the different intonation pattern re­ not immediately picked up. The other three ion of sulted in different "meaning," the South participants (two were raised Catholic in even­ Asian servers understood the reactions Southern California, a third is British) use a ~letely which had previously seemed to them in­ high-considerateness style. It was not poss­ ed in­ comprehensible, and the supervisors learned ible, however, based on this study, to de­ erges, that for them the falling intonation was scribe this style because the linguistic fea­ ~hting simply the normal way of asking a question tures of high-involvement style tended to 'ledge in that context. Because the use of inton­ interfere with the discursive practices of lperz' ational contours as a way of signalling pol­ high-considerateness features. The way this and iteness is conventionalized, it is obvious to happens, and the theoretical framework by and speakers of British English and invisible to which the pattern is analyzed, can be dem­ nmar South Asian speakers of a different variety onstrated with reference to interruption ­ iouth of English. This example, then, illustrates a linguistic phenomenon that has received ~trate both that contextualization cues such as a great deal of attention and analysis (eg g en- pitch and intonation are crucial to signalling James and Clarke 1993) - which is insepar­ and interpreting the meaning of words able from the related patterns of pitch, am­ s the spoken, and also that divergent contextual­ plitude, pacing, and co4 nversational rhythm. cen­ ization conventions contribute to the per­ Bennett (1981) points out that whereas over­ )mes; ception and the reality of social inequality lap - two voices speaking at once - is an ab­ 80 1. The Subject Matter of Sociolinguistics 8. lute) servable, descriptive linguistic category, in­ "You talking abollt me, or a deaf person," to (1) terruption is an interpretive category that which Deborah responds with increasingly implies a violation of speaking rights. fast pace and clipped syntax, "You. You.") Tannen (1984) describes segments of inter­ The result is that David's pace slows and he (2) DE action in which high- considerateness speak­ becomes inarticulate, whereas Deborah's (3) ers perceive themselves to be interrupted by pace quickens and she continues to overlap (4) De high-involvement speakers, but the overlap as she hastens to facilitate his reply by elab­ was not intended to be floor-stealing but orating on her question. It would be easy for (5) rather to be an encouraging show of interest an analyst to conclude that David is a con­ and attentive listening. In other words, for versational victim of Deborah's verbal ag­ high-involvement speakers, talking-along is gressiveness. In short, she interrupted him. (6) De often not interruption but rather cooper­ But Tannen contrasts this interchange (and ative overlap, a form of participatory listen­ others like it) with segments of the same Keye' ership. But for those who believe only one conversation in which a high-involvement ment ~ voice should be heard at a time, talking­ listener's overlaps have the effect of encour­ (like P along is always interruptive. The contrast aging rather than confusing a speaker - that the he~ becomes clear when segments of conver­ is, a high-involvement style speaker. In the ruptio sation in which three high-involvement following example (1984, 64) Deborah's to Del speakers participated are contrasted with "machine-gun questions" to Peter and Steve thermr segments of cross-style interaction. (In this are very similar to those she used with Debar study, as in many IS studies, the researcher is David, but the effect is very different. Again, grOWl also a participant.) For example, in the fol­ the conversation begins as a general dis­ a TV lowing exchange (1984, 69) 'Deborah's fast­ cussion of a phenomenon, the influence of Then} paced questions that come as overlaps create television on children. Steve remarks that answeJ stumbling, confusion, and a slowing of pace he believes the influence to be negative. De­ were y on the part of a high-considerateness speak­ borah then uses fast pacing (in this case ing "I er, David. David, a painter and sign-lan­ a latch - turn-taking with no perceptible Each ~ guage interpreter, has been explaining re­ interturn pause - rather than an overlap) and Ie. gIonal variation in signs, and has just to shift focus abruptly to Steve's (and his questi( demonstrated a sign for Christmas that de­ brother Peter's) personal experience. The reply t rives from Christmas decorations. Deborah question causes no disruption in the rhythm he swil overlaps to ask a question: of Steve's speech. He seamlessly finishes his in rer you .. lustrat (1) David: So: and this is the one that's Berkeley. This is (a fast the Berkeley ... sign for ... for [Christmas. or ave (2) Deborah: Do you figure out speakt those .. those urn correspondences? or do- when you ating ~ learn the signs, /does/ somebody tells you. astic 1 with, (3) David: Dh you mean [watching it? like (4) Deborah: Cause I can imagine knowing that sign, ... ment and not ... figuring out that it had anything to do with . rhythr with a the deconitions. stance rhythr (5) David: No. Y you kn6w that it has to do with the deconitions. ative ( such t Deborah's question in line 2 not only comes thought and then goes on to answer the device as an overlap but also abruptly changes the question. Because Steve and Peter are of the focus from sign language in, general to David brothers and the question addresses them perSOI as a sign-language learner in particular. both, they both respond. In the course of partic' the ov Tannen argues th~t David is caught off answering Deborah's first question, Peter guard by a range of features that character­ mentions that they had a television in "the descri ize this question as high-involvement: not quonset huts," thereby revealing that he and with0 1 only the overlap but also the rhythm and fast his brother lived for a time in the odd-look­ of the pacing and the shift of focus to David per­ ing temporary housing that was built for re­ of tht sonally. (A few turns later he asks explicitly turning veterans after World War II: notes olinguistics 8. Interactional Sociolinguistics 81

)erson," to (1) Steve: I think it's basically done ... damage to children. creasingly ....LThat what good it's done is ... outweighed by ... )U. You.") the damagel IWS and he (2) Deborah: Did you two grow up with television? Deborah's (3) Peter: Very little. We had a TV [ in the Quonset to overlap (4) Deborah: How old were you when your ly by elab­ parents got it? 1 Je easy for (5) Steve: We had a TV but we didn't watch it all I is a con­ the time: .... We were very young. 1 was four when my verbal ag­ parents got a TV. lpted him. 1 (6) Deborah: You were four? ange (and the same Key evidence of the logic of high-involve­ process ofconversational style as a system for volvement ment style lies in line 5. First, Steve's reply signalling interactive meaning as well as the )f encour­ (like Peter's in line 3) shows no evidence of consequences of conversational style as a tker - that the hesitation, false starts, and rhythmic dis­ source of systematic miscommunication that ~er. In the ruption that characterized David's reaction accounts in part for the stereotyping of New Deborah's to Deborah's machine-gun questions. Fur­ York speakers in general, and New York Jew­ and Steve thermore, Steve begins line 5 by answering ish speakers in particular, as "pushy" and Llsed with Deborah's first question (line 2: Did you two "aggressive." nt. Again, grow up with television?) by saying "We had Another study of subculturally-identifi­ neral dis­ a TV but we didn't watch it all the time." able discourse style in conversation that fluence of Then he proceeds with only a slight pause to also focuses on the conversation of East Larks that answer her second question (line 4: How old European Jewish Americans is Schiffrin's 'ative. De- were you when your parents got it?) by say­ (1984) analysis of "Jewish argument as so­ this case ing "I was faui when my parents got a TV." ciability." Just as Tannen shows that an os­ erceptible Each section of Steve's line 5 is syntactically tensibly non-cooperative strategy, conversa­ l overlap) and lexically tied to a separate machine-gun tional overlap, can have a cooperative intent (and his question. Steve begins "We had a TV" in and effect, Schiffrin demonstrates that :nce. The reply to the question "Did you two ...". Then for the speakers in her study, lower-middle­ Ie rhythm he switches to the first person "I was four ..." class Jewish Philadelphians, repeated dis­ nishes his in reply to the question "How old were agreement and competitive argument ac­ you ...?" These and many other examples il­ tually display the participants' solidarity lustrate that the same conversational strategy and reinforce their intimacy. Schiffrin (a fast-paced question that comes as a latch examines a series of conversations that took or overlap and abruptly switches focus to the place during sociolinguistic interviews she speaker personally) has a positive effect - cre­ conducted with a married couple, Zelda and ating shared rhythm and expressing enthusi­ Henry. In the following brief excerpt (321), astic listenership and interest - when used the topic of discussion is whether it is ap­ with a speaker who shares a high-involve­ propriate to prefer some relatives over ment style, but a negative effect (disrupting others. (It is not a coincidence, though rhythm and causing dysfluency) when used neither is it a focus of Schiffrin's analysis, with a high-considerateness speaker. In all in­ that many of the disagreeluents come as stances shown, overlap (with attendant pitch, overlaps.) rhythm and syntax) is intended as a cooper­ ative device; in one instance it is received as such but in the other it becomes a disruptive Henry: But don't they have a different attitude swer the device - an interruption. Crucially, the effect toward[the other? ~ter are of the overlap is the creation not only of one Zelda: Y- You're wrong Henry. :es them person - the overlapper - but rather of both Henry: Think about it. ourse of participants. Similarly, the interpretation of Zelda: No. You're wrong. Y- 1- look I have. U, Peter the overlap as an interruption may accurately the- I have uh f:four or five cousins l in "the describe the impression made on the speaker from one aunt! And I like Iny ­ t he and without accurately describing the intention Henry: you have- id-look­ of the listener - nor any inherent "meaning" Zelda: = cousin, one of thetTI better t4an the It for re­ of the linguistic behavior. Finally, Tannen [others! And ] that's how[it is: JYeh1 = notes that her analysis elucidates both the Henry: You do- you do- You do? 82 I. The Subject Matter of Sociolinguistics 8. Interf

After several more lines of dynan1ic dis­ Irene from Henry's potential criticism. In "frame agreement in this vein, Zelda suddenly says contrast, Henry's realignments are judgmen­ describ (overlapping Henry's prior turn), "Boy you tal, challenging, and divisive: They align him is impl need a shave so bad!" and Henry agrees with with the interviewer in opposition to Irene pret- a her criticism by saying (overlapping the end negatively evaluate her behavior, and gest th( of her statement) "I know I do. I like to take prompt her to reveal potentially compromis­ these s a shave every fOUf days!" Thus, Schiffrin ing information, although he too takes a with si1 shows, the interchange closes with a mixture protective stance toward her. For example, term in of humor, intimacy (a comment on bodily in answer to the interviewer's (Schiffrin's) ment-t< matters), and solidarity created through question about whether she'd be more likely on and rather than in spite of competitive disagree­ to discuss a problem with a friend or a In the ment (and, one might add, shared conversa­ family member, Irene says "I'm not that physicii tional rhythm and cooperative use of over­ family oriented. Like they are." Overlapping palsy it lap). the last phrase, Henry says, "She should be. identif) She's got a nice family." With this comment, charact 4.3. Framing Henry realigns himself with respect to his in­ with ar In analyzing the excerpt shown above, terlocutors not only by criticizing Irene's re­ examin Schiffrin describes Zelda's sudden reference sponse but also by addressing his assessment cian a) to Iienry needing a shave as a frame break to the interviewer rather than to her. Im­ standaI that glosses the preceding argument as non­ mediately after, Zelda intervenes to align what st serious. When disagreeing about attitudes herself with Irene by speaking for her: "Well ting rel toward cousins, Henry and Zelda are non­ she really- wait a minute she doesn't have residen: aligned, but when they agree that Henry sisters." Henry overlaps Zelda to clarify: tape; 2~ needs a shave, they are aligned. Thus Schif­ "She's got lovely children," and Irene over­ answer~ frin draws on Goffman's (1974) notion of laps "sisters" to agree with Zelda, "I'm not child's franling both to describe the shifting par­ talkin' about my children." (She goes on frame b ticipant structures by which Henry and to explain that a conflict arose between for eX3 Zelda display alignments and also to ident­ her husband and his brother when they were where t ify the frame by which their argument is to in business together.) Although the ways the vide be interpreted as non-serious (in Bateson's that Henry and Zelda speak for Irene thus same" 1 ternls, framed by a metamessage "this is differ and create differing alignments, frames. play"). nonetheless by speaking for Irene in the erallye In another study based on the same set of interview context, they both display and frame, 1 sociolinguistic interviews, Schiffrin uses and reinforce the closeness of their relationship about, develops the concept offraming to identify a to her and also transform the interview ear, an< conversational move she calls "speaking for frame. of the s another"- that is, voicing something about The linguistic creation and negotiation of to look someone else, in that person's presence, framing is the basis of Tannen and Wallat's ation i~ which only that person is in a position to (1993) analysis of interaction in a pediatric frame. , know. In an interchange among Zelda, examination/interview that took place and at least Henry, and their neighbor and friend Irene, was videotaped at Georgetown University examin( Schiffrin shows that speakers use "speaking Hospital's Child Development Center. This entertai for another" to accomplish frame shifts by study refines and expands a theory of fram­ latter e which participants are realigned with re­ ing in interaction and also demonstrates exampl, spect to each other, to the sociolinguistic in­ how an understanding of framing helps ac­ ach, th terviewer, and to their gender identities. In count for the participants' contrasting ex­ high-pi other words, macro-level relationships are perience of the same interaction - including butter: evidenced and created by micro-level lin­ why the physician's task is extremely chall­ laugh t] guistic strategies that align (a subtype of enging. Tannen and Wallat distinguish be­ exampl, frarnxng) participants in relation to each tween two types of framing and explicate doctor other. For example, when Irene rejects their interrelationship in interaction. Fram­ iug reg Henry's offer of a piece of candy, Zelda ing, in Bateson's and Goffman's sense, is there?" speaks for her in explaining, "She's on a the real-time exchange of metamessages by words 1 diet." Both Henry and Zelda speak for Irene which speakers communicate and interpret giggles (who is significantly younger than they), but how any utterance is meant. At the same child's Zelda's realignments are supportive and in­ time, researchers in psychology, artificial ation fr tegrative: by speaking for her, Zelda protects intelligence, and linguistics use the term encoun .nguistics 8. Interactional Sociolinguistics 83

;ism. In "frames" and the related term "scripts" to ance indicates that the latter frame "leaked" Idgmen­ describe the background knowledge that into the former. In this case, the encounter lign him is implicit in - and also necessary to inter­ proceeded without disruption, but in other o Irene, pret - any utterance. Tannen and Wallat sug­ instances a mismatch in knowledge schemas .r, and gest the term knowledge schemas to refer to triggers a shift in frames. For example, the promis­ these structures of expectation associated mother repeatedly expresses concern that takes a with situations, objects, people; they use the she fears her daughter is not getting enough xample, term interactive frames to refer to the mo­ oxygen because her breathing sounds noisy liffrin's) ment-ta-moment signalling of what is going at night. The mother refers to the sound of re likely on and how utterances are to be interpreted. her child's breathing as "wheezing." In re­ ld or a In the pediatric interview under analysis, a sponse, the doctor repeatedly reassures her Lot that physician examines a child with cerebral that the noise is not wheezing, which would 'lapping palsy in the mother's presence. The authors be evidence of difficulty breathing, but ould be. identify three distinct interactive frames that rather a normal symptom of cerebral palsy Imment, characterize the encounter, each associated by which muscle weakness results in "flop­ o his in­ with an interactive goal or activity: 1) an py" breathing. In the course of her examin­ ene's re­ examination frame by which the pediatri­ ation, the physician asks the child to breathe essment cian a) examines the child according to a deeply as she listens through a stethoscope. ler. 1m­ standard pediatric evaluation and b) relates This causes the child's breathing to take on o align what she is doing in a monotonous "repor­ the sound the mother has been asking about, r: "Well ting register" for the benefit of pediatric so she interrupts the examination to ob­ 1't have residents who might later view the video­ serve, "That's the particular noise she makes clarify: tape; 2) a consultation frame by which she when she sleeps." I use the verb "interrupt" Le over­ answers the mother's questions about her deliberately here, because the doctor ab­ I'm not child's health, and 3) a social encounter ruptly breaks the examination frame in ~oes on frame by which 'she manages the interaction, order to turn to the mother and once again )etween for example by telling mother and child reassure her that the noisy breathing is no­ ey were where to sit and monitoring the readiness of thing to worry about. (During playback, the e ways the videotape crew. What is ostensibly "the doctor noted that in order to avoid such in­ ne thus same" behavior can actually serve different terruptions of her examination, when she lments, frames. For example, the pediatrician gen­ sees patients in her private practice rather in the erally examines the child in the examination than the university clinic, she asks the adults ay and frame, but at one point the mother asks her accompanying a child to view the examin­ onship about a skin condition behind the child's ation through a one-way mirror and write terview ear, and the doctor interrupts the sequence down any questions they have to be address­ of the standard pediatric evaluation in order ed later.) , tion of to look at the child's ear. This act of examin­ Another study that demonstrates the Vallat's ation is therefore part of the consultation power of a framing approach combined ~diatric frame. At times the physician must balance with sociolinguistic microanalysis of inter­ ce and at least two frames, such as when she both action is Ribeiro's (1994) analysis of psy­ versity examines the child (examination frame) and chotic discourse. Comparing the intake and r. This entertains her (social encounter frame), the outtake interviews of a woman admitted to ~ fram­ latter enabling the former to proceed. For a psychiatric hospital in Rio de Janeiro, strates example, while examining the child's stom­ Brazil, Ribeiro shows that the woman's dis­ Ips ac­ ach, the doctor teases her by asking, in a course in the intake interview, which ap­ ng ex­ high-pitched teasing register, "Any peanut pears completely incoherent to a lay observ­ luding butter and jelly in here?" With a delighted er, can be seen as coherent when viewed chall­ laugh the child responds, "No." An amusing through a framing analysis. There are two sh be­ example of a leaky frame occurs when the frames operating in the interaction: In the plicate doctor says, in the same high- pitched teas­ interview frame, a psychiatrist is asking Fram­ ing register, "Is your spleen palpable over questions of the patient and the patient fails 1se, is there?" and the child, reacting not to the to answer them. In a psychotic crisis frame, ges by words but to how they were spoken, again the patient speaks to people who are not erpret giggles and laughs, "No." By referring to the present and as people who are not present same child's spleen being palpable (the examin­ (and perhaps not alive) or as herself at a dif.. ificial ation frame) in a teasing register (the social ferent age or in a different context {for term encounter frame), the pediatrician's utter­ example, herself as a child speaking to her 84 1. The Subject Matter of Sociolinguistics 8. Interac

mother or grandnlother). The psychotic dis­ dressing is 110t listening, whereas he is because course is incoherent in the interview frame, signalling only that he does not agree with wants to as the patient jumps from topic to topic, everything she has said. This silnple differ­ the intet chants and sings, and assumes different ence in use of mininlal responses (one might Dan'scoJ voices and footings. But it is coherent with­ call them contextualizatiol1 conventions) ex­ status: Jc in each psychotic frame in turn. Fur­ plains, in part, two common complaints: perior, a thermore, the patient makes accurate use of men's complaint that it is impossible to tell compute] knowledge schemas pertinent to each frame what women really think, and women's com­ nate. Wb such as the injunction not to make noise in plaint that men never listen. ments, h a hospital. Maltz and Borker's IS approach to cross­ with Dan Each of these studies both applies and ad­ gender communication provides the theor­ ning witl vances the theoretical concept of framing in etical foundation for Tannen's analysis of womenw sociolinguistic microanalysis of audio- or conversations between women and men as men who video-taped interaction. presented in the essays collected in Gender elements and Discourse as well as her books for gen­ linked: Je 4.4. Gender and Language eral audiences, You Just Don't Understand not only Among the essays by colleagues and stu­ and Talking from 9 to 5. In addition to the their rela1 dents that Gumperz collected in the volume conceptualization of cross-gender commu­ their con1 Language and Social Identity, the one that nication as cross-cultural, the notion of that men has had the most far-reaching influence is framing also elucidates male-female interac­ competen Maltz and Borker's (1982) application of tion. In the spirit by which IS resists the ane­ need no h Gumperz' IS framework to cross-gender ta-one correlation of a particular way of selves in ( comnlunication. After surveyin'g the find­ speaking with a single meaning, Tannen The co ings of research describing differences in (1994b), following Goffman (1977), suggests involves ways of speaking that typify boys and men that ways of speaking that pattern by gender mail cler on one hand and girls and women on are not sex-linked but rather sex-class other wo the other, Maltz and Barker suggest that linked. "Class" in this sense refers not to the Tina, the cross-sex communication can be understood notion of social class but rather to logical daughter as cross-cultural communication resulting types: the class of women and the class of who is ne in frequent miscommunication and eventual men, as distinguished from any individual temporar mutual negative stereotyping. In this view, woman or man. In other words, to say that a the conve "American men and women come from dif­ particular stylistic construction or conven­ research. ferent sociolinguistic subcultures, having tion is associated with the clas,s of women is sation tb learned to do different things with words in not to claim that any individual woman linked in a conversation, so that when they attempt to necessarily uses that convention. Tannen ar­ shopping carryon conversations with one another, gues, moreover, that framing is a way of ments ra even if both parties are attempting to treat serving the intertwined interactional goals says "Hi, one another as equals, cultural miscom­ of negotiating both status and connection. June's au munication results" (200). For example, She then contrasts two workplace examples getup. C prior research found that women tend to of small talk: one between two men, and an­ "that's CD utter nl0re minimal responses such as "yes," other among four women. In the first example, ment, "1 "mhm," and head nods. Maltz and Borker a subordinate (Dan) replies to his boss's then exp] claim that the significant difference is not in (John's) inquiry about his computer with the clothes a the number of minimal responses but rather colloquial complaint, "It sucks." John asks for But relati in the Ineanings these conversational signals specifics, then avers that he can fix it in "twen­ this exch: have for men and women. For women, mini­ ty five seconds." Disbelieving, Dan says, "I as the hig mal responses signal "I'm listening," where­ challenge you to do it." John responds, in a June into as for )men they signal "I agree," or at least mock-tough voice that frames his utterance as shift fron "1 follow your argument so far." As a result, non-serious, "You are a fool if you think you ing with 1 a man who receives a steady stream of head can challenge me, Mr. Computer!" Soon after, the mail nods and "mhm's" may logically conclude Dan, who knows that John has been suffering simultane that the woman he is talking to is agreeing from an intestinal ailment, shifts frames and action (l with everything he has said, whereas she is asks, "How are you feeling today, John?" In also cons demonstrating only that she is listening. response, John reports, in some detail, the Over, Junt Conversely, a woman who receives only oc­ symptoms of his intestinal distress, then adds, have the] casional nods and "mhm's" may under­ "It made me so glad that there was no girl­ key and standably conclude that the man she is ad­ friend around, nobody could take care me" manner. inguistics 8. Interactional Sociolinguistics 85

LS he is because he hates "being sick and somebody 4.5. Involvement strategies: Creating ;ree with wants to take care of me." Tannen argues that meaning in interaction le differ­ the interchange about John's ability to fix ne might Dan's computer reflects negotiation ofrelative Tannen (1989) details a theory of involve.. .ions) ex­ status: John is reinforcing his position as a su­ ment in discourse that is founded on the the.. nplaints: perior, and reinforcing Dan's position as a oretical tenets and analytic methods of IS. lIe to tell computer incompetent as well as a subordi­ She demonstrates that meaning is communi­ ~n's com­ nate. When he talks about his intestinal ail­ cated and interpreted in ordinary conver­ ments, however, John re-establishes rapport sation by means of linguistic strategies that to cross­ with Dan, and cements their solidarity by alig­ have traditionally been thought quintessen­ le theor­ ning with Dan man-ta-man in opposition to tially literary such as repetition, dialogue, alysis of women who annoyingly try to take care ofsick details, figures of speech, and patterned i men as men who only want to be left alone. All these rhythm. These strategies, which are shaped n Gender elements of the conversation are sex-class and elaborated in literary discourse, are per.. l for gen­ linked: John and Dan negotiate and reinforce vasive, spontaneous, and deeply functional 1derstand not only the differences in their status and in everyday conversation. Tannen calls them )0 to the their relative connection to each other but also "involvement strategies" because they draw commu­ their conformity to our culture's expectations on and simultaneously create interpersonal otion of that men talk about computers, demonstrate involvement. Details and dialogue combine e interac­ competence, engage in playful challenges, to suggest culturally recognizable scenes ; the one­ need no help even when sick, and define them­ (akin to frames) by reference to which utter.. r way of selves in opposition to women. ances are interpreted. Dialogue combines Tannen The conversation among four WOlnen also with repetition (including the repetition of suggests involves workers of varying rank: June, a prepatterned syntactic and intonational >Y gender mail clerk, ent~rs an office in which three contours) to create rhythms that involve in­ sex-class other women are engaged in conversation: terlocutors in rhythmic ensemble (Scallon lot to the Tina, the manager of the office and also the 1982; Auer, Couper-Kuhlen, and Muller o logical daughter of the company's owner; Heather, 1999) much as musicians create ensemble ; class of who is next in rank under Tina; and Janice, a when they play together. Together, the scenic ldividual temporary office worker who was recording and musical nature of speech, and the emo­ ;ay that a the conversation for her own sociolinguistic tions they evoke, are the dynamics by which conven­ research. From one point of view the conver­ linguistic strategies create meaning and in­ Nomen is sation that ensues is classically sex-class volvement in discourse. Repetition, dia­ l woman linked in that it is focused on clothing and logue, and details are the strategies Tannen lunen ar­ shopping and involves ritualized compli­ explicates in depth. Following Bolinger l way of ments rather than challenges. After June (1976), Hymes (1981), and Becker (1995), lal goals says "Hi," Tina calls admiring attention to Tannen sees patterned repetitions and luection. June's outfit: "Hey! Ah, we gotta see this contrasts as no less than the means by which ~xamples getup. Come on in." Tina compliments, speakers not only learn language but also , and an- "that's cute," and Heather echoes a compli­ structure language to convey meaning. She example, ment, "Love that beautiful blouse!" June illustrates the functions of repetition in the cis boss's then explains how she came to wear these production and comprehension of language with the clothes and where she bought the blouse. as well as in creating cohesion and maintain­ l1 asks for But relative status is revealed and created by ing interaction (for example, to get or keep in "twen­ this exchange as well. It is Tina's privilege, the floor, show listenership, stall, persuade, says, "1 as the highest-ranking participant, to invite gear up to speak, create humor, and so on). nds, in a June into the office and initiate the frame The multiple functions of repetition are il­ erance as shift from the discussion she had been hav­ lustrated in another segment from the hink you ing with her co-workers to one that involves Thanksgiving dinner conversation that pro­ )on after, the mail clerk. The complimenting ritual vided the excerpt presented in 4.2. In the suffering simultaneously includes June in the inter­ present example, Peter and Deborah are tmes and action (hence creating connection) and speaking with Chad, another participant ahn?" In also constitutes her subordinate role. More­ [raIn California. In discussing eating habits, ~tail, the over, June and Janice, the two women who Chad says "I go out a lot"; Deborah echoes, 1en adds, have the lowest status, speak in the lowest­ "1 go out and eat," and Peter responds, "You ; no girl­ key and paralinguistically most subdued go out?·" At the same time that all three align 3are me" manner. by addressing the topic of eating, they also 86 1. The Subject Matter of Sociolinguistics create and demonstrate involvement by re­ they are not reporting what in fact occurred peating the phrase, "go out." (The more but constructing dialogue to communicate common phrase that Deborah might have in the present. Communicating ideas by con­ uttered in a different context is "eat out"). structing dialogue creates involvelnent Peter then says, "The trouble with me is if I through its rhythmic, sonorous effect and by don't prepare and eat well, I eat a lot." This turning story into drama. Listeners then bel last phrase, "1 eat a lot," con1bines "a lot" come like members of the audience at a dra1 from Chad's opening statement "I go out a matic performance who use their own pas~ lot" and "eat" from Deborah's echo, "I go experience to interpret the meaning of out and eat." This is just a tiny exan1ple of events portrayed. This active participation what Tannen illustrates at length: that in senselnaking contributes to the creation through repetition discourse is woven from of involvement at the same that it relies on the thread of prior talk. Peter goes on to ex­ shared experience, or involvement, to pro­ SJ plain that if he takes the time to prepare vide meaning. Again, a single tiny example b himself a nice meal, he eats less than he can convey what Tannen illustrates with nu­ Ii} does if he eats whatever happens to be merous examples. In a lunch conversation, Ul available. (His actual utterance includes several women are discussing their experi­ ct more repetition: "And so if I'm just eating ences as mothers of children who are finally ro like cheese and crackers, I'll just stuff old enough to engage in play activities with­ atJ myself on cheese and crackers, but if I fix out an adult accompanying them: myself something nice, I don't 5. have to eat that n1uch." Debo­ DAISY The minute the kids get old enough to do these things rah's rejoinder and Peter's re­ themselves, Au! sponse to it evidence complex that's when Fra [ repetition: MARY "You do it yourself." and forc Deborah: Hmmm ... Well then DAISY Yeah that's when I start to say ... it works. Then it's a "Well ... I don't think I'll go in the water this time. Bate good idea. Why don't you kids go on the ferris wheel. and Peter: It's a good idea in I'll wave to you." Yod terms of eating. It's Beck not a good idea in terms of time. Mary provides a line of dialogue, "You do Tow! l~he bulk of this segment is made up of the it yourself," by way of listenership: She Benn repeated words and phrases "then," "it's a speaks as a mother talking ~o a child in just terpr good idea," and "in terms of." Peter high­ the constellation Daisy has set up. Then esses lights the contrast between eating, on the Daisy uses dialogue to instantiate a repre­ Blorn one hand, and time, on the other, by framing sentative situation, not a literally reported mean the contrasting terms in the repeated set-up one, as she moves from going "in the water" Norw sentences "a good idea in terms of." Fur­ to going"on the ferris wheeL" These speci­ perz, . therUl0re, whereas the message of his re­ fic instances of activities that children en­ Bolin! sponse to Deborah's comment is disagree­ joy, but which mothers are just as happy to in: Fo, forego, illustrate the larger point that the ment (she says it's a good idea and he is Brown saying it's not), this potential false note of mothers are jointly constructing in this Politen disagreement is overridden by the meta­ conversation. The lines of dialogue are hy­ Camhr message of ratification and involvement in pothetical, you might say, but by being pre­ his repetition of her words. That is, Peter rat­ sented as dialogue, they create a scene that Duran1 co-aut! ifies Deborah's contribution by stating his is recognizable to all the women present. cial Is disagreement in her frame - the words she Just as constructed dialogue conveys ideas 239-24' just spoke. In this way, repetition functions while also creating involvement, so do spe­ Erickso simultaneously on the four levels of produc­ cific details and the images they construct. the mel tion, comprehension, cohesion, and interac­ For example, in the preceding discussion of counter: tion. repetition, the point of Peter's utterance 44-70. Addressing dialogue as an involvement was paraphrased by using the general des­ -, (1986 strategy, Tannen suggests that the term "re­ cription, "whatever happens to be availa­ and Linj ported speech" is a misnomer. When speak­ ble." In the actual discourse, Peter used a Data, aT ers cast others' words in dialogue, or (as is specific instance, "cheese and crackers." Round 1 more often the case) cast their own thoughts You might say the general description is in: Tant as dialogue spoken by themselves or others, more specifically accurate, since Peter D. C., 2~ nguistics 8. Interactional Sociolinguistics 87

.ccurred clearly did not intend to say that cheese and -/Shultz, Jeffrey (1982) The Counselor ias Gate­ lunicate crackers are the only food he might eat keeper: Social Interaction in Interviews, New by con­ when he fails to prepare "something nice." York. Ivement But the specificity of "cheese and crackers" Garfinkel, Harold (1967) Studies in Ethnometh­ t and by makes it easier for a listener to visualize a odology, New York. then be­ scene of Peter eating too much of some­ Goffman, Erving [1964] (1972) "The n~glected It a dra­ thing that is not fully nourishing. So al­ situation. Language and social context", in: wn past though it is literally less accurate, it is more Giglioli, P. P., ed., Harmondsworth/Ba\1timore, tling of effective in communicating meaning by 61-66. ~ipation drawing on and simultaneously creating lis­ -, (1974) Frame Analysis, New York. creation teners' involvement. -, (1977) "The arrangement between the Sexes" 'elies on These representative studies are only a in: Theory and Society 4/3, 301-33]. to pro­ small number of the vast and ever-growing -, (1981) Forms of Talk, Philadelphia. ~xample body of work that uses interactional socio­ Goodwin, Charles (1981) Conversational !Organi­ Nith uu­ linguistic theory and method to deepen our zation: Interaction Between Speakers and Hearers, :rsatioll, understanding of how language works to ex­ New York. experi­ change meaning in interaction as well as the Grice, H. P. (1989) Studies in the Ways of Words, ~ finally role of language in constituting and negoti­ Cambridge. es with­ ating social relationships. Gumperz, John 1. (1982a) Discourse Strategies, Cambridge. 5. Literature (selected) ~e things -, ed., (1982b) Language and Socia/Identity, Cam­ bridge. Auer, Peter/Couper-Kuhlen, Elizabeth/Muller, Frank (1999) Language in Time: The Rhythm -, (1992a) "Contextualization revisited. The and Tempo of Spoken Interaction, New York/Ox­ contextualization of language", in: i Auer, ford. / P.ldi Luzio, A., eds., Amsterdam/Philadelphia, 39-53. time. Bateson, Gregory [1955] (1972) A Theory of Play and Fantasy. Steps to an Ecology of Mind, New -, (1992b) "Contextualization and understand­ York, 177-193. ing. Rethinking context: Language as an irtterac­ tive phenomenon", in: Duranti, A.lGoOdwin, Becker, A. L. (1995) Beyond Translation: Essays Ch., Cambridge, 229-252. V-ou do Toward a Modern Philology, Ann Arbor. relati~ity p: She -, (1996) "The linguistic and cultural of Bennett, Adrian (1981) "Interruptions and the in­ conversational inference. Rethinking linguistic in just terpretation of conversation", in: Discourse Proc­relativity", in: Gumperz, 1. J.lLevinson, 81. C., Then esses 4/2, 171-178. Cambridge, 374-406. repre­ Blorn, Jan-Petter/Gumperz, John 1. (1972) "Social -, (1999) "On interactional sociolinguistic meth­ ported meaning in linguistic structure: Code-switching in od. Talk, work' and institutional order: Discourse water" Norway", in: Directions in Sociolinguistics, Gum­ in medical, mediation and management settings", speci­ perz, 1. J.lHymes, D., New York. in: Sarangi, S.lRoberts, C., Berlin/New York, en en­ Bolinger, Dwight (1976) "Meaning and memory", 453-471. ppy to in: Forum Linguisticum 1/1, 1-14. Hymes, Dell (1974) Foundations in Sociolinguis­ at the tics: An Ethnographic Approach, Philadelphi:a. n this Brown, Penelope/Levinson, Stephen [1978] (1987) Politeness: Some Universals in Language Usage, -, (1981) "In vain I tried to tell you", in: EssiIys in re hy­ Cambridge. Native American Ethnopoetics, Philadelphia.: g pre­ e that Duranti, Alessandro (1986) "The audience as James, Deborah/Clarke, Sandra (1993) "Wo'men, men, and interruptions: A critical review", in: ·esent. co-author: An introduction", in: Text 6/3 [Spe­ cial Issue], Duranti, A.lBrenneis, D., eds., Gender and Conversational Interaction, Tannen, ideas 239-247. D., ed., Oxford/New York, 231-280. o spe­ Erickson, Frederick (1975) "Gatekeeping and Lakoff, Robin (1973) "The logic of politeness, or truct. minding your p's and q's", in: Papers from the ion of the melting pot: Interaction in counselling en­ counters", in: Harvard Educational Review 45/1, Ninth Regional Meeting ofthe Chicago Linguistics ranee 44-70. Society, 292-305. 1 des­ -, (1986) "Listening and Speaking. Languages Maltz, Daniel N.lBorker, Ruth A. (1982) "A:cul­ vaila­ tural approach to male-female miscommunica­ sed a and Linguistics: The Interdependence of Theory, Data, and Application", Georgetown University tion", in: Language and Social Identity, Gumperz, (ers." Round Table on Languages and Linguistics 1985, 1. l, ed., Cambridge, 196--216. on is in: Tannen, D.IAlatis, 1. E., eds., Washington, Philips, Susan U nnston (1983) The Invisible f;uI­ Peter D. C., 294-319. ture: Comrnunication in Classroom and Comnliuni­ 88 1. The Subject Matter of SOCIolinguistics 9. Discour ty on the Wann. Springs Indian Reservation, New , (1984) Conversational Style: Analyzing Talk conversa1 York/London. Anlong Friends, Norwood, N. 1. transcrib Ribeiro, Branca Telles (1994) Coherence in Psy­ -, (1989) Talking Voices: Repetition, Dialogue, and and spet chotIC Discourse, New York/Oxford. Imagery in Conversational Discourse, Cambridge. sations, Schiffrin, Deborah, (1984) HJewish argument as -, (1990) You Just Don't Understand: WOln.en and encounteJ sociability", in: Language in Society 13/3, Men in Conversation, New York. of analys 311-335. >-, ed., (1993) Fralning in Discourse, Oxford/New only frorr York. (analyses -, (1993) "'Speaking for another' in sociolinguis­ ica tic interviews: Alignments, identities, and -, (1994a) Talkingfrom. 9 to 5: Women and Men at pholog frames", in: Framing in Discourse, Tannen, D., Work, New York. pragmatic corporatiJ ed., New York/Oxford, 231-263. -, (1994b) Gender and Discourse, Oxford/New patterns ( Scallon, Ron (1982) "The rhythmic integration of York. ordinary talk. Analyzing discourse: Text and cations) t Tannen, Deborah/Wallat, Cynthia (1993) "Inter­ ways of sa talk", Georgetown University Round Table on active frames and knowledge schemas in interac­ t Languages and Linguistics 1981, in: Tannen, D., tion: Examples from a medical examinationlinter­ tering int ed., Washington, D. C., 335-349. view", in: Framing in Discourse, Tannen, D., ed., upon in d ~ -, (1981) "New York Jewish conversational New York/Oxford, 57-76. with) a style", in: International Journal ofthe Sociology of knowledge Language 30, 133-139. Deborah Tannen, Washington D. C. (U S. A.) search pre and meth( Kuhlen 2 2001, ScI (Ward anc 9. Discourse Analysis and Conversation Analysis (e.g. Chaft 2001), disc Diskurs und Konversationsanalyse 2001, WOt complianc! 1. Introduction cal, ideological)? Although linguists analyze (Ainswortl 2. Definitions: What is discourse? discourse at any - or all- of these levels, dis­ tural/inter( 3. Approaches to discourse analysis course analysts coming from the disciplines and ScoUe 4. Unity from diversity? noted above tend to focus primarily on the data, meth 5. Literature (selected) third level. be surprisiJ The variety of disciplinary c/ontributions tnultiplici t~ 1. Introduction to discourse analysis, as well as the three aries abou1 analytical levels, have the positive effect of This arti The growth of discourse analysis is well at­ extending the range of phenomena that can in discour~ tested 110t only in Linguistics and its sub­ be understood through an understanding several def field Sociolinguistics, but also within the of discourse. Yet they also have other con­ next, by various disciplines known collectively as the sequences that hinder the development of proaches tf human sciences (e. g. psychology, anthropol­ one single, united field of discourse analy­ conclusion ogy, sociology), the communication sciences sis. Not only do they encourage (indeed, ciples bro2 (e. g. rhetoric, communication), and the hu­ demand) attention to multiple sources and analysts to manities (e. g. history, literature). Assuming types of data, as well as the use of numer­ problems, for a moment that we define discourse as ous means of analysis, but they preclude types of da "language above the sentence" (but see 2.0), the availability - or appropriateness - of rrrethodolo! there are three levels at which the language focus on a single set of research problems of discourse can be analyzed. First is dis­ or development of only one discourse the­ 2-. Defini course as a linguistic text: what qualities and ory. properties distinguish a text from a random Represented in our recent handbook of Definitions set of sentences? Second is how text relates discourse analysis (Schiffrin, Tannen and help create, to lower level units: how is discourse (as Hamilton 2001), for example, are a variety described a paper~ either monologue or dialogue) related to of means of collecting data. Although vir­ sic clauses, sentences, and propositions? Third tually all discourse analysts avoid use of hy­ example, J 1~3) list 1( is discourse as part of other systems: how is pothetical utterance sequences, the means 10win'g: discourse embedded within, and partially of collecting instances of language use 1. langua constitutive of, larger domains of meaning include participant observation, accessing the cia (e. g. psychological, social, cultural, politi... on-line corpora (e. g. from the media, of