From context to contextualization

Links & Letters 3, 1996 11-28 From Context to Contextualization Peter Auer Gerrnanisches Serninar University of Harnburg September 1995 Abstract It seems to be a cornmon-place enough statement to note that language depends on context. For example, under the notion of deixis, certain linguistic structures (i.e., deictics like I, here, or now) show a dependence on context that is thought to be essential. However, the relationship between language and context is far from clear and has been the centre of a debate which has gained mornenturn over the past ten years. The controversy begins with the question of how much in language and which parts of it are context-dependent, but it also, and perhaps more s~bstantiall~, includes the question of how the relationship between language and context should be conceptualized in more theoretical terms. In this paper, 1 will follow some lines of argument of this debate. 1 will start with the traditional notion of context-in-language and show that it is too restrictive by listing linguistic structures beyond deixis which must be interpreted with reference to context in order to be understood properly. 1 will also give a typology of contextual elements (Le., co-textual features, physical surroundings of the speech situation, social situation, participants common background knowledge and the channel or mediurn) which may play a role in understanding. In the second part of the paper, 1 will deal with the context-text link and 1 will try to show that the notion of contextualization is superior to that of context-Apendence to account in an adequate theoretical and empirical way for this link. Key words: Language and Context, Pragrnatics, Contextualization, Deixis. Table of Contents 1. Prelirninaries 5. The nature of the contextual link: 2. The traditional point the notion of contextualization of view: deixis in language 6. The semiotic nature 3. Context-bound linguistic elements 0f cOntextualization beyond deixis: some examples 7. Conclusion 4. Which contexts? References A preliminary typology 12 Links & Letters 3, 1996 Peter Auer 1. Preliminaries What is to be considered a context and what the text (mote genetally, thefocal event or the object of participanti attention') which the context surrounds, is a question that cannot be decided on the basis of objective facts since observables do not neatly categorize themselves under these two labels. Instead, seeing something as a foca1 event and other things as its context is already an interpretation of the perceived stimuli in somebody's environment. In otder to underline the perceptual and interpretive character offocal events and contexts, it has been proposed to conceive of them in terms of a fipre-ground relati~nship.~ Focal events as figures are perceived according to Goodwin and Duranti (1992) as «well outlined, sharply defined, and well articulated)), while contexts as gounds «appear far more arnorphous, problematic, and less stable)). Another metaphot well suited to highlight the interpretive aspect of the notion of context is Husserl's horizon (Sinnh~rizont)~ where the meaning of any event or thing cannot be understood by someone who does not properly take into account its horizon. The horizon itself dissolves as soon as we attempt to describe or analyze it; for anyone who tries to reach the horizon will only find himself in another situation which opens up yet another horizon as far out of reach as the original one. Both the figurelground and the horizon metaphor hold true for lay identification offocal events against their background or context, just as well as fot linguistic theories, which usually work out the details of the linguistic datum (thejgure), but gloss over the context (the ground in which it is embedded andlor from which it receives its particular interpretation. Any attempt to move a part of the ground or horizon into the focus of attention will necessarily have to see this focus against another gound or horizon in which it is now embedded, and so on. 2. The traditional point of view deWs in language Linguistic theories of context may be categorized along three dimensions: (a) according to the aspects of context believed to be relevant for a pragmatic analysis of language (henceforth called the indexed features), (b) according to the aspects of language believed to be subject to a context-bound interpretation or meaning-assignment (henceforth called indexicals), and finally, (c) according to the type of relationship which is believed to hold between the first and the second. Although these three dimensions are theoretically independent from each other, certain triples of indexed features, indexicals and conceptualizations of the relationship between the two have established themselves in the 1. The term is borrowed from Goodwin and Duranti (1992). 2. Goodwin and Duranti (1992: 10) and following pages. 3. Compare with Gadamer (196011972: 286 and following pages). From Context ro Conrexrualization Links & Letrers 3, 1996 13 history of the discipline. A particularly important triple consists in some feature of the physical surroundings here-and-now, such as speaker, hearer, time and place (egoltu, hic, nunc) as the indexed feature; some deictic elements of a language (such as here, lor now) as the indexicals (denotational indexical), and a unidirectional relationship between the two (i.e., the context determines the meaning of the linguistic utterance). These three dimensions have come to be associated with what could have been called representational theories of language. The triple represents the most narrow theory/theories of context in linguistics, but also the onels that has received most attention. One reason is that the relevance of context is confined to restricted areas of grammar from which it can be expelled by proper paraphrase. Schneider (1993) speaks of the semantization of pragmatics in this case, consisting in a translation of relevant aspects of context into expressions of the object language, which is then subject to non-pragmatic, or truth value semantics. A second reason has to do with the fact that only those linguistic utterances which cannot be assigned referential meaning unless their context-of-occurrence is taken into account are seen to be in need of a pragmatic analysis. Non-referential aspects of meaning are excluded; linguistic indexicals for these aspects of meaning are neglected. A third consideration is that the relevant indexed elements are looked upon as real world objects out there, to which deictic structures refer. As a consequence, context features are regarded as existent prior to and independent of speakers' linguistic activities in or relative to them. The in~ompatibilit~ of such a notion of context with the above-mentioned Gestalt approach is obvious.* 3. Context-bound linguistic elements beyond deixis: some exarnples One way to show that the previous approach to context is restrictive is by enumerating linguistic structures other than deictic expressions in their denotational function, which nevertheless index entities outside the &cal event. What immediately comes to mind here are systems of honorifics which, in many languages, relate to participants' social roles; here, we may include structurally simple systems such as forms of address or the TUIVOS pronominal distinction, but also elaborate systems such as those of Japanese or Javanese, which affect major parts of the grammar and lexicon. In this case of what is sometimes misleadingly5 called social deixis, it is not a denotatum in the real world out there which is indexed, but rather a perceived social relationship between 4. For a thorough critique of the narrow approach to context, the reader is referred to Schneider (1993), and Silverstein (1976, 1992). According to Silverstein, the privileged position of the narrow conscrual of context in linguistics is related to (and even a consequence 00 the semiotically based limits of [speakersy awareness which biases their metalinguistic abilities towards referential, segmental, and maximally creative features of language. 5. The term is misleading if the notion of deixis is restricted to a denotational or referential indexical. 14 Links & Letters 3, 1996 Peter Auer the speaker and the addressee, or the referent, or of al1 three. But, of course, not only honorifics are chosen relative to social role relationships. Variationists and interactional s~ciolin~uists as well as linguistic anthr~~ologists have accumulated evidence for the claim that variation permeates grammar, from phonetics up to turn-taking; this variation (including its ideological underpinning as part of a habitus in the sense of Bourdieu 1977) is partly an index of speakers' and recipients' social categories, and of the social relationship that holds between them. The selection of a variety from a repertoire - be it a style, register, dialect, vernacular, or language (code-switching) is subject to the same complex of context variables. Another large area of linguistic structure which eschews the narrow reading of context-dependence may be subsumed under the heading of subjectivity.' Contrary to the narrow reading of context in which speakers enter only to the degree that they fix the here-now-and-there for denotational action, the impact of the speaking subject under this view extends to how his or her lifeworld, likes and dislikes, identification with persons or events referred to, is reflected in and indexed by syntax and morphology, lexicon and prosody. This is particularly clear in the case ofwhat Jakobson (1971) has called evidentials, which are the grammatical or morphological means by which a speaker signals his or her commitment to the truth of a statement (e.g., the Turkish dubitativeverbal affixes). In addition, work by Kuno (1987:203ff) and others has demonstrated how the selection of certain syntactic constructions (such as passives, subordination, sentence mood) and lexical items (certain reciprocal verbs, certain verbs of motion, and so on) can be explained by reference to the speaker's empathy. The function of prosody, particularly intonation, to display the speaker's point of view has been acknowledged since the beginnings of modern linguistics (arnong many others, Voloshinov 19261 1976). Only recently has it been shown that this expression of subjectivity in language is not individualistic and unstructured, but follows recurrent, conventionalized patterns.7 In addition to the speaker's point of view, grammatical structure also depends on and indicates the recipient'spoint of vieW pragmatic distinctions such as the one between given and new information or that between theme and rheme, which have been shown to be central for word order and other syntactic phenomena such as left- and right-dislocations, capitalize on precisely this aspect of context. Finally, syntax is an index to co-participants' shared background knowledge. Fillmore, Kay, and O'Connor (1988) have demonstrated this link in their analysis of the conjunction let alone (e.g., I wouldn't hire Smith, let alone Jones), which construes a scalar model of interpretation in which the second proposition expresses the answer to a factual or hypothetical question, but the 6. The term aludes to Benveniste's (1958) subjectivitédam le langage. Present-day terms would be empathy, perspective or point of view. 7. Compare with recent work by Günthner (1996), Selting (1994) on the prosody of indignation, vidiculing, expressive assessments and similar emotional aspects of language. Frorn Contexr to Contexrualization Links & Lerters 3, 1996 15 first proposition establishes some point of comparison, which by presupposed common knowledge is superior to the second. Without the knowledge that Smith is quite an alcoholic, and Jones even more so, the conjunction could not be understood correctly. The same argument can be made for other parts of syntax. Dependence on shared knowledge is also found in the structure of the lexicon, where single lexical items point to others to which they are bound by cultural convention and with which they form a semantic field (Trier 1934). In the famous mini-story The baby cried, the mommy picked it up discussed by Sacks (1972), a correct understanding is only possible when mommy/baby, but also mommylpick up and babylcriedare seen as parts of a frame-like whole, such that mentioning one of them activates the other, or the first (category-mem- ber) activates the second (catego y-bound activity), respectively. The effectiveness and elegance of the working of such a membership categorizdtion device depends on knowledge about the set-up of afdmily; in a culture in which only grandparents take care of the children, its interpretation would be quite different from what it is in a Western cultural context.* The few examples given here may be sufficient to show that the relevance of contextual factors for the understanding of linguistic structures is not restricted to the case of deixis. When we move from grammar and lexicon to a broader (and indeed, prapatic) conception of language as social action, this relevance becomes even less disputable.9 It is here that the semantization of pragmatics has failed in particularly obvious ways. Early attempts to describe the meaning of speech acts by relating them to underlying performative verbs are generally dismissed as misleading and inadequate today. The meaning of an utterance qua social activity (Handlungj cannot be reduced to a speaker's mental state (intention) to perform such an activity; nor can it be dealt with by the semantic description of a peformative verb which seems to correspond to this mental state. Instead, it is the joint achievement of both the speaker and his or her recipientls, to make an utterance meaningfiil in its context-of-occu- rrence. In Voloshinov's words, such an activity is not simply fit into, the result of, or caused by its context: it resolves it (192611976: 100; also compare with his materialistic notion of dialogue, Voloshinov 19271 197 1). The most radical alternative to the semanticizing approach to context has been formulated by ethnomethodologists (Garfinkel 1967) who assume the indexicality of any linguistic (or other) activity to be obstinately unavoidable and irremediable, whatever remedial actions investigators may engage in (Garfinkel and Sacks 1970: 349). Although lay members -or professionalsmay, for some reason and for some purpose, formulate parts of an interaction, (i.e., they may say-in-so-many-words-what-we-are-doingj, these accounts them- 8. Compare with Bilrnes (1993) for lexical and grarnmatical implicature from an ethnomethodological point of view. 9. As an early transition from the semanticizing to the pragrnatic point of view, note Benveniste's (1970) notion of énonciation . 16 Links & Letters 3, 1996 Peter Auer selves display indexical features; in this way, context becomes relevant at different hierarchical levels of (meta-)linguistic action, but it can never be expelled from it. Accounts are always informed by their occasions of use. 4. Which contexts? A preliminary srpology If we are willing to accept a wide notion of context, it is useful to distinguish types of indexed entities in order to come to grips with the complexity of the signlcontext interface.'' In a pre-theoretical, but intuitively plausible way, five dimensions of context suggest themselves: (a) linguistic contexts (sometimes called co-texts), (b) non-linguistic sense-data in the surroundings of the linguistic activity (the situation in a physical sense), (c) features of the social situation, (d) features of participants' common background knowledge other than (a)-(c), and (e) the channel of communication (the medium). Links lbetween a linguistic sign and its co-textual features have been thoroughly studied as means for establishing textual cohesion (Halliday and Hasan 1976); here, anaphoric and cataphoric pro-forms play a decisive role." The long-standing linguistic interest in these textual functions may be a consequence of the fact that the linguistic means employed for them overlap considerably with those used for deixis.12 Another fundamental co-textual feature is the sequencing of conversational act(ivitie)s. Research in conversation analysis has shown that conversational activities (moves) prestructure (to different degrees) the following conversational slot with respect to speaker as well as activity selection. While adjacen ypairs represent a particularly strong kind of sequential link, other activities (e.g., first parts in action chains) leave more alternatives for the sequentially next activity open.13 In the case of cohesion, as well as in that of conversational sequencing, foca1 events are related to their co-texts by a relationship of (immediate or mediate) adjacency on the same hierarchical level of text structure. What represents a cotext for a given linguistic sign may also be located on a superordinate level of linguistic structure. This is the case when utterances are parts of larger speech activities, speech events, or genres. These larger events will then provide the context for thefocal event, which is embedded in them. For instance, 10. Various proposals have been made to list the different components of those aspects of context that may be relevant for language. Del1 Hymes' SPEAKING acronym has been one of the most influential ones (Hymes 1972); other influential ones are given by Halliday (e.g., Halliday and Hasan 1985, Blom and Gumperz 1972, more recently also by Goodwin and Duranti 1992 and Auer 1992). The selection and discussion of context types is necessarily restricted here to the most fundamental ones. 11. Of course, it is well known that anaphoric and cataphoric links between full forms and pro-forms are not always based on referential continuity. 12. For a discussion of the difference between deixis and anaphora, see Ehlich (1982). Ir should be noted that the parallel treatment of anaphora and cataphora is indicative of a planar, non-linear (and basically literate) visualization of language as a non-temporal, textual form. 13. See Schegloff and Sacks (1973), or Pomerantz (1975). Frorn Context to Contextualization Links & Letters 3, 1996 17 an utterance may be co-textually embedded as an orientation to a rtory. In this case, the superordinate co-text informs the organization and interpretation of the subordinate one, just as the latter contributes and, in a way, helps to achieve the first. A final component of co-text which brings us to the fringes of the linguistic dimension of context is given by the intertextual relationship between texts produced on different occasions. Following Bakhtin (1986), it is well known that texts often (or, in some theories, always) respond to prior texts, and, at the same time, anticipate subsequent ones. Indeed, some linguists have proposed seeing context as yet another collection of texts indexed by the foca1 text. While Bakhtin's notion of intertextuality includes sequentiality in the sense of conversation analysis, the more interesting aspect of intertextuality refers to distant text relationships across situations. Here, texts may relate to actual other texts by referring or quoting them; or they may index prior traditions of formal structures in text production, as in the case of re-uses or adaptations, changes or amalgamations of one or various genres (Briggs and Baumann 1992).14 The second dimension of context is given by the physical surroundings of the speech situation, (Le., the things and events in the co-partici ants' sensual, particularly visual, reach). Everything that can be pointed to,lPinc1uding time, may become an indexed feature of a deictic expression. The second dimension of context therefore seems to be directly linked to the narrow construal of context. There is, however, an alternative tradition to this rather static approach to the situational environment of speech: Malinowski (1926) first drew linguist-ethnographers' attention to a language that does not have the dignity of many written texts (i.e., being detached from the social activities of everyday life) but which is part of a stream of verbal and non-verbal activities, both of which are intertwined and depend on each other for their interpretation. His famous description of the Trobriand islanders coming back into the lagoon after a fishing expedition gives an example of such hnguage in action (where the in refers both literally and idiomatically to action). Here, the verbal components of the situation as it develops in time are certainly not autonomous; and their relationship to the context-of-situation is far more intricate than could be analyzed on the basis of deixis alone. In fact, the verbal components are often only secondary -les essential to, less constitutive of the action than the non-verbal ones. Nevertheless, they may take on decisive importance at some points. Bühler (1 934: 154ff), who elaborated on this empractic use of language from a more linguistic perspective in his analysis of rituatio- 14. For further reading on the link benveen intertextuality and the construction of discourses, see Fairclough (1992). 15. To speak ofpointingin this case obviously requires a rather loose usage of the term, including metaphoricai extensions not present in everyday language. Since Bühler (1934), pointing gestures such as the voice of the speaker, eye-movements or body orientation are accepted parts of deixis. 18 Links & Letters 3, 1996 Peter Auer nal ellipsis (~resumabl~ without knowing Malinowski's work), aptly calls them diacritics on non-verbal activities. As a third dimension of context, the social situation was mentioned. It includes the constellation of participants, their social roles and the social activity they are engaged in. The analysis of the different alignments a co-partici- pant may establish with a particular linguistic utterance (i.e., his or her participant role) is one of the main topics in Goffman's work. For Goffman, a socialsituation is an environment of mutual monitoringpossibilities within a gathering (1964: 135). Within such a social situation, it is not enough to distinguish speaker and hearer, as used to be done in the traditional, cybernetically based models of communication. Instead, Goffman distinguishes, on the production side, between an animator who is the sound box for the message, an authorwho is responsible for its wording, and aprincipal apariy to whoseposition the wordr attest (Goffman 1979). On the reception side, the dddressed recipient and unaddressed recipients are ratified participants to an encounter, while over-hearers (bystanders) and eavesdroppers are non-ratified listeners of other people's encounters.16 Which participant role a person is in provides a context for how this person is permitted to act. While Goffman's approach is restricted to the realm of what he calls the interaction order, other ethnogaphers and linguists (e.g., in the tradition of the ethnogmphy of communication) would include participants' interactional and social roles and the type of speech event (e.g., medical consultation, birthday party, telephone enquiry) into a definition of the social situation as well. Interactional roles may be a function of the speech event; for instance, a medical consultation requires participants to take over, at least temporarily, the roles of doctor and patient. Other (aspects of one's) social roles, which tend to be trans-situationally more stable and which are not eo +o bound to the type of speech event co-participants are engaged in, are social class, caste, ethnic affi- Iiation, gender or age. The fourth dimension of context -that of participants' common background knowledgeis of complexity. Research on this dimension may be located in the tradition of phenomenological approaches to the structure of the lifeworld, the essential structural principles of which have been outlined in Alfred Schütz' work;17 it has also been elaborated on in the tradition of formal pragmatics and presupposition theory (Sperber and Wilson 1986), and more recently, there have been attempts to formalize this knowledge in artificial intelligence.18 There is an obvious overlap with the previous dimensions. What has been mentioned before in a text may become an indexed feature of the co-text of a later utterance; at the same time, it is part of the situation-specific common 16. Follow-up work on these distinctions can be found in Goodwin, Ch. (1984) , Goodwin, M.H. (1990) and Levinson (1988). 17. See for example Schütz & Luckmann (1976). 18. Reichman (1984), or Putnam (1988). From Context to Contextualizarion Links & Letters 3, 1996 19 background knowledge participants may rely on in the production and interpretation of future activities. Similarly, social roles can only become visibly relevant for an interaction because their attributes, including rules of linguistic conduct, are part of participants' shared knowledge. Thus, underlining the knowledge aspect is sometimes just another perspective on context which focuses, not so much on objective facts as indexed objects, but rather on (inter) subjective interpretations and typifications. From this perspective, a useful distinction is one that relies on the reach or domain (Gültigkeitsbereich) of a particular piece of knowledge.19 Knowledge is accumulated between participants during a particular interactive episode; this very specific knowledge may be partly forgotten after the episode, or it may be partly transferred to a stock of knowledge which accumulates between these same participants in the course of their histo y of interaction. A larger Gültkkeitsbereich is involved when knowledge which is characteristic to a certain profession (reflected, for instance, in a professional code or register), a neighbourhood, a sub-culture, becomes a relevant context of interaction. Finally, knowledge on how to behave properly within a given (ideal) community which is shared by al1 its members may be invoked for the understanding of afocal event. Here, we reach the maximal domain within which knowledge is shared among participants, (i.e., that of common culture). In the latter domain, looking at participants' background knowledge is not simply a different way of looking at the same indexed elements, but it covers an additional range of phenomena. The culture perspective is a central component of the FirthIHalliday tradition of linguistic research, but also of the ethno aphy of communication and other branches of anthr~polo~ical linguisiY Attempts have been made to formalize restricted components of this knowledge, using notions of scberna, script, or &me. The final dimension of context is that of the channel or medium in which the interaction takes place. For many, including the Westem cultures, the technology that has had the most impact on language is writing (see Ong 1982). The influence of modern or recent technologies -such as telephone, telegramme, e-mail, automatic answering machinesis only beginning to be investigated. 19. See Kjolseth's distinction between background foreground emergentgroundc and transcendent groztndc (1972). 20. Apart frorn earlier treatrnents in the Humboldtian tradition, it is once more Malinowski wLose ethnographic view oflanguage was a breakthrough towards the view on language that takes culture seriously. (He clairned that language is essentially rooted in the reality of the culture, the tribal life and custorns of a pe&le:and that it &nnot be explained without constant reference to these broader contexts of verbal utterance (Malinowski 1926:305).) See for exarnple Halliday and Hasan (1985) for an overview of this tradition; for anthropological approaches to cultural contexts see Geertz (1973). Irnportant contributions in rnodern linguistic anthropology towards a better understanding of cultural contexts and their relation to linguistic structure have derived from the interest in cross-cultural cornrnunication, particularly in the work of J. Gurnperz (Gurnperz 1982, and Gurnperz (ed.) 1982). 26 Links & Letters 3, 1996 Peter Auer EHLICH, K. (1982). ((Anaphora and deixis: same, similar, or different?)). Jarvella, R.J. and Klein, W. (eds.), Speech, place and action, p. 31 5-338. London: Wiley. FAIRCLOUGH, N. (1992). Discourse and social change. Cambridge: Polity Press. FILLMORE, C.; &Y, P. and O'CONNOR, M.C. (1988). ((Regularity and idiomaticity in grarnrnatical constructions: The case of 'let alone')). Language 64: 501-538. FIRTH, J.R. (1935). The technique of remantics. Reprinted in Papers in Linguistics, 1934-51. Oxford: University Press (1 957), p. 7-33. GADAMER, H.G. (1960). Wahrheit und Methode. Tübingen: Mohr. (3rd. ed. 1972). GARFINKEL, H. (1967). Studies in ethnomethodology. Englewood Cliffs: Prentice Hall. GARFINKEL, H. and SACKS, H. (1970). «On formal structures of practica1 actions)). McKinney, J.C. and Tiryakian, A. (eds.), Theoretical sociolog, p. 337-366. New York: Appleton Crofts. GEERTZ, C. (1973). The interpretatiun of culture. New York: Basic Books. GOFFMAN, E. (1964). «The neglected situation)). Arnerican Anthropologist66,6: 133-136. - (1974). Frame anahsis. New York: Harper & Row. - (1979) «Footing». Semiotica 25: 1-29. GOODWIN, C. (1984). ((Notes on story structure and the organzation of participation)). Atkinson, P. and Heritage, J. (eds.), Structures of social action, p. 225-246. Cambridge: Cambridge University Press. GOODWIN, C. and DURANTI, A. (1992). ((Rethinking context: an introduction)). Duranti, C. and Goodwin, C. (eds.), Rethinking context, p. 1-42. Cambridge: University Press. GOODWIN, M.H. (1990). «Byplay: participant structure and frarning of collaborative collusion». Conein, B.; de Fornel, M. and Quéré, L. (eds.). Lesformes de la conversation, Vol. 11, 155-1 80. Paris: CNET. GUMPERZ, J. (1982). Discourse sti-ategies. Cambridge: University Press. - (1992a). ((Contextualization revisited)). Auer, P. and Di Luzio, A. (eds.), The contextualization of language, p. 39-54. Arnsterdarn: John Benjamins. - (1992b). «Contextualization and understanding)). Duranti, A. and Goodwin, C. (eds.), Rethinking context, p. 229-252. Cambridge: Cambridge University Press. - (ed.). (1982). Language and social identity. Carnbridge: Cambridge University Press. GUNTHNEK, S. (1996). aThe prosodic contextualization of moral work -an analysis of reproaches in 'whylformats)). Couper-Kuhlen, E. and Selting, M. (eds.), Prosody in conversation. Cambridge: Carnbridge University Press. HALLIDAY, M.A.K. and HASAN, R. (1976).Cohesion in English. London: Longman. - (1985). Language, context, and text. Victoria: Deakin University Press. HINNENKAMP, V. (1 989). Interaktionale Sozio/ingukh% und interhlturelle Kommunikacion. Tübingen: Niemeyer. HYMES, D. (1972). ((Models of the interaction of language and social lifen. Hyrnes, D. and Gurnperz, J. (eds.), Directions in sociolinguistics, p. 35-71. New York: Acadernic Press. JAKOBSON, R. (1971). ((Shifters, verbal categories, and the Russian verb)). Jakobson, R., Selected writings II, p. 130-147. The Hague: Mouton. KENDON, A. (1 990). Conducting interaction. Cambridge: Cambridge University Press. KJOLSETH, R. (1972). «Making sense: natural language and shared knowledge in understanding». Fishman, J.A. (ed.), Advances in the socioloby of language, p. 50- 76 The Hague: Mouton. KUNO, S. (1987). Fzmctionalsyntax. Chicago: Chicago University Press. Frorn Context to Contextualization Links & Letters 3, 1996 27 LABOV, W. and FANSHEL, D. (1977). Therapeutic discourse. New York: Academic Press. LEECH, G. (1975). Semantics. Harmondsworth: Penguin. LEVINSON, S. (1988). ((Putting linguistics on a proper footing: explorations in Goffman's concept of participation),. Drew, P., and Wootton, A. (eds.), Ewing Gofian - Exploring the interaction order, p. 161-227. Cambridge: Polity Press. ~~HNOWSKI, B. (1926). «The problern of rneaning in prirnitive languages)). Supplement to: 1. Ogden, C.K. and Richards, A. (eds.), The meaning of meaning, p. 296-336. London: Kegan Paul. MOERMAN, M. (1968). «Being lue: uses and abuses of ethnic identificationv. Helm, J. (ed.), Essays on theproblem of tribe. Washington: University of Washington Press. OEVERMANN, U., (et al.). (1979). «Die Methodologie einer 'objektiven' Hermeneutik und ihre allgerneine forschungslogische Bedeutung in den Sozialwissenschaften)). Soeffner, H.G. (ed.). Interpretative Ve@hren in den So.zia1- und Textwssen-Schafin, p. 352-434. Stuttgart: Metzler. ONG, W.J. (1982). Orality and literacy. London: Methuen. HARTSHORNE, C.; WEISS, P. and BURKS, A. (eds.) (1960-66). The collectedpapers of Charles Sanders Peirce. 8 volumes. Cambridge: Haward University Press. POMERANTZ, A. (1 975). Second assessments. A study of somefedtures of agreements/disagreements. Ph.D. Thesis, University of California at Irvine. PUTNAM, H. (1988). Representation and reality. Cambridge: M.I.T. Press. REICHMAN, R. (1984). Getting computers to talk likeyou and me. Cambridge: M.I.T. Press. SACKS, H. (1972). «An initial investigation of the usability of conversational data for doing sociology». Sudnow, D. (ed.). Studies in social interaction, p. 31-63. New York: Free Press. - (1 976). «On formulating context)). Pragmatic MicroJiche 7, 1 : 76. SCHEGLOFF, E. (1987). ((Bemeen micro and macro: contexts and other connections)). Alexander, J.C. (et al.), (eds.). The Micro-Macro Link. Berkeley: University of California Press. SCHEGLOFF, E. and SACKS, H. (1973). ((Opening up closings)). Semiotica VIII, 4: 289-327. SCHNEIDER, H.J. (1993). ((Auspragungen pragmatischen Denkens in der zeitgenossischen Sprachphilosophie». Stachowiak, H. (ed.), Pragmatik - Handbuchpragmatischen Denkens, p. 1-37. Vol. 4. Hamburg: Meiner. SCHUTZ, A. and LUCKMANN, T. (1976). Strukturen der Lebenswelt. Neuwied: Luchterhand. SELTING, M. (1994). ((Emphatic (speech) style - with special focus on the prosodic signalling of heightened emotive involvernent in conversation». ]oumal of Pragmatics, 22, 314: 375-408. SILVERSTEIN, M. (1976). ((Shifters, linguistic categories and cultural description)). Basso, K. (ed.), Meaning in anthropology, p. 11-56. Albuquerque: University of New Mexico Press. - (1979). ((Language structure and linguistic ideology)). Proceedings of the Chicago Linguistic Society. Parasession on Linguistic Units and Leuels, p. 193-247. - (1992). The indeterminacy of contextualization. Auer, P. and Di Luzio, A. (eds.), The contextualization of kznguage, p. 55-76. Amsterdam: John Benjamins. SPERBER, D. and WILSON, D. (1986). Releuance. Oxford: Basil Blackwell. TRIER, J. (1934). ((Deutsche Bedeumngsforschung)). Goetze, A., Horn, W. and Maurer, F. (eds.), Germanischephilologie, p. 173-200. Bonn: Winter. 28 Links & Letters 3, 1996 Perer Auer VOLOSHINOV, V.N. (1926). «Slovo v zhizni i slovo vpokzii: k voprosom sociologic'eskoj poktiki)). Zvezda, 6: 244-267. Reprinted as ~Discourse in life and discourse in art (Concerning Sociological Poetics))) Freudianism - A marxist critique (1976), p. 93-1 16. New York: Academic Press. - (1930). Marksizm ifolosofi jazyka. Leningrad: English translation Mamism and thephilosophy of language (1971). London: Seminar Press.