Bringing discourse analysis into the language classroom

Links & Letters 3, 1996 77-101 Bringing discourse analysis into the language classroom' Josep Maria Cots Departament d'Angles i de Lingüística Universitat de Lleida September 1995 Abstract The aim of this article is to argue in favour of adopting the point of view of discourse analysis in order to describe and explain how language is actually used. After pointing out that discourse should not be considered as one more leve1 in the description of language but rather as a change of perspective, the author proposes three premises which must be assumed in order to bring discourse analysis into the classroom. In the central part of the article the author suggests and exemplifies a series of concepts from discourse analysis which can be easily introduced in order to enable learners to reflect upon discourse and communication. The last section in the article includes specific guidelines for designing a language syllabus based on a discourse approach to language use. Key words: Discourse analysis, Language Teaching, Pragmatics, Language Awareness. Table of Contents 1. Discourse analysis: 4. Some useful concepts one more thing to teach? for the analysis of discourse 2. Premises for bringing discourse 5. Guidelines for classroom analysis into the language classroom wOrk On discourse 6. Final remarks 3. Awareness of discourse Bibliographic referentes 1. Discourse analysis: one more thing to teach? Bringing discourse analysis into the language classroom cannot be reduced exclusively to the adoption of a series of new categories and analytical techniques. It requires the teacher, in the first place, and the students, subsequently, 1. 1 would like to thank M. Irún, E. Llurda, M. Moyer, L. Nussbaum, M. O'Neill and an anonymous reviewer for their comments on drafts of this paper. This paper has been greatly enriched with their suggestions, although it probably includes views with which they disagree. 78 Links & Letters 3, 1996 Josep Maria Cots to look at their teaching and learning task in a very different way: the main focus of study is not language but communication. Although this may sound like a very obvious thing to say after so many years of having welcomed to our classrooms the communicative and notional-functional approaches, in my opinion there still has not been a real change of approach to language teaching and learning. The reason for this is that, with some exceptions, the linguistic education of most of the present language teachers is still very much influenced by a structural approach, which prioritizes the study of language as an autonomous system of rules rather than as a means of communication. It may still take some years before we start to see the effects in the classroom of this new change of perspective which is progressively taking place in the linguistic sciences. This change is already reflected in the incorporation of specialists in discourse analysis and pragmatics in philology departments and the inclusion of university courses under these titles in the curricula of future language teachers. The introduction of the notion of communicative competence has been positive in the sense that communication is now conceived as a result of the successful application of not only grammatical but also pragmatic knowledge and skills (Cande 1983). Nevertheless, this has not changed the perspective many teachers had of language, because pragmatic information has usually been presented as an auxiliary component which is to be used only when grammatical explanations are difficult or impossible. The result is that for many language educators there is still a core element about which they have to be specially careful: grammatical competence. Pragmatic competence has become the equivalent to the topping of an ice-cream: it is nice if we can have it, yet it is not the ice-cream. The communicative approach to language teaching has succeeded in making us aware of some factors, which were not taken into account previously, but it has failed to integrate those factors into a new way of looking at language in use. One of the most clarifying definitions of the new perspective of language which is advocated by many discourse analysts has been put forward by Verschueren (1987: 38-39) in the form of what he calls an impertinent question as the way to a pertinent answer: What and how does language contribute to human survival on the level of the human race, smaller and larger communiries, individuals and day-to-day situations? Language, according to this author, is to be considered as a means for the human being to adapt to psychologicd/cognitive, physical and socio-cultural circumstances. This process of adaptation is not only in one direction; sometimes human beings can adapt those circumstances to their needs or goals by means of language. Grammaticai choices that people make when communicating constitute just one level of adaptation. Other levels of adaptation have to do with choices of sign system, channel, code, style, speech event, discourse, speech act, propositional content, word and sound. The pedagogic consequence of this new functional perspective is that language and communication are placed in a wider framework. They can no Bringing discourse analysis into the language classroom Links & Letters 3, 1996 79 longer be defined as pedagogic aims in themselves, but as a means of adaptation and survival in the world. Another consequence is that, if the choice of a specific lexico-grammatical structure is presented as one level of adaptation, side by side with other levels (choice of pronunciation, style, situation, channel, function), we should start thinking of distributing our teaching efforts accordingly, instead of giving grammar so much priority over the other levels. As some teachers have pointed out to me, the main problem they find in bringing a discourse frame into the classroom is not grammar-blindness, but the absence of systematic descriptions integrating al1 the levels of adaptation into pedagogical materials. 2. Premises for bringing discourse analysis into the language classroom According to the dictionary definition, a premise is something that you suppose is true and that you use as a basis for developing an idea or a statement. Every language teacher approaches hislher task with certain premises about what language is and how it is used. These premises are usually based on the teachers' academic experience in the description and learning of languages and inevitably penetrate their pedagogic task independently of the methodltextbook they follow. An example of teachers' premises can be seen in a study by Mitchell and Hooper (1992) carried out in England with secondary school teachers of English and foreign languages. Both groups of teachers equate knowledge about language with morpho-syntactic knowledge of a traditional kind, centred on written language. Whereas the English teachers were sceptical of the value of this type of knowledge in improving communicative performance, the foreign language teachers considered it essential for language learning. The ability to analyse hnguage variation was one of the English teachers' goals in their classes; this notion was totally absent in the case of the foreign language teachers. This, the researchers point out, w;s a rather surprising finding given the popularity of the concept of communit ative competence. Finally, when asked about what was involved in knowing about a language, neither of the two groups mentioned such important topics in present linguistic research as the structure of discourse beyond the level of the individual sentence, the spoken language in al1 its aspects, and firstlsecond language acquisition/development. My discourse view of language use is based on three general premises that can be summarized under the following headings: (i) communicative competente; (ii) context, language variation and real data; and (iii) negotiation of intentions and interpretation. It is true that these are familiar notions to most language educators and it is also true that many of them have tried to assume them in their pedagogic practice by introducing specific communicative tasks. However, it seems to me that, in spite of al1 the materials published (eg. Nolasco and Arthur 1987, Bygate 1987, Cook 1989, McCarthy 1991, McCarthy and Carter 1994), for many teachers, in the intimacy of their classrooms, discourse 80 Links & Letters 3, 1996 Josep Maria Cots premises have not yet become the real basis of their language syllabus nor of their evaluation system, both of which are still strongly influenced by a structural approach to language. If a teacher wants to adopt a discourse point of view in teaching a language helshe must make an effort to modiG some of hislher premises about language, and this will necessarily have an influence on the design of the syllabus and on the methodology employed. 2.1. Communicative competence Following Hymes (1971), we could say that the first premise involves the acceptance that a learneri capacity to communicate in a foreign language involves knowing not only (i) whether and to what extent a structure is formally possible (i.e., pmmatical), but also whether and to what extent it is (ii) psy~holo~ically fedsible (i.e., that can beprocessed without a gmt effort), (iii) appropriate and (iv) actually said. The presentation of each linguistic structure should not ignore any of these four types of knowledge because al1 of them are taken into account in a more or less conscious way by competent users of the language. These four types of knowledge constitute a whole, which Hymes defines as communicative competence, and which could be defined as the capacity to perform verbally in a successful and acceptable way. The implication of this is that a grammar which attempts to account for verbal performance rather than competence should not accumulate but integrate information about the four aspects mentioned above. Such a grammar should combine the following ingredients: a) The elaborate descriptive apparatus of the structural and generative approaches. b) The general approach of functional grammar according to which no language can be understood without taking into account its three basic functions: ideational (i.e., to reflect upon the world), interpersonal (i.e., to act upon the others) and textual (i.e., to construct adequate messages which can fulfill the previous two functions). c) The necessary sociolinguistic and ethnographic information to assess the extent to which a specific linguistic structure is used and the social connotation it carries. One of the most interesting attempts to analyse ~~stematically the notion of communicative competence with a view to its incorporation in a foreign language syllabus has been proposed by Canale and Swain (Canale and Swain 1980, Canale 1983)2. According to these authors, communicative compe- 2. Several authors share the view that the analysis of language into different areas of knowledge and the dassification of verbal features into hose areas can be a basis for syllabus design. These authors prioritize language use in the classroom via the implementation of suitable tasks. See, for instance, Munby (1978), Candlin and Murphy (1987), Littlewood (1981) and Yalden (1983). Bringing discourse analy& into the language classroom Links & Letters 3, 1996 81 tence can be considered, for pedagogic purposes, as the result of the successful application of four different types of knowledge and skills: grarnmatical, sociolinguistic, discourse and strategic. 2.2. Context, language variation and real data The second premise is based on the idea that language use is varied and has dzfferent functions depending on the context. Faced with this variation, the teacher can take two options: concentrate on just one type of use as the most important, or adopt a more realistic point of view and accept the fact that successful communication is based precisely on the capacity of the speakers to manipulate these different uses. The only way to capture and explain this variation is by defining the notion of context as precisely as possible and trying to find out in what way language use is influenced by the different contextual aspects and, at the same time, how these same contextual aspects can be altered by the action of the speakers. In order to be able to account for the relationship benveen the formal features of a text and their sociallcommunicative context McCarthy and Carter (1994) propose dividing the world of discourse into different genres which at the lexico-grammatical leve1 are realized by registers. In this sense, a genre is considered as a articular «communication format)) defined according to its social function. B The importance of context for communication forces us to accept that, if we really want to train our learners to become competent at communicating, the models of verbal behaviour that we show to them must be as contextualized as possible. Otherwise, in our attempt to simpliQ or idealize the data we will be creating an artificial kind of communication, thereby depriving the learners of contextual indexes such as dialect, social relationship between the participants, or social functionldefinition of the communicative situation, which we make use of in our everyday social encounters. The pedagogic consequence which results from this is that the best (and perhaps the only) way to contextualize language is simply to use real instances of language use, in which the full potential of language can be appreciated by looking at its social effects. 2.3. Negotiation of intentions and interpretations A third premise which needs to be assumed when we approach language as discourse is that communication cannot be explained as the simple transfer ofpreexisting meanings. Rather, a communicative event must be conceived as the locus where meanings are created through the negotiation of intentions and interpretations. As Nunan says (1993: 91), verbal interaction is the result of 3. See section 5.1 for a further development of the concept of genre. 82 Links & Letters 3, 1996 Josep Maria Cots the cooperative work of the speakers to make sure that their messages are being received in the way they were intended, and of the listeners to ensure that their interpretation coincides with the speakers' intentions. The devices that speakers use in order to ensure understanding go from accommodating one's discourse to the discourse of the interlocutor (by choosing the same style, register, dialect or discourse routines) to deploying specific strategies such as adjusting the level of explicitness to the needs of the interlocutor, requesting clarification or supplying backchannel, use of paraphrase and metaphor. Bygate (1987: 34-35) lists a series of strategies for negotiating meaning in spoken interaction, which may be deployed by the speaker (e.g., announce one's purpose in advance, clarify meaninglintention by summarizing, rephrase, exemplify) or by the listener (e.g., indicate understanding by gestures or verbal tokens, ask for clarification, indicate current interpretations). The issue of negotiation of meaning is especially important in foreign language teaching/learning for two main reasons: (i) It has been argued (i.e., Long 1983, Swain 1985 -cited in Nunan 1993) that the process of acquisition is enhanced when learners are put in a position where they have to negotiate meaning in order to make themselves understood. It seems that when learners participate in communicative situations where they have to negotiate meaning in order to make themselves comprehensible to their interlocutors (i.e., problem solving and information gap tasks) their competence is pushed to the limit and this encourages the acquisition process. (ii) Using a foreign language is very often synonymous with participating in an intercultural communicative event, where the risk of misunderstanding increases because the interlocutors do not share the same cultural values and have different expectations about how to do things with language. In this situation, the negotiation of meaning has an even more important role than in intracultural communication. 3. Awareness of discourse One of the innovations in the new curriculum in foreign languages for secondary education in Catalonia (Educació Secundaria Obligatoria, E.S.O.) has been the inclusion, as a pedagogic procedure, of the need to engage the learners in reflecting upon language and the processes of communication and learning (Diari Oficial de la Generalitat de Catalunya 1992). According to Nussbaum (1991: 62), one of the problems that the inclusion of this specific procedure rnay involve is that until very recently, the tasks and exercises intended to promote reflection upon language had a clearly prescriptive emphasis, telling the learners what is right and what is wrong instead of telling them what is appropriate and what is not according to the context. The reason for this is that the traditional way of reflecting upon language was done by applying the grammatical method and concentrating exclusively on the level of the form, Bringing discourse analysis into the language classroom Links & Letters 3, 1996 83 the structures of the language, and ignoring two other levels: function and socio-cultural meaning (McCarthy and Carter 1994). The inclusion of the promotion of awareness of communication as a procedure for teaching will force us to re-examine some teaching methods which assign little importance to the development of the learners' capacity to reflect upon language. The defenders of those methods argue that rather than promoting performance, an excessive degree of self-consciousness will inhibit it and, fürthermore, will make the learning task more complex by increasing the number of metalinguistic terms and categories with which the learner will have to cope. Nevertheless, James & Garrett (1991) show, by appealing to different authors and empirical studies, that the promotion of general language awareness not only enhances performance but also increases the learners' empathy towards the language studied, develops social and cultural tolerance and reinforces general cognitive capacities. The challenge that the inclusion of a procedure consisting of promoting awareness of communication poses for applied linguistics is that of showing language educators that discourse can be described and explained in a systematic way, by proposing a series of rules and principles which can hardly be found in a linguistic theory strictu senso but in a theory of human/social action. Some efforts have already been made in this direction (see, for example, Bygate 1987, Cook 1989, McCarthy 1991, Hatch 1993, Lomas et al. 1993, the new curriculum for secondary education -E.S.O., and the new syllabus for state teaching examinations for secondary schools and Official Schools of Languages in Boletín Oficial del Estado 1993). However, with the exception of a few «pioneers», many of us, afraid of losing the power and security that grammar confers, still seem excessively reluctant to adopt a discourse approach to language in which the concepts of wrong and right become very diffused. The situation in some cases may even be contradictory in the sense that while some kind of explicit formalization of grammar rules is considered necessary for an adequate learning process, we expect our students simply to absorb rules of discourse without any explicit formalization. This contradiction has been pointed out by Thomas (1 983: 109-1 10): Much effort is expended in writing nugatory texts explaining low-leve1 rules of grarnrnar, such as third person singular -S (which, since it is readily observable in the surface structure requires little explicit forrnalization). Pragrnatic failure, rneanwhile, like covert grarnrnatical error, often passes unchecked by the teacher or, worse, it is attributed to sorne other cause, such as rudeness, and the student is criticized accordingly. (. . .) this problern can be overcorne only by giving the student the tools to make the process ofpragmatic decision-making explicit. (rny italics) Promoting discourse awareness in the classroom should not imply that we must exclude other approaches to language learning which are in favour of an unconscious type of learning, focusing exclusively on meaning and use. Both approaches can be complementary. There are, however, two important peda- 84 Links & Letters 3, 1996 Josep Maria Cots gogic aspects (besides those mentioned in James and Garrett 1991) in which the development of language awareness can have a positive effect: (i) it allows us to cater for those individuals with a more analytical cognitive style and (ii) it is a way of integrating language more intimately in the school curriculum by showing that, like physics or history, it may be subject to rational enquiry. 4. Some useful concepts for the analysis of discourse In this section 1 will try to adopt a very schematic and, at the same time, practical approach. The reason for it is that 1 do not intend to present an overall view of discourse analysis4 but suggest some conceptual tools as clearly and simply as possible so that language teachers who know little about discourse analysis might feel comfortable and, therefore, be willing to try them out in their classes with the linguistic materials they are already using. For example, the very same text which has been used to introduce or reinforce a gammatical structure or a function could be used, instead, to talk about the type and distribution of speaking turns (if it is a dialogue), the levels of politeness, the construction of coherence and understanding by means of specific linguistic tokens or the type of communicative/social situation. Examples of how discourse and pragmatic information can be incorporated in the structural work that is usually done through typical textbook dialogues are found in Cots (1994). One of the dialogues that are commented on is the following (Hartely & Viney 1979): E: Excuse me! F: Yes! E: I'm lost! 1s this the way to Brighton? F: No, I'm afraid it isn't. You're going the wrong way. This is the Portsmouth Road. E: Oh dear. Can you te11 me the way to Brighton? F: Yes, turn around and go back to the roundabout. Take the third exit.. . that's the A272. E: TheA272? F: That's right. You'll see signposts to Brighton from there. This short conversation may be a good means to introduce ((getting-atten- tion» interactional routines (-Excuse me!-Ye!), a speech act idiomatic realization to express psychological state (Oh dear!). We can talk as well about the connotations in terms of speaker< empathy of saying Im aJTdid it isn't as opposed to No, it isn't or saying Thatj rightversus Es, There is also in the conversation one example of a strategy for negotiating meaning, intended to check understanding by repeating what the previous speaker said in an interrogative form ( The A272?). 4. See the following authors for excellent introductions to the subject: Brown and Yule (1983), Cook (1989), Hatch (1992), Nunan (1993). Bringing discourse analysis into the language classroom Links & Letters 3, 1996 85 Adopting a discourse approach to language teaching involves a new way of looking at it, but this does not mean that we should throw out the baby with the dirty water. It is important to continue those teaching practices which we have found effective through our practical experience and, at the same time, it is also important not to lose sight of the pragmatic and discourse theories of language use to better illuminate our practice. In the presentation of the concepts 1 am going to establish a distinction between (i) those which can be used to segment and find structure in discourse and (ii) those which can be used to explain discourse and communication in general. 1 will cal1 the first group descriptive and the second explanatory. In some cases, it might be suggested that some of the conceptual tools proposed are familiar to the learners from their first language and, therefore, there is no need to introduce them again in class. This would be the case for example of an explanatory concept like the conversationalprinciple/maxim of relevance. There are two posible answers to this. In the first place, if we take a grammaticalprinciple like subject-verb agreement, we can easily see that it applies to both English and Catalan. Nevertheless, this should not be an obstacle to its inclusion as something to be taught. The reason for this is that it applies in different circumstances and it is realized by different means. The same happens with conversationalprinciples/maxims: although they may exist in the two languages, it may also be the case that they are followed or violated in different circumstances; so whereas in one specific context it may be appropriate to «sound irrelevant~, in a similar context in the other culture it may be totally inappropriate. The second answer is that we must be careful not to take for granted that because we can al1 communicate in our own language it follows that we al1 share the same communicative values. As Wierzbicka (1991) shows, different cultures have different values and these are projected onto the communicative behaviour. Thus a feature like indirectionality in requests in order to avoid imposition on the addressee is appropriate in American culture but not in Israeli culture. The same may be applied to a feature like turn-taking. The fact that we al1 take turns when taking part in a conversation does not imply that we take them in the same way. Tannen (1984), for instance, points out that whereas American Jews see overlap as a sign of involvement and empathy, in the case of white Anglo-Saxon speakers this is seen a sign of lack of interest and lack of respect. 4. l. Descriptive concepts 4.1.1. Speech act Definition: An utterance considered from the point of view of its function(s) in the process of communication. Usefulness: When we analyse an utterance as a speech act we can incorporate, besides the basic literal meaning of the words contained in the utter- 92 Links & Letters 3, 1996 Josep Maria Cors (2) Cooperative Principle: quality A: Wouldyou like something to drink? B: Yeuh, pleuse. I'm dyingfor u coke. Here again, we have another maxim which is being deliberately flouted to produce a special effect. In this case, it is the maxim of quality: speaker B is not really dying, helshe is just very thirsty, but he makes the assumption that B is familiar with the principle of cooperation and that, therefore, helshe will interpret this violation as an attempt to increase the effectiveness of A's message. (3) Politeness Principle: tact maxim Lend me your cur > Could Ipossibly borrow your cur? If we simply follow the Cooperative Principle, we must say that the first utterance is perfectly valid because it obeys the four maxims (the speaker is being true, brief, relevant and clear). However, we should bear in mind that language users are not only interested in the effective transfer of information but also in the maintenance of social harmony. In the present example, the utterance fails to do that because it is an imposition. The second utterance diminishes greatly the imposition of the first by giving the addressee grounds to decline the compliance of the request (he might be unable to do it), presenting it as a hypothetical action ((ccould» and «possibly») and by omitting reference to the willingness of the addressee (((1 borrow» instead of «you lendn). Leech (1983) proposes six maxims for the Politeness Principle: tact (minimize cost to other, maximize cost to self), generosity (minimize benefit to self, maximize benefit to other), approbation (minimize dispraise of other, maximize praise of other), modesty (minimize praise of self, maximize dispraise of self), agreement (minimize disagreement between self and other, maximize agreement between self and other), and sympathy (minimize antipathy between self and other, maxiniize sympathy between self and other). (4) Politeness Principle: agreement maxim A: I think he shouídstuy here. B: - I can 't agree with you there. - ThutS just what I wus thinking. The two responses provided by B follow perfectly the agreement maxim. In the first case, the disagreement is minimized by appealing to the inability of the speaker rather than his willingness. In the second response, the agreement is maximized by pointing out the exact coincidencelsynchrony between A and B. 4.2.2. Presentation of self Definition: The public image that the members of a social group want to project according to a series of values which are dominant in that group. Bringing discourse analysis into the language classroom Links & Letters 3, 1996 93 According to Goffman (1967) this public image consists of a positive and a negative face. The positive face is one's concern to be considered a normal, contributing member of oneS social group. The negative face is hislher wish to be independent and free from the imposition of others. Usefulness: By means of the concept of presentation of self we can look at discourse as a means for the speakers not only to carry out transactions of goods or services but also to use their interpersonal skills in defining the situation, enacting a self or identity, and using strategies to accomplish other interactionai ends. Examples: (1) Avoiding assertiveness - 1 don't know, 1 guess he is not the bestperson to do it. - Maybe, he? not the bestperson to do it. - It? a kind of study about oldpeople. In the examples above we have three ways of diminishing the assertiveness of a statement. In this way the individual can tackle the risk of being understood as excessively independent, self-confident and even as wanting to impose hislher views about the world on the others. (2) Explaining and justifying - He? not vey bright, 1 don't mean to say he can't follow the classes, but his rnarks are a little lower than average. - We must be there at 9.30 because the concert starts at 10.00. Explaining and justifying are two ways of preserving aposteriori a positive public image. The first utterance is an example of an explanation which is added to an initial statement which could have been understood in a very negative sense and therefore as a rather aggressive act for the public image of the person being talked about. The subordinate clause of the second utterance is intended as a justification for what might be interpreted as an act of arbitrariness. (3) Revealing one's inner self - Oh my goodness! I love thatpicture! It? wonde@l. - I'm real4 enthusiastic about my new job. - I hope you will not do it. A very important aspect of an individual's presentation of self is connected with the revelation of details of one's inner self (beliefs, wishes, opinions, likes, dislikes, etc.) on interacting with others. This is an aspect of discourse which has to do with one's personal style but aiso with the cultural background of the speaker (Tannen 1984). 94 Link. & Letters 3, 1996 Josep Maria Cots 4.2.3. Communicative strategy Definition: The systematic adoption of a series of verbal actions which respond to a more or less conscious plan or communicative routine to achieve a specific goal. Usefulness: It allows us to conceive verbal behaviour as consisting not of independent, isolated actions which ate the result of sudden decisions by the speaker, but rather as the implementation of a plan or communicative routine in which a series of verbal actions are mutually dependent and they al1 contribute to the achievement of a specific goal. Examples: (1) Asking for directions Action 1: A: Excuse me! B: Yes? Action 2: A: Im lost. This is not the way to Brighton, is it? B: No, I'vn afiaid it isn 't. This is the Kent Road. Action 3: A: Oh dear. Can you tell me the way to Brighton? B: Yes, turn around andgo back to the roundabout. Take the third exit. ThatS the A4. Action 4: Many thanks. It is al1 too often the case that a function like «asking for directions)) is presented as requiring only one utterance (i.e., Can you te11 me...?). However, as the example shows, it is much more realistic to present it as the result of a series of concatenated actions including utterances with the function of attracting someone's attention, explainingljusti~ing oneself and thanking. (2) Making an invitation Action 1 : A: Hello Tea? B: Oh, bello! Action 2: A: Are you doing anything tonight? B: No, not really. Action 3: A: Wouldyou like to go to the cinema? B: Ok, what time shall we meet? Action 4: A: I'llpick you up at 9.00, ok? B: All right. Action 5: A: See you then. Bye. B: Ok. Bye. This conversation, as in the previous case, responds to a plan which speakers follow unconsciously because they have internalized it as a communicative routine from previous occasions of language use. What is important in terms of explaining discourse in the classroom is to point out that this plan includes different stages (greeting, preparation, invitation, meeting and leave-taking) and that it requires the cooperation of the two participants in the conversation in order to be effective. Bringing discourse analysis into the language classroorn Link; & Letters 3, 1996 95 4.2.4. Cultural inferencing Definition:Process by which we derive information which is not literally included in the message but which can be derived from it by applying sociocultural or general knowledge. Since the listener has no direct access to the intended meaning of the speaker, helshe must rely on the process of inference to arrive at an interpretation. Usefü1ness:The systematic study of examples of real communication, in which inference inevitably plays an important part, offers us a way of introducing specific socio-cultural knowledge not as an extra component of a language course (the ((civilization and culture component))) but as an essential aspect of the ability to communicate in real contexts. Examples: (1) Im not afiaid of ramboes like you. In order to fully understand the message conveyed by this utterance it is necessary to apply non-linguistic knowledge related to the film industry and one of its popular characters known as Rambo. (2) Unbelievable! Springtime and you split up with your girpiend. In this case again, in order to find the two parts of the utterance coherent, it is necessary to apply non-linguistic knowledge which has its origin mainly in the literary tradition. This tradition has usually associated the spring season with the awakening of love. (3) 1 went into the restaurant, Ipaid my meal and then lookedfor a table to eat it. This utterance may be incoherent to many people who are not familiar with fast-food restaurants in which the meal is paid in advance. (4) It was lunch time and all the pupils took out their sandwiches. Of course, the smell ofpeanut butter andjelly spread out immediately in the room. The presence of the particle «of course)) in this utterance can only be understood if we have information about the favourite type of sandwich for young children in the U.S. 5. Guidelines for classroom work on discourse Bringing discourse analysis into the language classroom cannot be an end in itself but rather a means to make our pedagogic task more efficient and effective in developing the learners' communicative competence. As 1 have pointed out previously, this competence must be understood as different sets of knowledge and skills (linguistic, sociolinguistic, discourse and strategic) that the speaker must make use of in order to produce coherent, unified and socially appropriate messages. According to Lomas, Osoro and Tusón (1993: 80-81) in order 96 Links & Letters 3, 1996 Josep Maria Cots to acquire these four types of competence, the learner will have to develop the following skills: (i) Identify the relevant elements of a communicative situation. (ii) Identify and use different genres, styles and registers. (iii) Use language as an instrument of interpersonal relations, to construct one's thought and to discover and interpret reality. (iv) Identify the value, origin and social function of the different discourse practices. In the rest of this section 1 will comment on three issues with which the teacher might be confronted when it comes to designing a syllabus based on a discourse approach to language: (i) the segmentation of discourse, (ii) sequencing, (iii) selection and presentation of instances of language use. 5.1. The segmentation of discourse When organizing a language curriculum according to a discourse approach we should first distinguish four basic modes: spoken-written, monologic-dialogic. Although these modes can be prototypically defined in terms of specific formal features, we should not conclude that the possibilities of expression of the language users are reduced to four possibilities. Rather, the four basic modes hnction as the extremes of nvo axes of variation. Thus, we can have a formal spoken text with typical features of written language, or a monologic text with dialogic features. Depending on the context of use and the intended effect of the message, the speakerlwriter will decide, by choosing certain formal features, the position of hislher discourse on the two axes. After the mode, the next leve1 of segmentation of discourse is genre, which can be defined as a socially recognized type of communication with a specific social function which determines both the rhetorical structure and the choice of gammatical and lexical elements. Carter and McCarthy (1994) mention a series of «core generic functions)) to organise the syllabus: reporting, narrating, persuading, arguing, describing, instructing, etc. Each of these core generic functions is further subdivided into specific genres. For instance, information report, weather report and progress report are three different realizations of reporting. Each of these realizations adopts a specific register according to the subject matter, the medium and the type of audience. Finally, it is posible to find texts which are the result of generic blends such as reporting and predicting, narrating and arguing or explaining and persuading. The advantage of presenting language use through different genres is that, besides emphasizing the tight relationship between social function and linguistic form, we can situate our text as part of a cultural and textual tradition which is decisive in many cases in order to explain the form and contents of certain ((ritualized)) forms of communication. Cook (1989: 95) presents a list of ((discourse types)) or genres which can be useful to exemplify the kind of genres 1 have in mind: Bringing discourse analysis into the language classroom Links & Letters 3, 1996 97 recipe joke anecdote label poem letter advertisement report message note chat seminar manifesto toast argument The third level of discourse segmentation is that of strategy, through which we can integrate the participants in a communicative event as real actors who decide, according to their goals and the circumstances, what to say and how to say it. The concept of strategy also implies a systematic way of using the language and, in this sense, we can integrate lexico-grammatical choices and socio-cultural constraints in the description and practice of specific verbal actions. In order to define the specific strategies that the syllabus must contain McCarthy and Carter (1994: 180-8 1) propose the following «presyllabus» with groups of strategies and questions that can be asked in order to specifj different discourse features. The six groups of strategies proposed are the following: 1. Genre-related strategies: most frequent media, modes and genres; patterns of interaction (narrative, problem-solution). 2. Coherence-related strategies: topic management, turn-taking, cohesion. 3. Politeness strategies: face, forms of address, reciprocity. 4. Planning strategies: anticipatory strategies (enumeration, cataphoric use of articles and pronouns), sequences of tense, aspect and voice, appropriate degree of creativity and risk-taking. 5. convergence strategies: informational and cognitive convergence (theme, mood and modality), affective convergence (solidarity routines, agreementdisagreement), shared cultural knowledge. 6. Repair strategies: risks of communicational problems and cultural misunderstandings, self-repairlcooperative repair. 5.2. Sequencing Another issue that must be faced in the design of a language curriculum has to do with the sequence in which the different elements constituting discourse will be presented. Here we can basically distinguish two main approaches which are based on theories of how we process language: bottom-up and topdown. A bottom-up approach to describing discourse conceives communication as divided into different levels and it proposes an analysis which begins with the smallest units of language (i.e. soundslletters) and continues through a series of inclusive levels up to the levels of the text or discourse. Cook (1989: 80) mentions the following levels: soundslletters, grammar and lexis, cohesion, conversational mechanisms, discourse function, discourse structure, discourse type, shared knowledge, social relationships. In my opinion, many attempts to bring discourse into the language classroom have failed because of this approach, since the teachers feel that they cannot move onto the next level until 98 Links & Letters 3, 1996 Josep Maria Cots the learners have mastered the previous one and usually the great «divide» appears benveen the sentence and discourse levels. The alternative approach, known as top-down, considers thar language users make sense of discourse by moving from the highest units of analysis to the lowest, starting with background knowledge and expectations about the purposelcontents of the message and the strucrure of the text and looking for confirmation of these at the lower levels. This seems a more realistic approach based on the experience of how we understand difficult discourse: we take some linguistic or situational detail as the cue which allows us to make a general hypothesis which must be progressively confirmed through our analysis (readingllistening) of the text. This second approach seems much more useful in the case of foreign language learning, where subjects already have a great deal of communicative experience in their own language which they must be able to use in order to compensate for deficiencies at lower levels. When we want to understand difficult discourse we usually start with knowledge about the sender, the intended receiver, the purpose, rhe genre. This knowledge triggers some expectations about the content and form of the text which are confirmed or not through our processing of the text. Rather than choosing between one or the other approach when organizing a syllabus, it is probably better to combine them in what Stanovich (1980, cited in Nunan 1993) defines as ((interactive-compensatory model». According to his model, discourse comprehension is not as simple as moving from higher to lower levels or viceversa. It is an interactive process, in which we use information from more than one level simultaneously, and deficiencies at one level can be compensated for by any other level, independently of its rank. 5.3. Selection andpresentation of instances of language use The presentation of the different instances of language which constitute the basis of the syllabus is also affected by the approach we adopt. McCarthy and Carter (1994) propose five main principles which, in their opinion, should be central to discourse-based language teaching: (i) The contrastive principle: Understanding of how language operates in a particular situationltext is increased when it is compared and contrasted with different situationsltexts in the foreign language and in the learner's first language. (ii) The continuum principle: The series of texts which are produced by a speech community do not constitute clearly separate compartments but a continuum in which the freedom of the user has a very important role when it comes to ~manipulating)), according to hislher intenrions, the elements which define the prototypical form of the text. (iii) The inferencing principle: Any sample of real language carries a cultural load which must be taken into account in order to make sense of it. Thexefore, it is necessary to devote some effort not only to ((teaching culture» but also to teaching actual procedures to infer meaning. Bringing discourse analysis into the language classroom Links & Letters 3, 1996 99 (iv) Familiar to unfamiliar principle: Following Vygotsky (1979, cited in Lomas, Osoro y Tusón 1993: 91) and his notion of «zone of proximal development)), this principle states that effective learning takes place when the starting point is discourse practices with which the learners are familiar, to continue in later stages with discourse practices which are more complex, elaborated and decontextualized. (v) The critical principle: Language should also be presented as a means of manipulation and creation of ideology for those who traditionally have more access to the media for verbal communication (governments, bureaucracies, teachers (?!), etc.). By developing their critical capacity to look for traces of ideology beyond language we are empowering our students and helping them to become freer educated citizens. 6. Final remarks My aim in this article has been to explore how the work of discourse analysts can be incorporated into language teaching. As it is clear from the title, the article is not an attempt to show how discourse can be brought into the classroom but rather how the discourse analysis can help language educators and learners to better understand language use and, therefore, contribute to making the teachingtlearning experience more effective. It has been my intention to show that discourse analysis can inform language educators from three different points of view: general ~nderstandin~ about language, metalinguistic reflection and syllabus design. 1 have started by suggesting what 1 think are three very important premises about language use that characterize discourse analysis as different from senteiice linguistics: (i) communicative competence vs. linguistic competence; (ii) contextual variation vs. homogenization; and (iii) negotiation vs. transfer of meaning. After pointing out the importante of promoting awareness of how discourse works as part of a more general language awareness programme in language education, 1 suggest a series of concepts from discourse analysis which can be useful to ((talk)) in the classroom about language use. These concepts should allow us not only to describe discourse but also to explain it in a way which can be easily accessible to the learners. Finally, 1 consider three specific issues that need to be taken into account when designing a teaching syllabus based on a discourse approach to language: segmentation, sequencing and selection and presentation of discourse. If we accept that our goal as teachers is essentially to help our learners to become efficient and effective participants in real communicative situations and that this involves more than the knowledge and skills necessary to manipulate the structures of a language, then we must inevitably appeal to a theory of language use that attempts to integrate its social, cognitive and linguistic strands. This is precisely what discourse analysis does in trying to discover the system underlying our verbal productions. 100 Links & Letters 3, 1996 Josep Maria Cots BOLET~N OFICIAL DEL ESTADO (BOE) (1993). No. 226, Se~ternber 21st. CANDLIN, C. and MURPHY, D. (eds.). Language learning tasks. Englewood Cliffs, NJ: Prentice Hall. BYGATE, M. (1987). Speaking. Oxford: Oxford University Press. CANALE, M. (1983). ((From cornrnunicative competence to cornrnunicative language pedagogyn. Richards, J. and R. Schrnidt (eds.). Language and communication, p. 2-27. London and New York: Longrnan. CANALE, M. and SWAIN, M. (1980). ((Theoretical bases of communicative approaches to second language teaching and testing)). Applied Linguistics 1, 1: 1-47. COOK, G. (1989). Discourse. Oxford: Oxford University Press. COTS, J.M. (1994). ((Un enfoque socio-pragmático en la enseíianza de una lengua extranjera)). Signos 11: 46-5 1. Dwu OFICIAL DE LA GENERALITAT DE CATALUNYA (DOGC) (1992). Decret 9611992, no. 1593, April 13th. GOFFMAN, E. (1967). Relations inpublic. New York: Anchor Books. GRICE, H. (1975). aLogic and conversation~. Cole, P. and Morgan, J. (eds.). Syntax and semantics, volume 9: Pragmatics, p. 41-58. New York: Acadernic Press. HARTLEY, B. and VINEY, P. (1979). Streamline Englirh: Connections. Oxford: Oxford University Press. HYMES, D. (1972). «On communicative cornpetence)). Pride, J. and Holrnes, J. (eds.). S~ciolin~uistics, p. 269-293. Hardmonsworth, England: Penguin. LAKOFF, R. (1973). «The logic of politenes: rninding your p's and q's». Papersjiom the 9th Regional Meeting, Chicago Linguistics Society: p. 292-305. LEECH, G. (1983). Principies ofpragmatics. Harlow, England: Longrnan. LITTLEWOOD, W. (1981). Communicative language teaching. Carnbridge: Carnbridge University Press. LOMAS, C.; OSORO, A. and TUSÓN, A. (1993). Ciencias del lenguaje, competencia comunicativa y ensezanza de la lengua. Barcelona: Paidós. MCCARTHY, M. (1 99 1). Discourse analysisfor kznguage teachers. Cambridge: Cambridge University Press. MCCARTHY, M. and CARTER, R. (eds.) (1994). Language as discourse: Perspectivesfor language teaching. London and New York: Longrnan. MITCHELL, R. and HOOPER, J. (1992). ((Teachers' views of language knowledge)). James, C. and Garrett , P. (eds.). Language Awareness in the Classroom, p. 40-50. London and New York: Longman. MUNBY, J. (1 978). Communicative syllabus design. Cambridge: Carnbridge University Press. NUNAN, D. (1993). Introducingdiscourse analysis. London: Penguin. NUSSBAUM, L. (1991). «De cómo recuperar la palabra en la clase de lengua: Notas para el estudio del uso oral)). Signos 2,2: 60-67. STANOVICH, K. (1980). ((Toward an interactive cornpensatory rnodel of individual differences in the developrnent of reading fluency)). Reading Research Quarterly 16: 32-71. TANNEN, D. (1984). Conversational style: Analyzing talk amo-jiiends. Norwood, New Jersey: Ablex. Bringing discourse anal~sis into the language classroom Link; & Letters 3, 1996 101 THOMAS, J. (1 983). ((Cross-cultural pragmatic failure)). Applied Linguistics 4,2: 9 1- 1 12. VERSCHUEREN, J. (1987). ((Pragmatics as a theory of linguistic adgpatation)). IPRA Working Docurnent. Antwerp, Belgium: University of Annverp. VYGOTSKY, L. (1 979). El desarrollo de los procesos psicoldgicos superiores. Barcelona: Crítica. WIDDOWSON, H. (1984). Explorationr in applied linguistics 2. Oxford: Oxford University Press. WIERZBICKA, A. (1 99 1). Cross-culturalprapatics: The semantics of human interaction. Berlin and New York: Mouton de Gruyter. YALDEN, J. (1 983). The communicative syllabus: Evolution, design and implementation. New York: Pergarnon.