Zambeziu (2001), XXVIII u)-MANAGING 'FACE'1 IN URBAN PUBLIC TRANSPORT:POLITE REQUEST STRATEGIES IN COMMUTEROMNIBUS DISCOURSE IN HARAREPEDZISAl MASH1RIUniversity of'Zimbabwe, Department of African Languages and LUerutureAbstractThis article examines the request strategies in commuter omnibus discourseinvolving the bus crew (conductors, touts and drivers* and passengers, inHarare The present study considers requests in commuter transport as 'uct:threatening acts (FTA), hence the need, for the commuter crew to strategical!}shape their communicative actions to achieve their overall discourse goal ofgetting passengers to perform actions that are in their own interest withminimum resistance or confrontation. The crew present itself by usingcommunicative devices that prompt the passengers to evalua.e it positivelyas warm, friendly, modest and respectful. However, the passengers' responsesto the requests range from compliance to resistance depending on theirinterpretation of the speaker's motive and the probable social consequences.INTRODUCTIONThe study explains polite request strategies of commuter drivers,conductors and touts (henceforth the crew) to (potential) passengersand the passengers' responses to these requests in commuter omnibusdiscourse. This omnibus discourse is of interest and importance for anumber of reasons. First, it contributes to knowledge in one ratherneglected area of research in sociolinguistics: communication in publictransport discourse. While interest in research in the study of spokendiscourse within specialized, institutionalized contexts - Christiansermonic discourse (for example. Mashiri, 2000c; Dzameshie, 1995),political discourse (for example, Love, 2000; Barton, 1999; Lu, 1999).television interviews (for example, Owsley and Myers-Scotton, 1984J,family discourse (for example, Ervin-Tripp et al., 1984; Blum-Kulka. 1990)1 The notion of 'iace' is derived from Goffman's (1967) theory of the seii as a social/publicor interactive construction. As Goffman notes, face is the positive social value a personeffectively claims for oneself through what others see him/her do or hears him/her sayThus, face is one's public self-image that can be lost, maintained, or enhanced, ininteraction, it is in every participant's interest to act or speak in ways that do nothumiliate or embarrass others or themselves. In discourse, politeness is one speechstrategy for maintaining face.8586 POLITE REQUEST STRATEGIES IN COMMUTER OMNiBUS DISCOURSEhas increased in recent years, discourse in the commuter omnibus speechsetting, unique in many parts of Africa, has generally been overlooked.Thus, the information provided by this study about the types of politenessrequest strategies used in this setting contributes to knowledge on theethnography of communication in one important social context inZimbabwean society.Secondly, this article is important from a theoretical viewpoint. Bycombining key tenets from Labov and Fanshel's (1977) rules of requestsmodel, Brown and Levinson's (1987) politeness theory and Myers-Scotton'smarkedness model, this study shows that "a multiple model of discourseis an effective way of explaining linguistic choices" (Dzameshie, 1995,192).Thirdly, one of the challenges that has confronted Zimbabwe sinceindependence in 1980 is the transport crisis, which affects some 60-80%of the 1.5 million workers in Harare2 who commute daily. The crisismanifests itself in many different forms that include overcrowdedcommuter transport services and violence and sexua! harassment againstcommuters at the bus ranks and in the vehicles, especially when returninghome from work in the evening.Given these dep'orable conditions and the unfriendly relationsbetween the crew and the passengers, the primary research questionaddressed by this study is: why do the commuter omnibus crew adoptpolite or mitigated requests in communication with passengers? As inSouth Africa (Khosa, 1998, 81) and in other cities in developing countries,the experiences of female commuters in Zimbabwe are different fromthose of male commuters. Therefore, this study also intends to find outwhether or not the requests used for male commuters differ in any wayfrom those used for female commuters, and if they do, what the motivationfor the distinction is. As the majority of the users of public transport inZimbabwe (Harare, in the case of this study) are African, and mainlyShona speaking, Shona is considered as the primary language of commuteromnibus discourse examined in this study.THEORETICAL FRAMEWORKThe primary theoretical framework adopted here is Labov and Fanshel's(1977) request model. The core idea of the request theory is that thereare "speech events in which speaker, A, uses verbal means to accomplishThese estimates were obtained from personal conversation with Dr Odero of the Ruraland Urban Planning Department of the University of Zimbabwe. Although the 1997 Inter-censual Demographic Survey Report estimates the population of Harare to be 1 871 943(Inter-censual Demographic Report, 1997, 23), a significant number of people working inHarare commute from/to Chitungwiza, Norton and Ruwa daily.P. MASHIRI 8^the end of getting the listener, B, to do something" (Labov and Fanshel,1977, 77). All such speech acts are subsumed under a single categorycalled "request for action".3 Structurally, requests for action may containimperatives or mitigated forms and, indexically, solicit information,conformation, attention or approval. In response to a request from thespeaker, the listener has various options: s/he may give the speaker theinformation, conformation, or whatever is requested, or s/he may carryout the action or suggestion, s/he may put off the request with anaccounting, or s/he may refuse it, with or without accounting. Labov andFanshel's theory limits requests and responses to speech acts. The presentstudy refines it to include extralingustic behaviour as part of politenessstrategies that may be used by either the speaker or the listener.In order to explain the reasons for the commuter omnibus crew'schoice of mitigation strategies in making (their) requests, it has beennecessary to adopt Brown and Levinson's (1987) politeness theory. The'kernel idea of the politeness theory is that certain speech acts areintrinsically threatening to face and thus require 'softening' by means ofpoliteness strategies (PSs)" (Brown and Levinson, 1987, 24; Mashiri, 2000a,61-62). The theory focuses on the effects of linguistic choices on the facewants of the hearers, "whether they feel approved of, liked, respected, orotherwise" (Dzameshie, 1995, 193). Thus, the theory is audience-centredand it emphasizes softening of the request as the main purpose for usingPSs.However, the present study redefines the overall motivation for theuse of PSs. We argue that in addition to using PSs in order to maintainface, speakers use them to accomplish their overall discourse goal, thatis, "what the speaker aspires to achieve in the speech act (Saville-Troike,1982; Hymes, 1972). This overall-discourse-goal explanation for the use ofPSs is made more explicit by integrating the politeness theory withMyers-Scotton's (1983) markedness model. Unlike the politeness model,Myers-Scotton's model is speaker-centred. It shows how speakers managetheir linguistic choices in order to achieve their discourse goals. Thismodel is based on a number of underlying assumptions, two of which arehighlighted here.Requests are determined by rules for making these requests directly or indirectly. Therules of request show the audience when they are seriously requested to perform anaction or not. There are four preconditions for making valid requests; needs, abilities,obligations and rights. These preconditions also provide the major modes of mitigationand aggravation. There are general principles that seem to determine whether or not aform is mitigating or aggravating. Needs and abilities are generally mitigating and rightsand obligations, aggrava'ing. In Shona there are inorphosyr.tactL dev.cej for mitigatingand aggravating requests.88 POLITE REQUEST STRATEGIES IN COMMUTER OMNIBUS DISCOURSEFirst, like Labov and Fanshel's model, Myers-Scotton's assumes that"all linguistic code choices are indexical of a set of rights and obligationsholding between the participants in a communicative event (Myers-Scotton, 1988a, 152). Secondly, the markedness model assumes that allspeakers have implicit social knowledge and facts that guide them inchoosing one linguistic variety over the other, as part of theircommunicative competence. According to Myers-Scotton, all speakershave mental representations of a correlation between code choices andrights and obligations sets. Thus, they tacitly know what code choicesconstitute the marked or unmarked linguistic realization of "an expectedrights and obligations set" (Myers-Scotton, 1988a) between the speakerand the hearers in a specific interaction.4DATA COLLECTIONData were obtained from six ranks, three for commuter buses travellingto high-density suburbs and the other three for those travelling to low-density suburbs. Additional data were collected from the interactionbetween commuter personnel and passengers on ten different busesequally apportioned to the fore-mentioned destination categories travellingduring both peak and non-peak periods. Both the ranks and the buseswere randomly selected. But, the rationale oi sampling ranks andomnibuses according to destinations and commuting time was to find outwhether or not patterns of requests and responses changed with thesevariables. The data were obtained by different methods.Participant observation by the present researcher a ul studentresearch assistants since 1997 was the main method of collecting examplesof requests and responses to the requests. All identifiable requests andresponses occurring in the chosen settings between sets of partu ipantswere systematically written down. Unstructured interviews. m-J informaldiscussions were made with randomly selected commuter ti: ivers and/orconductors and commuters and recorded to find out thei .'.uthation,intentions for and interpretations of the requests and re ;-.y. ses thatensued from the interactions on the buses.WHY USE POLITENESS STRATEGIES?The use of direct requests (commands) often creates an activeconfrontation situation, both on the rank and on the bus. The c >nfrontation4 The relative markedness of linguistic choices is interaction-type specific A s.ilient aspectof Myers-Scotton's markedness model is her negotiation principle (see \1vers-Scotton,1983, 116). In this study, negotiation refers to the making of ratior.il c-^il-orientedlinguistic (and extra-linguistic) choices as a means of achieving a spe.iki is specificdesired discourse goal.P. MASHIRI 89arises from the clash between what the bus crew m?/ perceive as validrequests for action and their prerogative, on the one I ind, and what thepassengers see as their willingness and rights, or the other hand. AsDzameshie (1995, 206) notes, "within a given discourse, participants'tacit knowledge of their rights and obligations partly influences theirlinguistic choices". If a conductor, for instance, makes a direct requestthat s/he expects (a) passenger(s) to comply with but receives a negativeresponse or a rejection, s/he may feel challenged or affronted by theresponse. Invariably, the conductor, in the same discourse or insubsequent contacts may utilize politeness strategies (PSs) to soften his/her requests or to achieve compliance and/or co-operation.TYPES OF POLITENESS STRATEGIESThe most recurrent PS used by commuter crew is the use of morpho-syntactic constructions. In the commission of face threatening actions(FTAs subsequently), the hortative inflection nga- 'let' together with theinclusive 1st person plural subject pro-form -ti- 'we/us' occur veryfrequently as mitigating markers. The use of the hortative marker and theinclusive pro-form is not only meant to encourage the passengers torespond favourably, but to promote cooperation between the speakerand his/her hearers. Consider the following examples:1. Mungabatanidzewo here kumashure uko?'Could you people at the back seat put together your bus fare andpass it forward please?'2. Ngatibatanidzei.'Let us put our bus fare together and pass it forward.'3. Ngatisebedzanei, tikwane tose.'Let us push so that all of us may fit.'The meaning of the pro-form -ti- would be:-SpeakerBut in the interest of the speaker+ AddresseeThe principle function of the use of -ti- is to get others to do an actionthat is in the speaker's own interest. In this case, the use of both thehortative marker and the pro-form has a mitigating effect. The interviewsthat were made with some conductors reveal that this linguistic choice isprompted by previous experiences of rejected requests. In some instances,the mitigation is emphasized either by accounting as in, tikwane tose in(3) above, or by 'style shifting' (Pearson, 1988, 79). Style shifting includesthe marked use of relational terms of address and slang terms, and asBrown and Levinson (1978,116) claim, by using slang, the speaker evokesall associations and attitudes that he and the hearer share towards the90 POLITE REQUEST STRATEGIES IN COMMUTER OMNIBUS DISCOURSEdesired goal. Social distance is reduced. The most commonly used termsare muface 'my acquaintance' and mudhara 'pal but, lit: old man', in theutterances below:4 Pinda tiende muface.'Get in, so that we go, my acquaintance.'5. Handei kuback seat mudhara.'Let us go and sit at the back, pal.'Young male conductors in addressing male passengers that theyperceive as their age-mates invariably use these slang terms. As Lederer(1997, 5) observes, "slang allows us to break the ice and shift into a morecasual and friendly gear." In this case, the conductors use slang to evokesolidarity while negotiating their own returns. The slang terms also enablethe conductors to invoke and manipulate the traditional assumptionsabout manhood that are shared by them and the hearers. Men are expectedto take risks and to endure pain and rough conditions. Therefore malepassengers tend to respond favourably to the requests to travel as standingpassengers, squeezed in corners or seating on falling seats, for fear ofbringing their masculinity into question. Female conductors do not usethe same politeness markers as their male counterparts to refer to malepasbengers nor do they use any for female passengers. Similarly, maleconductors do not use slang terms for mitigating their requests to womenpassengers. Instead, they use kin terms as relational social honorifics.For example, consider the requests:4. (a) Pindai tiende sisi'Get in so that we may go, sister.'5. (a) Handei kuback seat amai.'Let us go and sit at the back, mother.'Although, denotationally, the terms sisi sister'and amai 'mother' areused to designate family relations among relatives, honorifically, however,they (and others) are used connotationally to maintain and enrich socialinteraction among both related and unrelated participants. Commentingon the use of relational social honorifics in Jordanian Arabic, Farghal andShakir (1994, 242) said, "in fact, the best way to get the attention of astranger on the street is to use an honorific kin term". In the examples ofutterances given in 4 (a) and 5 (a), above, the utterances are doublymarked for politeness by using them along with politeness formulas. Kinterms are used as politeness-enhancers. In all the cases that the researcherobserved the kin terms being used, the speakers' intended goals wereachieved with minimum rejection or huff.The commuter crew also softens their requests by using other slangterms that connote a higher status for the people so addressed. The termvarungu 'lit: white people' is used to refer to both female and maleP. MASHIRI 91passengers, but shasha, 'prominent person', bigaz, 'lit: big' and mudhara,'lit: old man' refer to men only. The following are examples of requests inwhich these slang words are used:6. Farirai mota yenyu varungu. Ngatipindei tibvu iasimuka. Parirenyatwa,Avondale neGreencroft zvose. Zvarnatouya kudayi zuatonaka varungu.Tatoraramaso. One asara! Pinda tichienda shasha. Ngatisebedzaneikabigaz.'Get onto your bus happily our employers. Let's get on so that weleave immediately. All those wry": go to Parirenyatwa, Avondale andGreencroft please get in. We are happy to see you our employers.We depend on you for survival. One last person! Jump in so that weleave immediately pal. Lets push so that we all fit, big one.'7. [At a rank in the city centre: a young male !out - to a well-dressedelderly man who appears to be a regular patron] Maswera seimudhara? Mava kumbonozororaka? Mirai zvenyu ipapo mugokwira irikuuya iyo. [To the conductor of an approaching bus] Kumberi kunemunhu kudhara Masvingo. [To the regular patron] Pindai mberimudhara. [The man sits on the front seat and the tout hangs aroundchatting to him. The man gives the tout a twenty-dollar note and hedisappears quickly to control the other passengers].[At a rank in the city centre: a young male tout - to a well dressedelderly man who appears to be a regular patron] 'How was your day oldman? I see you are now returning home, aren't you? You may waif for awhile then you will get onto that bus coming after this one. [To theconductor of the approaching bus] The front seat is reserved for someone,hey Masvingo. [To the regular patron] Take the front seat old man.Mawadza (2000b, 95) rightly observes that,Passengers on a commuter bus are often referred to as uarungu, becausethey are regarded as employers. This stems from the fact that whitepeople were, for the most part, the employers in the colonial period.Thus, without Varungu' on the commuter bus, there would be nobusiness. The term murungu 'lit: white person' is sometimes also usedto flatter an individual in order to elicit favours. Those who use theterm in this manner expect that the person referred to will feel flatteredenough to offer a big tip for the service rendered. This is because, inthe colonial period, the term was associated with the popular image ofthe white person as being wealthy and powerful, and, therefore, likelyto give generous tips . . .The utterances in (6) and (7) function as formal casual requests,while those in (4) and (5) function in informal requests. The formalityparameter is often governed by extralinguistic factors; for example, theuttering of (6) or (7) is usually prompted by the fact that the addressee iswell-dressed and assumes an air of importance, thus deserving the term92 ROUTE REQUEST STRATEGIES IN COMMUTER OMNIBUS DISCOURSEiigu. shasha, bigaz or mudhara. instead of muface. The terms bigaz,siiushu and mudhara raise the hearer to a status where complaining,overtly rejecting the tout's request or failing to otfer him a tip embarrassesor belittles the hearer Thus, the use of these terms, in addition toevoking politeness, guarantees compliance and accommodation.lii addition to style shifting, the commuter crew uses directimperatives qualified by accounted statements as a politeness device.Consider examples 8-10 below where the account is in bold in eacii case.8 Musamire pamusiwo, driver haazoona zvakanaka.'Do not stand on the door way because you will obstruct the driver.'9. Sebedzanai, vamwe vakwanewo.Please, push backwards so that there is space for other people.'10. Vari kudzika paPolice fambai muuye mberi, vamwe vanonokakubasa.Those disembarking at the Police bus stop, move forward please toavoid delays that make others late for work.'Three iorms of responses were observed. To the requests in (8)-(10),silence was the most common form of putting off the request if the buswas full and was obvious that there was nowhere to move to. Simpson(1989, 171) and Labov and Fanshel (1977, 63) concur that silence is oneform of resisting expressing agreement and disagreement with the speaker.Interviews conducted with both male and female passengers revealedthat silence is motivated, either by the desire to avoid dispute with thecrew or the tear of being attacked or embarrassed if the disputedegenerates into a verbal or physical fight. Sometimes, if a conductor, forexample, is annoyed by the noncompliance, he may either repeat therequest or challenge the passenger(s) concerned to drop off the bus atthe point of dispute. To avoid the latter, passengers may reinforce theirrefusal with an accounting as in (11)-(13), in bold.M.Hapana kwokuenda, bhazi rakazara.'There is nowhere to move to since the bus is already full.'12. Tichisebedzana tichienda kupi iro bhazi risingatatamuke?There is nowhere to push to because the bus doesn't stretch, doesit?'l3.Handikwanise kufamba, ndine mwana mukuwasha.'1 cannot move (while the bus is moving) because I have a smallchild with me, sir.'When a conductor isolates an individual whom he perceives asthreatening his status and rights by maintaining the challenge, theconductor may counter the challenge by exercising his power to orderthe passenger to get off the bus. The present researcher witnessed severalincidences where either a passenger, (male or female) refuses to obeythe conductor's order to get off the bus and vicious insults are exchangedP. MASHIRf 93all the way to the passenger's destination, 01 the conductor ..tops the busin order to physically drag the passenger off the bus. In five such cas^sobserved by the present researcher, three of them involved interventionby the rest of the passengers to defend the passenger(s) implicnied. Ineach case, the conductor succumbed to the 'mass' protest, which rangedfrom questioning his action, perceived as threatening the negative 'face'of the passengers, to threats of performing a citizen's arrest. This act ofsolidarity by commuters is most common where the implicated passengeris a woman, an elderly person or one in the company of friends orrelatives.Further, polite questions are widely used as requests by bus commuterconductors, as seen in (14) below.(14a) Tabhadhara tose here?(Have we all paid our fares?)(14b) Pane vasina matiketi here?(Is there anyone who has no bus fieket?)(]4c) Mungaitewo hare mari yakachinjika?(Would you please have small denominations?The utterances in (14) above function as a polite reminder and apolite question, respectively. For instance, a conductor to an absent-minded passenger, thus reminding her/him to pay the bus fare may utter(14a) and (14b). Interviews with some conductors revealed that manypassengers tend to present large bills, e.g. a $100 note when paying for atrip costing only $20. While passengers believe that the bus crew isobliged to find change for whatever amount is presented, the latter insiston the correct fare or small denominators. This conflict of obligationsand expectations often result in confrontations, hence the need to usepolite questions as in (14c) as a PS.CONCLUSIONThe findings and explanations presented in this article revealed thepredominance of polite requests as a speech act in commuter omnibusdiscourse in Harare. This analysis reaffirms the usefulness of a multi-dimensional model of discourse in which a speaker's intent is primary.The intent may derive from, among other things, more static situationaland relational constrains. In the commuter bus setting in Harare theinteractional relations between the commuter crew and passengers isvery volatile, hence requests tend to be FT As. In dealing with passengersthe commuter crew negotiates both co-operation (solidarity) andcompliance by adopting PSs.The importance of PSs in requests is explainable in terms of theillocution. First, the commuter crew considers them the most effective94 POLITE REQUEST STRATEGIES IN COMMUTER OMNIBUS DISCOURSEcommunicative acts needed to make passengers perform or not performcertain actions. In seeking a favourable hearing to/ its requests, the crewemploys various linguistic forms that include inclusive proforms, relationalterms and style shifting. However, other factors such as the perceivedage, social class and gender of the addressee(s), the sex of the conductor,especially, and the cultural norms also determine the linguistic choicesrelating to requests and the responses to these requests.Second, the conflict between the commuter crew and the passengerson perceived rights and obligations also set the tone for politeness incommuter bus discourse. By forgoing their inherent rights and privilegesassociated with their role as transport 'managers' the commuter crewstrengthens their negotiation power, preserves face, averts altercationswith passengers, maintains their clientele and achieve their discoursegoal.Since the emphasis here has been on FTA-related PSs, facepreservation is an important motivation for the use of PSs by the commuteromnibus crew. However, the present research also recognizes that asrational goal-oriented speakers, the commuter crew uses PSs primarilyas strategic linguistic choices that they "skillfully manipulate in attemptsto achieve their discourse goal" (Pearson, 1988, 71) of gaining favourableresponses to their messages. In other words, face preservation is only ameans to an end: the attainment of the speakers' discourse goal.ReferencesBARTON, ELLEN, L., (1999) "Information and Interactional Functions ofSlogans and Sayings in the Discourse of a Support Group", in Discourseand Society: An Interactional Journal or Discourse and Communicationin the Social, Political and Cultural Contexts, 10 (4). 461-86.BLUM-KULKA, SHOSHNA (1990) "You Don't Touch Lettuce with Your Fingers:Politeness in Family Discourse", Journal of Pragmatics, 14, 259-88.BROWN, PENELOPE AND STEPHEN C. LEVINSON (1987) Politeness: Universal inLanguage Usage (Cambridge. Cambridge University Press).DZAMESHIE, ALEX K. (1995) "Social Motivation for Politeness Behavior inChristian Sermonic Discourse", in Anthropological Linguistics 27 (ii),192-215.ERVIN-TRIPP, SUSAN, MARY C. O'CONNOR AND JARRETT ROSENBURG (1984)"Language and Power in the Family", in Cheris Kramarae, MurielSchulz and William M. O'Barr (eds.), Language and Power (BeverlyHills, Sage Publications), 116-35.FARGHAL, MOHAMMED AND ABDULLAH SHAKIR (1994) "Kin Terms and Titles ofAddress as Relational Social Honorifics in Jordanian Arabic", inAnthropological Linguistics 36 (ii), 240-53.GOFFMAN, ERVING (1967) Interaction Ritual: Essays of Face to Face Behavior(New York, Garden City).P. MASHIRI 95HYMES, DELL (1972a) "Models of the Interaction of Language and SocialLife", in Gumperz, John J. and Dell Hymes (eds.) Directions inSociolinguistics (New York, Holt, Rinehart and Winston), 35-71.KHOSA, MESHACK, K. (1998) "Sisters on the Wheel: Gender Relations in theTaxi Industry in Durban", in Anita Larsson, Ann Schlyter and MatselisoMapetla (eds.), Changing Gender Relations in Southern Africa: Issues ofUrban Life (Lesotho, The Institute of Southern African Studies).LEDERER, R. (July 1997) "Sailing on Slang", in Writers Digest (July, 77) (7).LOVE, ALISON (2000) "Democratic Discourse? Realising Alternatives inZimbabwean Political Discourse", in Zambezia 27 (I), 27-45.Lu, XING (1999) "An Ideological/Cultural Analysis of Political Slogans inCommunist China", in Discourse and Society: An International Journalof Discourse and Communication in Their Social, Political and CulturalContexts, 10 (4), 487-508.MASHIRI, PEDZISAI (2000a) "Street Remarks, Address Rights and the UrbanFemale: Sociolinguistic Politics of Gender in Harare", in Zambezia 27(I), 55-70.(2000c) "The Value and Relevance of Orality in the Methodist ReligiousPerformance", in Chiwome,.Emmanuel M., Zifikile Mguni and MunasheFurusa (eds.), Indigenous Knowledge and Technology in African andDiasporan Communities (Harare, Southern African Association forCulture and Development and National Council for Black Studies Inc,California).MAWADZA, AQUIUNA (2000b) "Harare Shona Slang: A Linguistic Study", inZambezia 27 (I), 93-101.MYERS-SCOTTON, CAROL (1983) "The Negotiation of Identities inConversation: A Theory of Markedness and Code Choice", inInternational Journal of the Sociology of Language 44, 115-136.OWSLEY, HEIDI H., AND CAROL MYERS-SCOTTON (1984) "The ConversationalExpression of Power by Television Interviewers", in The Journal ofSocial Psychology 123, 261-71.PEARSON, BETHYL (1988) "Power and Politeness in Conversation: Encodingof Face-Threatening Acts at Church Meetings", in AnthropologicalLinguistics 30 (I), 68-93.SAVILLE-TROIKE, MURIEL (1982) The Ethnography of Communication: AnIntroduction (Oxford, Basil Blackwell).SCHIFFRIN, DEBORAH (1994) Approaches to Discourse (Oxford, Blackwell).SIMPSON, PAUL (1989) "Politeness Phenomena in lonesco's the Lesson", inCarter Ronald and Paul Simpson (eds.), Language, Discourse andLiterature: An Introductory Reader in Discourse Stylistics (London, UnwinHyman).