Situated language Language does not occur in a vacuum. It is situated, contextuatized. There are speakers and hearers, a time and a place, aj^k andaj>urpo§g. This chapter introduces the main ways in which sodolinguists have tried to describe and make sense of how everyday language is moulded by - and moulds - the situation of its usage. We then go on to examine several approaches to how language operates as part of interaction, The chapter deals with the use of language to do things through speech acts, and the ways in which politeness (or impoliteness} affects how we address another person. It looks in detail at the sociohnguistics of interaction, how that can lead to miscommunication between people, and the ways in which gender dynamics affect conversational patterns such as turn taking. Underlying this is a concern for the micro detail of everyday conversation that is informed by the methods and theories of anthropology, in particular by ethnographic approaches. 6.1 Situations, contexts and domains A situation is the extended social occasion within which speech occurs (Hymes 1972). Situations are as variegated as human encounters - 'card games, ball-room couplings, surgical teams in operation, and fist fights', as Erving Goffman put it in a brief early paper on 'the neglected situation' (1964: 135). Language is one player in situation, although most situations^ are not primarily defined by language but by the shared activity. Tbey may, however, contain conventional language elements, and speech may even be the main activity or prerequisite of what occurs. A committee meeting, a trial in court, a school class ate all speech situations - they cannot in fact take place without speaking, and speaking that is appropriate to the specific situation. The Guidebook to Saciolinguistics, First Edition. Allan Bell. © 2014 Allan Bell. Published 2014 by Blackwell Publishing Ltd. 132 Situated Language Situated Language 133 'Context' is a term that is widely used in sociolmgusstics even when scholars question its appropriateness • How justifiable is it to treat language and context as separate entities? • How practical is it not to treat them as separable9 See for example the essays in Duranti and Goodwin (1992) Closely related to situation is the notion of context. This assumes that a situation con-sists oftwo elements: a core event or activity, and the field of action in which it is embed -ded - the context. Traditionally, these have been treated as two relatively separable components (Goodwin and Duranti 1992), and interaction between them has been regarded as essentially one-way, from the context to the language. But sociolinguists became increasingly aware that knjuage^itselt' shapes_andjej^hapes Jhe. .contextijnd therefore: the overall situation of its own use. Even at the most basic level of who counts as a participant in a situation, language choice can define people as in or out of the frame. In an early phrasing that could represent much of the sociohnguistic work of the early twenty-first century, John Gumperz wrote: 'Sociohnguistic variables are themselves constitutive of social reality' (1982a: vh). Fishman (e.g. 1965) offers a series of questions about language use in society which have been widely used as a shorthand way of characterizing sociohnguistic situations: • who uses • what variety « of what language • to whom • when • where • about what • and to what end. a good deal of sociohnguistic enquiry can be pursued through these questions, and throughout this book we address the issues they raise in different guises (see Gals study in Chapter 5.6, for example). Such a series of questions seeks to investigate 'the norms of language usage - that is to say, the generally accepted social patterns of language use and of behavior and attitude toward language' (Fishman 1971: 219). These questions can be used, for example, to examine the choice between polite and familiar second person pronouns, a common feature of many languages. The so called T/V pronouns in European languages (e.g. French tu versus vous) were the focus of a pioneering study by Brown and Gilman (1960) which characterized them as encoding solidarity versus power. Zierotas £ 2 - -iorass "/stems In a classic article, Susan Ervin-Tripp (1972), one of the,founding sociohnguistic scholars, proposed a framework of 'alternation and co-occurrence: She built:oii the work of Brown and Ford (1964) on address terms in American English^These terms involve choices between addressing your interlocutor by their first name, or by title plus last name, and so on - 'Julia' versus 'Mrs Child! Ervin-Tripp diagrams a flow chart which offers a series of binary alternatives according to the addressee's gender, relative age, etc. Address systems differ across languages and in different-dialects of English. Read the relevant section at the beginning of Ervin-Tripp's article, then interview each other about address terms in your respective languages and dialects: 1 What terms are available in the address system of your informant's language or dialect {e.g first name, titles, etc.)? 2 Give examples of how these terms combine fo make different forms of address. 3 What are the factors that influence choice of address terms in your informant's language or dialect? 4 List these factors and diagram them in a flow chart similar to Ervin-Tripp's If there are too many factors or outcomes, Just diagram the main parts of the system. 5 Report back and draw generalizations and conclusions on the basis of the class's findings. Domains of use Fishman also approached context through his concept of domains (1972). The choice of which code to use in habitual situations in a bilingual or multilingual society is not random. Fishman writes about a Belgian official who tends to use one code (standard French) at the office, another (standard Dutch) at his club, and a third (a local variety of Flemish) at home. The domain proposal aims to capture the regularity of these uses without losing the possibility of individual choice and fluctuation. The description here may represent the official's typical code choices, but he will occasionally use Flemish at the office, French at the club, and standard Dutch at home. Note the links between domain and 'language function, which I introduced in Chapter 3. Function is a macro-level concept which focuses on how languages operate across a whole nation or society. Domain is a parallel micro-level notion, which is concerned with the code choices that individuals or small groups make in particular interactions. Typical combinations of three main elements feed in to make up a domain - the topic of talk, the locale or physical setting and the expected role relations between participants (Table 6.1). The different factors buttress each other, and the likelihood of a particular language code occurring becomes stronger. In the Education domain, for example, there is typically a clustering of participants in defined role relationships (teacher and students), meeting in an educational institution and discussing the content of a course of study. In a bilingual 134 Situated language Situated Language 135 Table 6.1 Characterizing domains Domain Participants Setting Topic Education teacher school coursework Religion priest church, temple religious practice Employment employer workplace tasks at hand Family parent home domestic matters Friendship friend café common interests Source: Adapted from Fisliman (1972), 445 Apply the concept of sociolinguistic domains in your setting: 1 How well do the five domains in Table 6.1 apply to interactions in your society? Do they involve the listed participants, settings and topics, or different ones? 2 What further domains can you identify? How are they defined in terms of participants, settings and topics? 3 Do a domain analysis of one event in your culture such as a workplace interaction or a religious service. Evaluate domain as a concept: 4 Does 'domain' further your understanding of how languages are used by people and groups in your own society? How? 5 Are there any ways in which it limits or skews our understanding? For example, is it too static to capture some - or many - situations? Does it close off alternative ways of looking at individuals' language choices and behaviours? society, this educational domain will characteristically favour one language rather than another - for example, English in Malawi (Chapter 3), or Spanish in Mexico. Research has tended to emphasize those domains associated with institutional public areas such as religion, education or media. However the more private situations with family and friends are also included, as Table 6.1 shows. Domains continue to be used as an organizing principle, for example in Kelly-Holmess study (2010) of marketing language. 6.2 ETHNOGRAPHIES OF COMMUNICATION Situated language has long been the interest of broadly anthropological approaches to sociolinguistics which provide much of the substance of this chapter. The most comprehensive early proposals came from Dell Hymes. Starting with a pioneering article in 1962, he developed a framework which he first called theJ^mg^v£o(_m^}M-Ethnography is the core method of social anthropology. It involves a close description of a community's way of Jife - what they do togejherf how they interact with each other, how they see themselves and the world. It is mainly the fruit of participant observation: the researcher observes the everyday life of the group through long-term involvement and experience, whether a plateau village in Madagascar (Keenan 1989) or a gang of teenage Latina girls in northern California (Mendoza-Denton 2008). The ethnographer learns the group's language and practices, often over the course of a year c>r more and with subsequent return \ istts. She keeps detailed fieldnotes of what she sees and hears, makes audio or video recordings, asks questions, probably conducts interviews. She aims to set aside her own preconceptions and to understand the group in its own terms, but also to maintain an independent view on what she finds. The result is a 'thick description which leadsja arumderstanding of the group and its practices - including clarifying in what sense it is a group' at all. Blending ethnography into the study of language, Hymes's formulation of the ethnography of communication was typical for the early stages of an academic discipline. He advocated the value of developing ajaxonomi^a structured and stratified hst of the elements of a communication situation, like that used to classify plants into families, species and so on. This taxonomy aimed to help the ethnographer of communication observe, specify and interpret the patterns of a group's communicative behaviour. In a series of publications Hymes offered related - although not identical - listings of the 'components' of the speech situation. The best-known and most complete list contains 16 terms (Hymes 1972), which he groups under eight headings whose initials form a mnemonic spelling out the word SPEAKING (Table 6.2). Many of these components have been mentioned in earlier chapters, and others will come up later. This is no surprise since they include most of the basic parameters surrounding the use of language. The research activity at the end of this chapter applies the SPEAKING taxonomy to analysing the classroom situation. Situation consists of the circumstances in which a particular communication occurs. Setting (1 in Table 6.2) means the physical parameters of time and location - for example, 6.00 pm on a commuter train in New York City. Scene (2) is the culturally defined occasion that is involved - language happens very differently at a rock concert compared to a cocktail party. Participants are the people who are taking part in a situation where language is being used. Classifying and illustrating participant roles and the part they play in language situations has attracted attention from successive sociolinguists because participants are arguably the key component of any interaction. I deal with Hymes's version of these roles only briefly, and return later in the chapter to look at more detailed formulations. Two of Hymess four categories identify the people producing the language. The Speaker/ sender (3) is the authority behind something that is being communicated, the person responsible for the message. The Addressor (4) is the physical producer of the speech, the voicebox. Two categories distinguish types of receivers - Hearers/audience (5) who are not being directly targeted, and the Addressees (6) who are. Ends are the purposes of a particular communication. Hymes divides this into two: Outcomes (7) are the general goals of an activity within a culture. For example, a university graduation ceremony involves staff and students acting out a rite of passage in front of a supporting audience. Goals (8) on the other hand are individual participants' 136 SiTUATED LANGUAGE Situated language 137 Table 6.1 The SPEAKING taxonomy 10 11 13 14 15 16 Setting time, place, circumstances Scene 'psychological setting' - the cultural definition of an occasion Speaker/sender person/authority behind the message Addfessor person who voices the message, spokesperson Hearer/audience person/s who hear the message Addressee person directly addressed Purposes - outcomes general purposes within the culture Purposes - goals individual participants' goals Message form how things are said Message content topic, theme Key tone or manner in which something is said Channels spoken, written, etc. Forms of speech language codes - varieties, languages, etc. Norms of interaction cultural 'rules' for conducting conversations Norms of interpretation listeners or observers' interpretations of the rules Genres language associated with particular speech events SITUATION PARTICIPANTS ENDS ACT SEQUENCE KEY INSTRUMENTALITIES HORMS GENRES Source: Drawn from Hymes (1972) intentions in relation to the overall event, for example family members who come to support a graduand. The difference between outcomes and goals is most obvious when there is a clash between the general and the individual purposes. Say a political dispute is underway between students and university management, and some students at the graduation ceremony voice their protest - then their goals differ from the outcome and from the goals of most other participants. The importance of participants' goals in communication has been a particular interest of social psychologists of language. Act sequence involves two components. The Message form (9) covers the way things are said - in fact, the stuff of most language analysis. The Message content (10) is what is said - the topic or theme. Different topics wili often trigger different ways of speaking - we discuss road deaths differently from local gossip. ■'.;)Exeroise 8-4' 'Topic and language choice Topic is an important/category in sociolinguistic: enquiry -note for example its role in-Fishman's domain construct. In many workplaces around the world, technology may be discussed in. English because technological advancement: is associated with English. And surprisingly, in a traditional society an external language such as English may be used for intimate topics because traditional norms amply do not allow them to be discussed in She indigenous language. Rickford and McNair-Knox (1994) detail how one speaker, Foxy, shiftedher linguistic style depending on whatshe. wastalking. about from teen pregnancies to career issues to drugs to race relations. • . Do: you believe thetopic. you are discussingaffectsthe way you andother people, in your speech community talk? Mow? For-example,.aresome.topics discussed in one language but not another? Why? Key (11) is the manner or tone in which something is presented. What is superficially the same thing may be said in different ways. Conversationalists, like actors, can impart to a line widely different 'readings' through different keyings. The most obvious manifestation of key is when the way something is said subverts the meaning so that it becomes the opposite of what the surface of the words appears to intend - that is, irony. A change of key is often marked in the intonation. In English 'That's brilliant!' with a falling intonation is a criticism, the opposite of the same exclamation with a neutral rise-fall intonation. Instrumentalities include the Channels (12) by which the language is carried, such as writing versus speech, or sung versus spoken. This has become an increasingly important dimension as technological change has created new communication media, especially through the many affordances of the internet. Forms of speech (13) covers the repertoire of codes that are available to speakers in a community (see Chapter 5). Norms are the rules governing the production and interpretation of language in a situation. Norms of interaction (14) are the community's rules for communication, and are most easily visible when someone breaks them, for example in an encounter between two contrasting cultures. They include such things as when it is appropriate to take a turn at speech and when to keep silent. For example, Athabaskans leave more silence between speaking turns than do mainstream North Americans (Scollon and Scollon 1981). Such rules for turn taking in conversations are studied in particular detail by Conversation Analysts. The Norms of interpretation (15) show how inter-actants interpret each other's practices. When Athabaskans and mainstream North Americans interact, both follow their own norms for silence after each turn. The Athabaskans find the others rude or dominating because they themselves can never get a word in edgeways. The others think the Athabaskans are sullen or superior because they will not talk. Both groups end up confused and frustrated by each other. Genres involve the language associated with a particular kind of speech event: a story, a job application, a blog, and so on. Different Genres (16) have different rules 138 Situated Language The internet has opened jp new channels of communication that did not exist until the late twentieth century. It has also created new genres, and changed existing genres (see for example Rovve and Wyss 2009. Thurlow and Mroczek 2C11). Marty people believe it is having a major impact on our language (see e.g. Crystal 2006). 1 What new genres have been developed on the internet? List as many as you can, and describe their similarities and differences, using appropriate terms from the SPEAKING framework. Have they introduced linguistic changes? 2 Also consider how new these genres in fact are - did they originate in pre-internet channels and genres? (see for example McNeill 2009 on the origins of the blog 3 How have traditional mass media (such as press, radio, television) colonized the internet? How has this changed them and their language? (Use the SPEAKING framework to compare for example a hard-copy edition of a newspaper and its internet site on the same day - see Bell and Smith 2012.) for language choice and use. New technologies create new genres, most obviously in the burgeoning of internet forms. Genre is a slippery concept which eludes easy definition but which many analysts have found indispensable to understanding language use. The research activity later in the chapter offers the chance to apply and evaluate Hymes's taxonomy. In the process, it will become clear that some of the terms are overlapping, obscure, counter-intuitive, or simply selected to fit the mnemonic. Despite this, Hymess listing serves as a wide range where sociolinguists may browse and find something appropriate and usable for their analysis. Later sociolinguists have adapted this kind of taxonomy for their own purposes, for example Scolkm and Scollon's 'grammar of context' in intercultural communication (2001). Applied judiciously and perceptively, the analysis of such components makes a good starting point for understanding how language works in specific situations. 6.3 SPEAKERS IN SITU Earlier we saw briefly how Hymes, as part of his SPEAKING taxonomy, distinguished just two speaker and two hearer roles. A more detailed approach comes from Erving Goffman, the sociologist of everyday life. Goffman's construct of the 'interaction order' was an important influence on sociolinguistic approaches to speaking situations, and in his later work he himself engaged extensively with sociolinguistic issues. In an essay on 'footing', Goffman (1981) laid out what^he called the 'production format' in which speakers operate, recognizing three speaker roles: the Animator is the voice box, the Author is the strategist who formulates the message and the Principal is the authority Situated Language 139 Table 6.3 Roles in language production-" Hymes 1972 Goffman 1981 Bell 1991b Sender Principal Principal [Author Author lEditor/s Addressor Animator Animator Source: Bell (1991b), table 3.1 behind the words. Goffman thus splits Hymess role of sender' into two (see Table 6.3). In the default case of ordinary face-to-face conversation, the speaker is the sole 'first person, playing all three of Goffman's roles simultaneously. But even these prove to be not fine enough for the complexities of many institutional production formats, especially in mediated communication, where a fourth role is required. In my own research on the generation of news language (Bell 1991b), I subdivided the author' role again to allow for the fact that there can be numerous authors' but not all of them play the same function towards the text (Table 6.3). The principal is the institutional authority behind a speech or text, whose position or stance is being expressed. There is an initial author who originates the first draft of a news story. But that story then passes through the hands of a number of editors who process and modify the text until it is presented or published in its final form by the animator. These roles can be applied to the analysis of language events of all kinds, including internet texts. The straight listing of roles implies a linear chain, which in Table 6.3 looks like a production line from Principal ->Author —» Editors -> Animator. But what is going on can better be regarded as an embedding or layering or nesting of one role within the next rather than as^ a line, as_G^ffinan noted (1981: 147f£). To illustrate again through the news process: each separate role in fact represents its own distinguishable speech event which could be charac-terrzed using the full apparatus of Hymes's SPEAKING taxonomy. Journalists work in a particular setting, with particular purposes and norms of interaction. They pass a text to copy editors, who are working in their own, different situation, then on to newsreaders in theirs, and so on. And each of these situations and their resultant texts is embedded within the next (Bell 1991a). Clearly such a detailed description would be large and cumbersome.Jjut traces of the many_layers sometimesshow through in the finished product, the published text. When a story published in New Zealand describes a location as '129 kilometres west of London', we can safely deduce that the original said 'SO miles' but the copy editors have neglected to round up the conversion to '130 kilometres'. The concept of embedding is extraordinarily important for understanding how a lot of both everyday conversation and mediated language production work. Goffman describes the jajJEijngs or 'laminations' which speakers routinely introduce into their talk as they recount stories or incidents about themselves or others. The production f format of the story scenario is embedded into that of the story's telling. Each such embedding changes the speaker's footing - their 'alignment, or set, or stance, or posture, pr projected self (Goffman 1981:128). Contemporary sociolinguistic work also focuses 140 Situated language Situated Language 141 1 Review the three approaches to speaker roles (Hymes, Goffman, Bell) outlined in Table 6.3. 2 Choose one situation of institutional face-to^ace communication -e.g. a service •-■v. encounter in a government office, or a trial in court - and apply each of the three :: IrameworkatoltintLim, especially:with:reference to. any instances of'embedding!: 3 Or: choose. one: genreoftraditional media communication,'such as proriuction .. of a newspaper advertisement or television commercial, and apply each of the three frameworks to It. 4 . Or: choose one internet genre such as production of a blog or social network site wall, and apply each of the frameworks to It. 5 Use your analysis to compare and assess the usefulness of each framework. Are the greater complexities in the Goffman and Bell frameworks justified by the; increases they bring in insight? How appropriate and helpful is the concept of 'embedding/layering; 6 How would you modify the frameworks in the light of the analyses you have done? ::'.v^y:' on how speakers take up different footings, for example in the course of a us presidential campaign where inconsistency or 'flip-flopping' of candidates' stances may itself become an issue (Lempert 2009). Bakhtin (1981) has maintained that much of our spontaneous speech consists of wordsthat...come:_fromJhejmouJ^of others, and this is nowhere more evident than in news discourse. Most news is based on talk. The information that journalists write up is gleaned mainly from interviews and documents. Fragments from these texts are embedded into the. story, most obviously as written quotations or audio and video dips. As a result, the news story you read or watch consists of a set of textual laminations which would be almost impossible to unravel to its origins. 6.4 audiences FOR language Speakers require listeners. The audience makes up the complementary dimension of language interaction, the second and third persons of the pronoun paradigm. We saw in Table 6.2 how Hymes distinguished between addressees and a wider set of hearers who are not being addressed. Goffman, however, points out that an utterance does not carve up the world beyond the speaker into precisely two parts, recipients and non-recipients' (1981: 137). He makes finer distinctions between ratified and ujirati-tied participants (Table 6.4). Building on Goffman's proposals, I developed a set of roies as part of the 'Audience Design' framework, systematizing and labelling them as in Table 6.4. Audience Design (Bell 1984) treats the speaker, the first person, as the primary participant at the moment Table 6.4 Audience roles Hymes 1972 Goffman 1981 Bell 1984 Participant status Addressee Addressed recipient Addressee Addressed [ Unaddressed recipient Auditor Not addressed Hearer/audience { Bystander: Overhearer Over hearer Not ratified Bystander: Eavesdropper E avesdropper Not known of speech, qualitatively apart from other interlocutors. It then ranks the audience roles according to whether or not the persons are addressed, ratified or known by the speaker. The main audience role is the second person, the addressee, who meets all three criteria -^known, ratified arid addressed. Other, present third persons may be licensed to participate, but they are not dlrecdy addressed - I term these auditors. Beyond them, talk often has a wider audience of unratified third parties. Those whom the speaker knows to be present, but who are not ratified participants in the conversation, are overhearers - parties within innocent earshot of the conversation. Other parties whose presence is unknown to the speaker are eavesdroppers, whether they are intentionally and maliciously hiding 'behind the arras', or whether they hear by chance because they just happen to be within earshot but out of sight. The specific configura-tion of all these roles and relations during any given utterance is, to use Goffman's term, the participation framework' of that moment of talk. These four audience roles are impSicationally ordered according to their status as addressed, ratified and known. Often in an interaction, the physical distance of audience members from the speaker coincides with their role distance, with the addressee physically closest and eavesdropper farthest away. Audience roies are assigned by the speaker, and .visual criteria such as physical placement, body orientation or direction of gaze are often crucial to identifying who is who. The differences between these audience roles have significant sociolinguistic repercussions. We have seen in preceding chapters how speakers choose their language code depending on who their specific addressee is. This is most obvious when the repertoire is multilingual rather than monolingual. In a bilingual group, the language you choose to use may in fact be enough to nominate who it is that you are addressing. Gal (1979: 121) reports an instance from her Oberwart study where two younger people in a farm kitchen were joking together in German. The young man switched out to Hungarian | when he wanted to ask his mother something. The change in language choice briefly and [directly re-classified the mother as his addressee, whereas she had been just auditor to | the German conversation of the younger ones. ? One case where production and audience roies quickly become very complex occurs with translation or interpretation from one language to another. Vigouroux (2010) describes the role and footing complexities involved in a Pentecostal church service for Congoiese migrants in South Africa, where the pastor's French-language sermon must be simultaneously translated into English. Within a monolingual code, speakers can adjust their pronunciations very finely to cater for changes in audiences (Bell 1984: 172ff.) - we return to these kinds of styling abilities in Chapter 11. 142 situated language Situated Language 143 Use the questions from Exercise 6.6 but apply them to audience instead of producer roles. • Specify audience roles in terms of the addressed-rarified-known criteria (Table 6.4), which are especially relevant to distinguishing different internet genres or sites such as blogs or chatrooms. • What light do these roles throw on the issue of what is public and what is private on the internet, e.g. on a socia) media site? Tanguage choice can be influenced not just by the addressee but by the outer audience roles, the auditor and the overhearer. It is common practice in many bilingual communities for even a single monolingual speaker joining a group (thus becoming an auditor) to trigger a switch by the whole group to the monolingual s language so a-, to include die person as a parijanant Continuing use of a language that one participant does not understand would define the uncomprehending hearer as unratified, outside the group. Even overhearers can have this kind of effect on speakers' language choice. Gumperz found an absolute overhearer-determined rule operating in a bilingual Slovenian/German community in southwestern Austria: 'So strong is the injunction against speaking Slovenian in mixed company, that tourists can live in the village for weeks without noticing that any language except German is spoken' (Gumperz 1982a: 47). But overhearer-oriented choices are generally less determined. Gal and Dorian record identical incidents from their two communities, but with opposite results. In Oberwart, a group of bilinguals at an inn switched from Hungarian to German when overhearers at a nearby table requested it. But in East Sutherland, a group of bilinguals in a bar refused a similar request to switch from Gaelic to English. Just as the media situation complicates the roles we need to describe the language producers, so it complicates the audience roles through the kind of embedding discussed earlier, jgnce each interaction situation becomes embedded inside the next, audience roles as well as speaker roles are laminated. A concert performance in front of a live audience can be broadcast (in real time or recorded). The live performance is a situation with its own full set of audience roles. When that performance is then broadcast, embedding it in the mediated situation, the wider audience generates its own, further set of roles. This can lead to trouble: the concert becomes technologically available to an audience which includes groups who were not targeted in the live event. Through mass media, everyone may become overhearers. And through the internet, everyone is potentially an eavesdrop-j. per - unaddressed, unratified, unknown but there - as many have discovered to their cost. 6.5 SPEECH ACTS AND POLITENESS Speaker and audience roles are central to the study of speech acts - the abiding interest of the field of pragmatics, which is one of sociolinguistics' nearest disciplinary neighbours. Pragmatics focuses on language useaticl users. In its narrowest definition, it is concerned with those aspects of the context of communication that get encoded in the language itself - most obviously deictics such as pronouns (who 'she' means in a discourse can only be specified by reference to the context). More broadly, pragmatics deals with the many ways in which discourses are internally organic and rplatpd :o tln-ir cordc.Nt. Its interests therefore overlap with much sociolinguistic work on syntax and discourse. The study of speech acts is founded on the work of the British philosopher, J. L. Austin. In 1955 he presented his central ideas as the William James Lectures at Harvard University, although he died before these were published in 1962. His How to Do Things with Words is a book written with that shining clarity of style which some phUosophers have achieved on difficult linguistic subjects. Speech acts are performatives, Austin argued - utterances that actually perform the action they express. If I say, 'I promise to pick you up at midday', j ust by saying the verb 'promise; I perform the act of promising. Austin suggested that while statements can be true or false ('I picked you up at midday' - 'No you didn't' - 'Yes I did'), performatives can be either 'happy' or unhappy'. That is, they have to meet certain 'felicity' conditions, such as the right configuration of participants. Consider the words, 'I pronounce you husband and wife.' In most countries, the conditions for 'happy' performance of this kind of utterance are severely restricted. The law prescribes who is allowed to say it (a limited set of qualified celebrants), who it can be said to, where it can be said, and even the attitude of the couple. Austin (1962) distinguished three aspects of the speech act: • The locutionary act is the activity of uttering the appropriate wording. 8 Thejjfocjrhorian^^ denotes what act is performed through the utterance, such as complimenting or requesting. • ^n^?lJL°_?iti™lrl.,.?ife£i indicates what the act actually achieves with its hearer - an apology accepted, an offer taken up. The difference between the locutionary and illocutionary act is important. Many differ-■ehlforjnsjrTworto -aine ij.-ocutiim. 1 ask you to do x may be the most overt form of requesting in English, but phrases such as 'would you please' are the more common locutions. Austin found that his initial distinction between constatives (statements) and performatives could not be easily sustained. This opened the way for John Searle, a student of his, to develop a general theory of speech acts (1969). including dividing them into five classes: • assertives (or representatives), such as to state or describe " directives: order, permit, request 8 . commissives: promise, pledge • expressives: thank, apologize • declarations: appoint, declare. (Searle 1976) 144 SITUATED LANGUAGE Situated Language 145 Sociolinguists and speech acts Pragmatics scholars and philosophers approach speech acts quite abstractly, using invented examples to illustrate points, but sociolinguists engage with them as they occur in natural language data. Apologies, compliments and complaints offer a very obvious zone for investigating the social workings of language. The earliest research counted and compared the numbers of particular speech acts used by different groups, along with their topics and functions. Much of that work focused on gender language differences in English (see Exercise 6.11). In many speech acts, the exact configuration of participants is crucial to the success or interpretation of the discourse and interaction, bringing us back to the detail of speaker and audience roles laid out earlier. Some speech acts may be incomplete unless the addressee responds appropriately. 'I bet you is successful only if the addressee accepts the bet. When a complaint is produced before an auditor, that person functions as witness to the complaint, and their response helps determine what the outcome will be. Researching interactions within a Montreal family, Laforest (2009) found that the auditor/witness usually affiliates with the complaint and so collectivizes it, but also works to ensure the situation does not deteriorate too far (Exercise 6.8). Even an unratified over-hearer lurking unintentionally on the edge of interaction may play a role, for example inducing a speaker to mitigate an uttered threat. But researching speech acts does raise issues for sociolinguists. Real recorded data is much less tidy than constructed examples, and it can be difficult to decide which speech act is being performed. Very many iocutionaryacts seem to pertorni more than one iUpi5rtonayhe_sam^^ Is 'H°w do you feel about coming with me?' a question, request or offer? Or is the speaker rcKiiig on just that ambiguity to hedge her positioning betweenimposing on the heareror^exposmg herself to a face-threatening refusal? Researchers are now very conscious that speech acts are multi-functional, that they need to take thorough account of the wider discourse and context in which they occur in order to interpret their meaning, and that the same speech acts may operate differently in different languages and societies. Politeness A major strand of the pragmatic approach to language in use, and especially speech acts, is the politeness model proposed by Brown and Levinson (1987). The term unfortunately conjures up everyday perceptions of 'polite society', etiquette and manners, but here it receives a broader definition. The central construct is 'hce\ a pej$ Mere Tu to laisses pas le temps de reflect* [entre les:entre tes : ■ 26 Pere [Bien oui, Qa e'est: faut que ca retombe la poussiere a un moment donne. 23 * Father I think you do a Sot of things at the last minute. 24 Daughter ■ Not really. 25 -» Mother You don't give yourself the time to think [between your: between your 26 Father [Yeah. Well that's: the dust has to settle sometime. * : complaint -> witness's reaction [ overlapping speech • What is the nature of the daughter's response to her father's complaint? • Who does the mother align with? • What stance does she take towards the complaint itself? • What effect does that have on the force of the complaint? • What function doss the father's closing remark have? but geirtk^sp^h a^ssuch as threats £to the hearers, negative face). FTAs can be classified on whether they threaten speaker or hearer, and their positive or negative face. An apology constitutes a threat to the speaker's own positive face, since it implies a past shortcoming of theirs. An FTA is said to be on record when a clear single intention is obvious behind it, while for off record the intention remains ambiguous. The most direct, bluntly expressed FTA is bald - 'Sit down' - compared with the manyj^reMve jirtionsiy_^kb_s^eatesjmtiga^the fHe...%eat;. 'Please pull up a chair', 'Won't you take a seat.' The politeness model has triggered a great deal of sociolinguistic research over more than 30 years in many cultures, languages and genres, often focusing on gender differences. It has also faced many refinements and challenges (e.g. Watts 2003). Probably the main issue has been Brown and Levinson's original claim to universality - their subtitle was 'Some Universal in Language Usage' (1987). In particular, face wants are not necessarily the drivers of politeness in all cultures. An ongoing site for research on the nuances and relationship of politeness and face is the use of honorific forms in 146 Situated language Situated language 147 Poiiteness interacts with power, as Watts's text argues (2003). Despite a charter which entitles patients in Kenyan hospitals to respect, Ojwang, Matu and Ogutu (2010) found patients are subjected to frequent face-attacks by nurses. This reflects : the power which nurses as professionals hold over, patients, overriding usual consid- : e rations of politeness. Ojwang et al. give examples of eight types of speech acts from their data (2010:505): Why do you keep disturbing us? Don't you know that you must pay for drugs? I don't want to see that woman here. The problem with you is that you don't understand. Why follow me in hospital as if I were your debtor? Take that woman out of here now. You are adults but you behave like children. if you try to be clever here, you will regret. • What are the speech acts involved here? In Brown and Levinson's terms, which are on- or off-record - with a clear single intention versus ambiguous? Which are baldly put, and which have some mitigating redressive wording? In what cases is it hard to say? While some patients tried to repair the damage to their dignity or launched a counterattack on the nurse's face, most responded with: a silence that reflected the power the nurse held over them. A teenage girl in labour was reprimanded and humiliated when she appealied for assistance (2010:506): a. Patient: Sister nisaidie nakufa na maumivu. 'Sister, help me I am dying of pain.' b: Nurse: Si ungomaliza shuie kwanza ndiyo ushikc mimha wcwe ■ msichana? 'Why didn't you complete school before conceiving you girl?' (Silence) c. Nurse: Sasa ndio utashika adabu 'It is now that you will learn manners.' (Silence) • Interpret how the Nurse's words in this extract threaten the patient's positive and negative face. Japanese, by which levels of respect are encoded in grammatical forms such as prefixes (see for example a Journal of Pragmatics theme issue, 43/15, 2011). 6.6 THE SOCIOLINGUISTICS OF INTERACTION Sociolinguistic research into speech acts involves attention to the fine detail of how participants interact in everyday situations. It forms part of a wider interest in how we make ourselves and our communicative intentiojis""understood in daily interaction. This was the pioneering work of John Gumperz, whose name is inextricably linked with Dell Hymes because of their co-editorship of the foundational Directions in Sociolinguistics collection (1972). But their work, approaches and contributions were very different. Hymes saw the big picture of the emerging field, taxonomizing and theorizing, and presenting little empirical language analysis of his own. Gumperz's focus was micro, as my Chapter 5 discussion of his research on code switching showed. He examined the minutiae of everyday conyersations, and how these functioned to enable individuals to communicate and understand one another. His 'Interactional Sociolinguistics' was hands-on rather than programmatic. Gumperz's work relates, often quite explicitly, to several of the approaches presented in this chapter. Goffman's 'interaction order' (section 6.3) offered a framework for studying everyday social dealings for which Gumperz looked to work out principles specific tcf language use. Gumperz (1982a: 156) also built on the thinking of speech act philosophers such as Austin and Searle (section 6.5), and particularly the approach of H.P. Grice. Grice had developed the 'Cooperative Principle' (1975), which spelt out four maxims for the conduct of conversations: that contributions need to be informative, true, relevant and clear. Gumperz's consideration of what was below the surface of conversations drew on Grice's other main construct - 'conversational implicature'. He took the proposals of the philosophers and sociologists and operationalized them for actual everyday conversation. Gumperz's leading concept was the contextualization cue, presented in his foundational book Discourse Strategies (1982a). Contextualization cues are constellations of surface features of message form ... by which speakers signal and listeners interpret what the activity is, how semantic content is to be understood, and how each sentence relates to what precedes or follows' (Gumperz 1982a: 131). They offer overt clues to wtoaj^otherwseody presupposed or implied below the surface of a conversation. The cues can be of almost any linguistic (and non-lmguisticj kind^- vocabulary, phonology, syntax, code switching, use of formulae, and (particularly) intonation patterns or prosody', Theyfunction to 'key' (as in Hymes's SPEAKING framework, Table 6.2) thejnter-jjretation which the speaker expects the hearer to reach. At the start of an interaction, a speaker and a listener will spend some time negotiating the basis of continuing conversation. In this process, the interpretive weight placed on contextualization cues is much greater than their surface linguistic importance. If the interactants coordinate successfully, they have achieved 'conversational sy nchrony'and moveforward in tune wjjji^achj)ther. But if interactants do not share a common set of contextualization cues, trouble follows. Miscommunication brings to the surface presumptions that are usually hidden. Gumperz focused on such misfires, investigating communication problems and seeking to interpret where, how and why such breakdowns had occurred. Much of his early empirical work looked at interactants who share the same language but come from very different background cultural situations, such as Indian speakers of English in conversation with British or American speakers. He analyses in depth (1982a: 174) one employment counselling interview between a (white) British counsellor and an Indian teacher which lasts over an hour in a state of continual misunderstanding and increasing conflict. Their purposes for the conversation clash in the manner of Hymes's Goals/Outcomes distinction (Table 6.2). 148 SITUATED LANGUAGE Here are two examples from Gumperz's early research on contextuaiization cues {1982a: 135): A A husband sitting in his living room isaddressinghlswife. Husband Do you know where today's paper is? Wife I'll get it for you. Husband That's O.K. Just tell me where it is. /'// get it. Wife No, I'll get it. B A mother is talking to her eleven year old son who is about to go out in the rain: Mother Where are your boots? Son In the closet. Mother I want you to put them on right now. 1 What sort of speech acts do you think these are? - questions seeking information, or indirect requests for action? 2 Notice the occurrence of strong word stresses in the last utterance of each example. What do these express? 3 When Gumperz played the two interactions to sets of listeners, those who heard the opening line of each example as a request also felt the tone adopted in the closing lines was justified. Why? 4 The husband in A is American, while the wife in A and mother in B are both British. Listeners who were British tended, to give different.interpretations; from Americans. Which variety do you think would favour the question reading, and which the request? {See Gumperz's finding at 1982a: 139.) What could this mean for cross-varietal/cultural communication? The difficulties that questions 1-3 pose for non-native English speakers, and 4 for non-Americans/British, illustrate how subtly and culturally determined is the knowledge required to interpret such everyday interactions, Contextuaiization cues are central to the communicative competence of speakers -recall Dorians characterization of the interpretive skills of semi-speakers in her Gaelic community (Exercise 4.7). Manvofthese cues are subile. *pedfk i.. sun-groups, and learned through long-term, face-to-face interaction with ingroup members. But the very absence of these skills contributes to limiting the amount and kind of contact that would produce their acquisition by the outsider. Gumperz's wider research interest was in the AC^iorjoliUca]Lr^ercu^r^pfjhese kinds of cross-cultural rmsnndeitfandjnff, though his legacy is in many ways the least sharply defined of the sociolinguistic founders. Interactional Sociolinguistics is;_the name SITUATED LANGUAGE 149 given to the body of work that has builton Gumr^rz^me_thods_(2001). It shares not so much a cohesive theory as a methodological conviction that the detailed workings of eyeryday conversations are This it has in common with the fine-grained methods of Conversation Analysis. It has fed into the approaches of socio-linguists whose work we have discussed earlier, such as Gal, Levinson, Heller and Auer (some of whom were Gumperz's students). Best known is JJeborah Tannen, collaborator with Gumperz on the development of contextuaiization cues, whose study of'conversational style' (19841 is a comprehensive application of the interactional approach. It analyses talk over a dinner (for American Thanksgiving) among six friends in California. Three came from New York, two from California, and one from England. The New Yorkers had what Tannen called a 'high-involvement style': fast and simultaneous speech, rapid questions, quick turn taking. This left the other three participants - who had a 'high-considerateness style'.- feeling.dominated_and In Hymes's terms (Table 6.2), there was a clash of norms. 6.7 GENDER AND CONVERSATION Such research on communicative cross-purposes led Tannen to publish books like You Just Don't Understand (1990), which popularized the role of gender difference in conversation to the broader American and international public, particularly with her contrast betweenjnjms 'reporUtalk^and.women's 'rajjpcrrt-f^LK This was part of an upsurge in sociolinguistic work on gender and language, one strand of which treated gendered distinctions in language as a form of cross-cultural difference. Extending Gumperz's approach into cross-gender communication, Maltz and Borker (1982: 212) treated Jknerican boys and girls as being socialized in different worlds: What we are suggesting is that women and men have different cultural rules for friendly conversation and that these rules come into conflict when women and men attempt to talk to each other as friends and equals in casual conversation. Early research examined differences, for example in speech acts such as compliments in English (e.g. Holmes 1988, in New Zealand). The general findings were that women give and receive more compliments than men. Rees-Miller (2011) replicated such studies in the United States decades later. In unstructured' settings, compliments by women to women were largely (74%) about appearance, while male-to-male compliments on appearance were very rare. Most compliments between men were about performance (64%), particularly in sport (see Exercise 6.11). A second approach takes a deficit line. The earliest gender language research implied that women's language is in some ways deficient rather than just different. Robin Lakoff's short book Language and Woman's Place (1975) focused on language forms which indicated subservience and, she believed, were typical ofwomen's English rather than men's, for example: 150 situated language | Exercise 8.t1 Gcfrvalm^n.sar*tge**8* - ■■ I Consider these three compliments and their reception (Reea-Miller 2011)- 1 M: Hey bro. good game tonight; you played really well ■ ••' . . M Hey thanks, man, it was a tough one. v.- -. 2 F: That color goes great with your eyes. F: Thank you for noticing. 3. M I like that shirt on you M: Thanks, I like your face .Describe and.interpret the. gender mix. of ■parLcipants;.^tH'©ito|Mes.-;-iai^^h&.T^"6p^'i t/on of these compliments. Note the linguistic details such as the use of solidary address terms in 1 (see Kicsling's study on 'dude; 2004) . » Exactly how and why is the response to 3 odd? What do you think it means socially? - for speaker, for hearer, and tor the relationship between them?, (see v: Rees-Miller 2011:2685 for the researcher's interpretation) . Why were compliments 1 and 2 accepted and 3 not? Pragmatic markers you know, sort of Tag questions isn't it? can't we? Intensifiers so much, just great 'Empty' adjectives sweet, adorable Precise colour terms beige, lavender While her book was the foundation of gender language studies, l.akoff was soon criticized on at least three grounds: aspects of her approach implied women's deficiency, the linguistic forms were not at all neatly distributed according to gender, and their functions were much more variegated than just indicating lack of confidence. Many of these are typical Gumperz-typecontextualization cues, the pragmatic lubricant which eases the course of conversation, while others are politeness moves which attend to an interlocutor's face needs. And it was not long before feminist sociolinguists questioned why women's language should be regarded as the deviation from men's language rather than vice versa. The third approach views gender language as involving dominance. Lakoff s explicit statements on the matter make it quite clear that she believed the root cause of the 'deficiencies' was in fact dominance: The ultimate effect of these discrepancies is that women are systematically denied access to power, on the grounds that they are not capable of holding it as demonstrated by their linguistic behavior along with other aspects of their behavior. (Lakoff 1973: 48) Situated Language 151 Gender differences - both in language and generally in society - are not random or neutral alternatives. Gender is itself a significant form of social stratification: there is no known society where women have more power than men (Giddens 2009). That power relation often surfaces in the patterns of conversation. In an article on 'the work women do' in interaction, Pamela Fishman (1978) found that women providejl jnost of the support in conversations with their partners - for example, they asked three quarters of the questions. On the other hand, men did three-quarters of the interrupting in West and Zimmerman's research (1983) on the small insults' embodied in conversational interruptions. It was not long before O'Barr and Atkins (1980) concluded, on the basis of comparing male and female witnesses in court, that it was_not a matter of women's language but of 'powerless .language'. Gender language research is now a large and varied field, and we return to different facets of it in the following chapters, including nuancing the definition of gender and how it is enacted. We will also see that the deficit-difference-dominance_paradigms apply as well - not surprisingly- to other dimensions of social differentiation, particularly ethnicity (Chapter 7). 6.8 THE CASE of SLANG IN RIO The South Zone is the wealthiest area of the Brazilian mega-city of Rio de Taniero. It has spectacular hills and beaches and the most expensive real estate in South America. It is also the location of a number of favelas - shanty towns. These side-by-side contrasts of rich and poor have, as so often, an ethnic and gender reflex. Poor black young men in the area are some of Brazil's 'most disenfranchised citizens', according to an ethnographic study by the American sociolinguist Jennifer Roth-Gordon (2007: 323). Research into the language of disadvantaged black male urban youth has a long tradition in sociolinguistics, stretching back to pioneer studies in the 1960s of New York City teenage gangs (Labov 1972a - see Chapter 7, this volume). The methods needed to research language in such situations are necessarily ethnographic. Researchers have to win access to and earn the trust of the people they wish to work with. Roth-Gordon's approach was typical of a linguistic-ethnographic study in its use of multiple data sources, long-term involvement, and collaboration with the participants. She interacted with her informants over a period of IS months. Data collection began with written surveys and interviews at a high school. This led on to recording casual conversations with a peer group, the central data pool of the project. One group member acted as the researchers mentor, guide and interpreter to the group and its language. He reviewed the recordings, identifying and translating target expressions from the Portuguese, which made participants' intuitions and judgements central to defining the data of the study. Finally, the researcher took, selected excerpts from the recordings and played them back to different groups -the participants themselves, their parents, other adults, and middle-class youth. 152 Situated language Pragmatic markers The Rio de Janiero study focused on the role of giria or 'slang' in the margfnalization of these young men of thefavelas, and specifically on the role of four 'pragmatic markers' in glria. Sometimes also called discourse markers or pragmatic particles, these are the small bits of language v.hkh provide much of the: discoursal and interactional glue that hold our conversations together (Scbiffrin 1987). In English, expressions like 'all right, 'actually', T mean^ 'and so on' pepper our conversation, constantly signposting the structure and significance of what we say and the way in which we want to relate to our interlocutors. They are, in Gumperz's terms, some of the leading contextualization cues. Ethnographic, qualitative methods are suited to researching theseksnds of hnguistic forms, which require a situation where speakers are at ease and analysts can retrieve enough context to interpret the usages. But there is a sociolinguistic issue here as well as a pragmatic one:_the markers are ojten denigrated, especially in the speech of non-mainstream groups. They change frequently and spread quickly - and are commonly despised by, you know, older people as sort o/indicative of the inadequate language skills of, like, younger people? There can be little doubt, however, that these critics will themselves be using pragmatic markers in their own conversation. Everyone does. The pragmatics of giria The outsider groups routinely described the giria users' speech as incomprehensible, and pointed especially to their high use of several pragmatic markers that are both textual and interactional. Roth-Gordon studied four expressions: sound words, obscenities, address terms and addressee-oriented tags. Example 6.2 shows multiple occurrences of the first three of these. We will look at just two of them. Sound words are in fact non-words, phonologieally graphic expressions that have also been studied in other languages such as Greek, where they are frequent in narratives (Tannen 1986). Most sound words imitate noises of contact or conflict like bam! or pd! Often preceded or followed by a pause in giria, they behave almost like spoken commas, punctuating narratives and foregrounding new and important information (see Example 6.2) - classic contextualization cues. These loud, fast-spoken stems are widely used in the male peer group, although speakers of standard Portuguese would not even regard them as words. The parents and middle-class Brazilians, whom Roth-Gordon surveyed for comment, know that these are part of stigmatized giria speech and say that they hamper the coherence of the discourse. Since the function of pragmatic markers is to do just the opposite - to organize and structure discourse - the perception of incoherence appears to be a reaction to the speakers rather than to their speech. Addressee-oriented tags are .among_the most conunon interactional devices in ianguage - classically in English y know (Example 6.1). The most distinctive addressee-oriented tag in giria is ta ligado 'are you connected/paying attention?', which has been popularized by Brazilian rappers to reference black unity and political consciousness. It I therefore functions both linguistically and sociopolitically in the same way as two prag- I matic markers in African American Vernacular English - arte 'all right' and na mean 'you '.know what I mean' (Reyes 2005). Situated Language 153 181 311 ■Hi Pragmatic markers in English: Pragmatic markers: have been -heavily, researched in sociolinguistics for decades, in English1 and: other languages - irvwomen's and men's speech (Holmes 1995), academic lěctures-ÍChia-Wp^ place discourse (Holmes 2006), and (Spanish) classroom interaction (De Fina 1997). . One study of my own looked at markers in the peer conversations of young New Zealanders (Bell and Johnson 1997), especially addressee-oriented markers which serve to check the listener's comprehension of and empathy with what the speaker is saying. Y'know turned out to be especially common, occurring up to 200 times in-a single interview.. (Discourse, markers often;:ba«e: reduced phonetic realizations, so linguists write them accordingly -y'know 'you know; arte 'all right") One speaker produces frequent clusterings of y'know as he talks, sometimes up to 10 tokens in a run, and even the occasional pair of two y'knows directly after each other. By contrast,.tag questions such as aren'f they? are very rare in real talk despite their being taught to learners as essential colloquial expressions. Hustling in Brazilian giria The giria speaker 'Karate* describes watching street vendors hustle bystanders into playing a gambling game. He uses three of the four pragmatic markers studied--soundwords (three times), obscenities (three times) and address forms, (twice). Their frequency 'dramatically alters the rhythm of his speech' (Roth-Gordon 2007: 327) and constitutes a distinctive giria style. Karate: Caralho, tirou dez mingau do bolso. Pum\ Botou na parada. Sum! Caraca, ganhou os dez mingau. Caralho, ganhou quarenta merrsis. ihi mane; e mote! /..../ Os outros cara tao apostando iambem junto com ela. Ba> Ganhando, perdendo, mas o mesmo, cara. Caralho (shit'), he took ten bucks out of his pocket. Puml Put it in the game. Bumf Caraca ('gosh'), he won the ten bucks. Caralho ('shit'), he won forty big ones. Oh, mane ('man'), it's easy!-. [,...] The. other guys are also betting j along with her. Sal Winning,: los^ cara ('man'). 154 Situated Language 'One normal word in ten However, the salient thing for listeners is not so much that giria speakers use their own distinctive addressee-oriented tags as the frequency with which they use the standard tags, particularly sabe '(you) know' Some of the outsiders that Roth-Gordon surveyed describe slang speakers as speaking one normal word for every ten words' (2007: 332). The markers are indeed frequent in Example 6.2, but in fact they constitute there only one word in five rather than nine words out often. Exaggerating the prevalence^of a handful of features is a common stereotype of youth speech in different languages and locations, from Brazilian giria to Californian female youth language to British 'chavspeak' (Mendoza-Denton 2008) - see Exercise 6.12. Yet study after study has found that discourse markers are widespread in everyday speech from all kinds of speakers. As Roth-Gordon concludes, these giria speakers 'present a coherent linguistic style that makes extensive use of pragmatic expressions to organize textual information, manage interpersonal relationships between speaker and listeners, and convey speaker stance towards content and audience' (2007:339). Standard Portuguese sJí'^}^I^I}}É:^^Ilx]^&^at^SÍL'aS'3 socul noUiniguistic grounds^- on the margin-.ilizetl position of it; speaker-,, and on the; wayin which^na challengesj:onyention. 'Discourse markers in general suffer from an image problem.' observes Norma Mendoza-Denton (2008:285), whose study investigated Latina gang use of various r markers, t'■ vh^PB'M^M^: 1 Are Ihere any such frequent but stigmatized markers in your language and dialect? If so. what are your impressions of how they are used and by whom, what are their discourse and social meanings, and what are general attitudes to them? ;^tjSeT irjter^ for example articles and books by Janet Holmes, and many papers in the Jmrrta%of0ag$aifc^ 3 Locate YouTube dtps of media representations of this kind of language - such as Little Britain and The Catherine Tate Show tor British chavspeak (see also Bennett 2012). and Saturday Night Live performances ot young California i;% female talk. Assess the accuracy of those performances, and their part in the 4 Consider the role that these pragmatic markers play in stereotyping and marginalizing their users. Why do they attract such negative evaluations? What are the means by which these evaluations get circulated in a society? Should See Chapter 10 for further discussion of language ideologies and standards and their repercussions. situated Language 155 6.9 RESEARCH ACTIVITY ETHNOGRAPHING THE CLASS Apply Hymes's framework as displayed in Table 6.2 to the class you are in. I have expanded Hymes's terms for participants by introducing the finer distinctions shown in Tables 6.3 and 6.4 Then use the activity to assess the framework. How much does it further your understanding of language use in the classroom? What are its limitations? Can you suggest improvements? Setting Time and place where the class is held. Physical setup, e.g. of seating. Does place' signify more than just location? Scene What is the cultural definition of this occasion? Is it a formal lecture? Or some other format - seminar, tutorial? What are the rules? Animator, Editor, Author, Principal Who does the talking? Who voices the messages? Who is the author? Are there any 'editors'? To what extent is the lecturer conveying content generated by others? Is there a tradition' being handed on? Is there much quotation, direct or paraphrased? Who is the authority standing behind the messages? Who is responsible for the content of the class? Do class members generate content? Does the institution (department, university management) have a role? How and when might that become an issue? Addressee, Auditor, Overhearer, Eavesdropper Who hears the message? The whole class? And only the class? Are all class members being addressed, or are some of them auditors (ratified but not addressed)? Are any unratified hearers present - overhearers? Are all the hearers being addressed all of the time? Are there any eavesdroppers?! How do both producer and audience roles change when lecturer or student asks a question? Or when the class breaks out into small groups? Outcomes What is the overall purpose of the class? Goals Do all class members share this purpose? Does the lecturer? How can differences between outcomes and goals show up? 156 situated language Message form How are things said - by lecturer, by class members? What is the mix of formal and informal? Message content What is the overt theme, topic? How is this approached and presented? Is it adhered to? Key What is the manner or tone of what is said? Is it always straight and serious, or are there other keyjngs such as humour, irony? Are there non-verbai aspects? Channels What are the channels of communication - spoken, written, audio, visual? Are there handouts, graphs, powerpoint, video playback, internet materials, etc.? Forms of speech What languages are used, and what varieties? Who by? What codes from participants' repertoires are not used? Norms of interaction What are the rules for turn taking in the ciass? Who is licensed to speak? When and how? Do all interactions take place across the whole class, or are there 'breakouts' such as pair work? Norms of interpretation Does everyone interpret the norms in the same way? Are there any clashes between lecturer and students, or between different class members? Why? Genre Should this really be classified as a lecture? Or is it some other related genre? What are the criteria? 6.10 summary This chapter has interwoven several broadly anthropological approaches to language use and users, introducing a large battery of concepts and terms. • 'Situation' is the social occasion where language occurs. It involves interplay between language and its context, with language in part shaping the situation of its own use. * Fishmans concept of domains reflects that there are typical combinations of participants, topics and settings in, for example, educational or religious contexts, and these tend to trigger particular code choices. situated language 157 The founding approach to situated languageytasDeil Hymess 'ethnography of communication' focused in the SPEAKING taxonomy which identifies the components of speech situations. While Hymes recognized only two speaker and two hearer roles, Goffman then Bell advocated for finer divisions. Within the construct of 'speaker', roles from the Principal (authority behind a text) through Author and Editors to Animator (the voicebox) need to be differentiated. Audience roles are also ranked in centrality from the targeted Addressees outwards / through the unaddressed Auditors and unratified Overhearers to any unknown | Eavesdroppers. These finer roles can be seen at work in media and internet produc-l tion and reception, and are best conceived as embedded within or layered upon each) other./ \ Participant roles have major implications for language choices, and also for the operation of speech acts, a focus in the neighbouring field of pragmatics. Austin identified performative expressions like 'I promise' which do the action they name. Speech acts have different 'illocutionary forces' such as asking or threatening. Early sociolinguistic research in speech acts focused on counting and comparing gender differences in acts such as apologies. Later research attends to the multiple functions and nuances that such speech acts may carry in the context of discourse. In the 'politeness' framework of Brown and Levinson, speech acts relate to participants' 'face' needs, especially when confronted with face-threatening acts. Gumperz's work in Interactional Sociolinguistics focuses on how we make our communicative intentions understood through the minutiae of everyday interaction. His concept of the contextualization cue identifies surface features by which speakers signal to listeners what is otherwise only implied in a conversation. If interactants do not share a common set of contextualization cues, trouble follows. Gumperz focused on such communication breakdowns especially in cross-cultural encounters. Gendered language differences show in conversation. The 'difference' approach regards men and women as coming from communicatively distinct subcultures, shown for example in their complimenting behaviours. The deficit perspective evaluates women as conversationally unconfident and subservient, while the dominance approach sees the differences as the result of mens power in society, exemplified in more frequent interrupting behaviour. The chapters case study is Roth-Gordon's ethnography (2007) of disadvantaged black youth in Rio de Janiero, involving long-term close observation of their way of life, recording their interactions, and eliciting their interpretations. Giria or 'slang' is a key part of their identity and marginalization, and is especially expressed in the high use of pragmatic markers, which is denigrated by standard Portuguese speakers. 6.11 Further reading For overviews of ethnographic/anthropological approaches, see the complementarily titled texts by Duranti (Linguistic Anthropology, 1997) and Foley [Anthropological 158 Situated Language situated language 159 Linguistics, 1997). Duranti has also edited two readers {2004, 2009). The five-volume Schieffelin and Garrett compendium (2011) assembles an invaluable set of classic and recent articles. For considerations of situation and context, see Duranti and Goodwin's collection (1992) and van Dijk's recent monograph from a text discourse perspective (2009). SaviUe-Troike's The Ethnography of Communication (2003) gives an extensive exposition of Hymes's SPEAKING framework plus several worked examples. The book-length studies by Eckert (2000) and her one-time student Mendoza-Denton (2008) offer insightful expositions and reflections on the doing of ethnography. The Journal of Linguistic Anthropology gives the flavour of current research, as do articles in Language in Society and Journal of Sociolinguistics. Like his contemporary, Fishman, Hymes was central in shaping the field through his work as an essayist (e.g. 1972), journal and collection editor. In 1972 he founded the original socioliiiguistic journal, Language in Society, and edited it for over 20 years. He also (co-)edited still-relevant early collections (Hymes 1964; Gumperz and Hymes 1972). Hymes's work is most accessible through his 1974 collected works Foundations in Sociolinguistics. He is not a tidy thinker or writer, but his thought is stimulating and foundational. See biographical essay in Wodak, Johnstone and Kerswill (2011). Studies of participation and audience frameworks trace their roots back to Hymes (1972) and Goffman (1981). My own take is in Bell (1984,1991a), and sociolinguists continue to revisit and refine these basic constructs (see articles in the Journal of Pragmatics). On speech acts, Austins founding book How to do Things with Words (1962) is an essential and elegant read. Collected readers at the more abstract/philosophical end of pragmatics include Kasher (1998) and Allan and Jaszczolt (2012). For empirical socio-linguistic studies, the Journal of Pragmatics - at nearly 4,000 pages per year in 15 issues -is a compendium of up-to-date research, especially on speech acts, politeness and discourse markers. The International Pragmatics Association runs conferences and publishes (see website). Brown and Levinson (1987 - first published in 1978 as the major part of a collected volume) is the foundation document of the politeness framework. There is a Journal of Politeness Research devoted to the topic. Watts's reframing text (2003) challenges many of its assumptions and presents politeness as involving a discursive struggle for power. Janet Holmes's research, especially on workplace interaction, forms a major body of work on speech acts, discourse markers and politeness (e.g. 1995, 2006; Holmes and Stubbe 2003; Holmes, Marra and Vine 2011). Gumperz's 1982(a) monograph is the basic text of Interactional Sociolinguistics, twinned with his edited collection (1982b) of chapters by students and colleagues. The biographical and contextual essay by Gordon in Wodak, Johnstone and Kerswill (2011) provides a good overview of Gumperz's work. Eerdmans, Prevignano and Thibaulfs Discussions with John J. Gumperz (2002) offers a fascinating roundup of his own approach and its successors. Volume 3 of Coupland and Jaworski's collection (2009a) covers Interactional Sociolinguistics (although oddly Gumperz's own work appears in other volumes). Tanneris books, both academic and popular (1984,1990,1994) probably constitute the most sizeable corpus of overtly interactional socioliiiguistic work. Gender/ language reading will be summarized in Chapter 7, except to note Pavlidou's very useful overview of gender and interaction' (2011). References Allan, Keith and Kasia M. Jaszczolt (eds), 2012. The Cambridge Handbook of Pragmatics. Cambridge: Cambridge University Press. Austin, J.L., 1962. How to Do Things with Words. Cambridge, MA: Harvard University Press. Bakhtui, M.M., 1981 [1935]. 'Discourse in the novel'. In M.M. Bakhtin, The Dialogic Imagination. Austin, TX: University of Texas Press. 259-422. Bell, Allan, 1984. 'Language style as audience design'. Language in Society 13: 145-204. Bell, Allan, 1991a. Audience accommodation in the mass media'. In Howard Giles, Nikolas Coupland and Justine Coupland (eds), Contexts of Accommodation: Developments in Applied Soeiolinguistics. Cambridge: Cambridge University Press, 69-102. Bell, Allan, 1991b. The Language of News Media. Oxford, UK; Basil Blackwdl. Bell, Allan and Gary Johnson, 1997. 'Towards a sociolinguistics of style". University of Pennsylvania Working Papers in Linguistics 4: 1-21. Bell, Allan and Philippa Smith, 2012. 'News discourse. In Carol A. Chapelle (ed.), The Encyclopedia of Applied Linguistics (1st edn). Oxford and Maiden, MA: Wjley-Blackwell (online at http:// onlinelibrary.wiley.com/book/10.1002/9781405198431; retrieved 1 February 2013). Bennett, Joe, 20! 2. '"And what comes out may be a kind of screeching": the stylisation of chavspeak in contemporary Britain! Journal of Sociolinguistics 16: 5-27, Brown, Penelope and Stephen C. Levinson, 1987. Politeness: Some Universalis in Language Usage. Cambridge: Cambridge University Press. Brown, Roger and Marguerite Ford, 1964. 'Address in American English'. In Dell Hymes (ed.), Language in Culture and Society. New York: Harper & Row. 234-44, Brown, Roger and Albert Oilman, 1960. 'The pronouns of power and solidarity'. In Thomas A. Sebeok (ed.), Style in Language. Cambridge, MA: MIT Press. 253-76. Coupland, Nikolas and Adam laworski (eds), 2009a. Sociolinguistics (Critical Concepts in Linguistics), 6 vols. London: Routledge. Crystal, David 2006. Language and the Internet (2nd edn). Cambridge: Cambridge University Press. De Fina, Anna, 1997. 'An analysis of Spanish bien as a marker of classroom management in teacher-student interactions', journal of Pragmatics 28: 337-54, Duranti, Alessandro, 1997. Linguistic Anthropology. Cambridge: Cambridge University Press. Duranti, Alessandro (ed.), 2004. A Companion to Linguistic Anthropology. Maiden, MA: Blackweli. Duranti, Alessandro (ed.), 2009. Linguistic Anthropology: A Reader (2nd edn). Maiden, MA: Wiley-BIackwell. Duranti, Alessandro Duranti and Charles Goodwin (eds), 1992. Rethinking Context: Language as an Interactive Phenomenon. Cambridge: Cambridge University Press. Eckert, Penelope, 2008. 'Variation and the indexical field', Journal of Socialinguistks 12: 453-76. Eerdmans, Susan L„ Carlo L. Prevignano and Paul I. Thibault (eds), 2002. Language and Interaction: Discussions with John J. Gumperz. Amsterdam: lohn Benjamins. Ervin-Tripp, Susan M., 1972. 'On sociolinguistic rules: alternation and co-occurrence'. In lohn J. Gumperz and Dell Hymes (eds), Directions in Soeiolinguistics. New York: Holt, Rinehart & Winston. 213-50. Fishman, Joshua A., 1965. 'Who speaks what language to whom and when'. La Linguistique 2: 67-88. Fishman, Joshua A. (ed.), 1971. Advances in the Sociology of Language, vol. 1: Basic Concepts, Theorys and Problems - Alternative Approaches. The Hague: Mouton. Fishman, Joshua A. (ed.), 1972. Advances in the Sociology of Language, vol. 2: Selected Studies and Applications. The Hague: Mouton. 160 SITUATED LANGUAGE SITUATED LANGUAGE 161 Fishman, Pamela M., 1978. 'Interaction: the work women do. Social Problems 25: 397-4-06. Foley, William, 1997. Anthropological Linguistics: An Introduction. Maiden, MA; Blackwell. Gal, Susan, 1979. language Shift: Social Determinants of Linguistic Change in Bilingual Austria. New York: Academic Press. Giddens, Anthony, 2009. Sociology (6th edn). Cambridge, UK: Polity Press. Goffman, Erving, 1964. 'The neglected situation'. American Anthropologist 66/6, part 2: 133-6. Goffman, Erving, 1981. forms of Talk. Philadelphia, PA: University of Pennsylvania Press. Goodwin, Chaťles and Alessandro Duranti, 1992. 'Rethinking context: an introduction'. In Alessandro Duranti and Charles Goodwin (eds), Rethinking Context: Language as an Interactive Phenomenon. Cambridge: Cambridge University Press. 1-42. Grice, H.P.. 1975. 'Logic and conversation. In Peter Cole and Jerry L. Morgan (eds), Speech Acts. New York: Academic Press. 41-58. Gumperz, John J., 1982a. Discourse Strategies. Cambridge: Cambridge University Press. Gumperz, John J. (ed.), 1982b. Language and Social Identity. Cambridge: Cambridge LJniversity Press. Gumperz, John J., 2001. 'Interactional sociolinguistics: a persona! perspective'. In Deborah Schiffrin, Deborah Tannen and Heidi E. Hamilton (eds), The Handbook of Discourse Analysis. Maiden, MA: Blackwell- 215-28. Gumperz, [ohn ). and Dell Hymes (eds), 1972, Directions in Sociolinguistics. New York: Holt, Rinehart & Winston. Holmes, Janet, 1988. 'Paying compliments: a sex-preferential politeness strategy'. Journal of Pragmatics 12: 445-65. Holmes, Janet, 1995. Women, Men ami Politeness. London: Longman. Holmes, fanet, 2006. Gendered Talk at Work: Constructing Gender Identity through Workplace Discourse. Maiden, MA: Blackwell Publishing. Holmes, Janet, Meredilh Marra and Bernadette Vine, 2011. Leadership, Discourse and Ethnicity. New York: Oxford University Press. Holmes, Janet and Maria Stnbbe, 2003. Power and Politeness in the Workplace: A Socioiinguistic Analysis of Talk at Work. London: Longman. Hymes, Dell H.. 1962. 'The ethnography of speaking'. In Thomas Gladwin and William C. Sturtevant (eds), Anthropology and Human Behavior. Washington, DC; Anthropological Society ofWashington. 13-53. Hymes, Dell (ed.), 1964. Language in Culture and Society: A Reader in Linguistics and Anthropology. New York: Harper & Row. Hymes, Dell, 1972. 'Models of the interaction oflanguage and social life'. In John J. Gumperz and Dell Hymes (eds), Directions in Sociolinguistics. New York: Holt, Rinehart & Winston. 35-71. Hymes, Dell, 1974. Foundations in Sociolinguistics: An Ethnographic Approach. Philadelphia, PA: University of Pennsylvania Press. Kasher, Asa (ed.), 1993. Pragmatics: Critical Concepts (6 vols). London: Routledge. Keenan, Elinor, 1989. 'Norm-makers, norm-breakers: uses of speech by men and women in a Malagasy community'. In Richard Bauman and Joel Sherzer (eds), Explorations in the Ethnography of Speaking (2nd edn). London: Cambridge University Press. 125-43. Kelly-Holmes, Helen, 2010. 'Rethinking the macro-micro relationship: same insights from the marketing domain! International Journal of the Sociology of Language 202: 25-39. Riesling, Scott E, 2004. 'Dude. American Speech 79: 281-305. Labov, William, 1972. Language in the Inner City: Studies in the Black English Vernacular. Philadelphia, PA: University of Pennsylvania Press. Laforest, Marty, 2009. 'Complaining in front of a witness: aspects of blaming others for their behaviour in multi-party family interactions'. Journal of Pragmatics 41: 2452-64. Lakoff, Robin, 1973. 'Language and woman's place'. Language in Society 1: 45-80. Lakoff, Robin, 1975. Language and Woman's Place. New^Yorlc Harper & Row. Lempert, Michael, 2009. 'On "flip-flopping": branded stance-taking in U.S. electoral polities'. Journal of Sociolinguistics 13: 223-48. Lin, Chia-Yen 2010. ' .. that's actually sort of you know trying to get consultants in .. .': functions and multifunctionality of modifiers in academic lectures'. Journal of Pragmatics 42: 1173-83. Maltz, Daniel N. and Ruth A. Borker, 1982. 'A cultural approach to male-female miscommunica- tion'. In John J. Gumperz (ed.), Language and Social Identity. Cambridge: Cambridge University Press. 196-216. McNeill, Laurie, 2009. 'Diary 2.0? A genre moves from page to screen' In Charley Rowe and Eva L. Wyss (eds), Language and New Media: Linguistic, Cultural and Technological Evolutions. Cresskill, Nf: Hampton Press. 313-25. Mendoza-Denton, Norma, 2008. Homegirls: Language and Cultural Practice among Latina Youth Gangs. Maiden, MA: Blackwell. O'Barr, William M. and Bowman K. Atkins, 1980. "'Women's language" or "powerless language"?' In Sally McConnell-Ginet, Ruth Borker and Nelly Furman (eds), Women and Language in Literature and Society. New York: Praeger. 93-110. OJwang, Benson Oduor, Peter Maina Matu and Emily Atieno Ogutu, 2010. 'Face attack and patients' response strategies in a Kenyan hospital'. Journal of Sociolinguistics 14: 501-23. Pavlidou, Theodossia-Soula, 2011. 'Gender and interaction. In Ruth Wodak, Barbara Johnstone and Paul Kerswill (eds), The Sage Handbook of Sociolinguistics. London: Sage, 412-27. Rees-Miller, )anie, 2011. 'Compliments revisited: contemporary compliments and gender'. Journal of Pragmatics 43: 2673-88. Reyes, Angela, 2005. 'Appropriation of African American slang by Asian American youth'. Journal of Sociolinguistics 9: 509-32. Rickford, John R. and Faye McNair-Knox, 1994. 'Addressee- and topic-influenced style shift: a quantitative socioiinguistic study. In Douglas Biber and Edward Einegan (eds), Socioiinguistic Perspectives on Register. New York: Oxford University Press. 235-76. Roth-Gordon, Jennifer, 2007. "Youth, slang, and pragmatic expressions: examples from Brazilian Portuguese'. Journal of Sociolinguistics 11: 322-45. Rowe, Charley and Eva I. Wyss (eds), 2009. Language and New Media: Linguistic, Cultural and Technological Evolutions. Cresskill, NJ: Hampton Press. SaviDe-Troike, Muriel, 2003. The Ethnography of Communication: An Introduction (3rd edn), Maiden, MA: Blackwell. Schieffelin, Bambi B. and Paul B. Garrett (eds), 2011. AnthropologicalLinguistics (Critical Concepts in Language Studies), 5 vols. London: Routledge. Schiffrin, Deborah, 1987. Discourse Markers. Cambridge: Cambridge University Press. Scollon, Ron and Suzanne Scollon, 1981. Narrative, Literacy and Face in Interethnic Communication. Norwood, NJ: Ablex. Scollon, Ron and Suzanne Wong Scollon, 2001. Intercultural Communication: A Discourse Approach (2nd edn). Maiden, MA: Blackwell Publishers. Searle, John R„ 1969. Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. Searle, John R., 1976. 'A classification of illocutionary acts'. Language in Society 5:1-23, Tannen, Deborah, 1984. Conversational Style: Analyzing Talk among Friends. Norwood, NJ: Ablex. Tannen, Deborah, 1986. 'Introducing constructed dialogue in Greek and American conversational and literary narrative'. In Florián Coulmas (ed.), Direct and Indirect Speech. Berlin: Mouton de Gruyter. 311-32. Tannen, Deborah, 1990. You Jusi Don't Understand: Women and Men in Conversation. New York: William Morrow. 162 Situated language Tannen, Deborah, 1994. Talking from 9 to 5: Women and Men at Work - Language, Sex and Power. New York: William Morrow. Thurlow, Crispin and Kristine Mroczek (eds), 2011. Digital Discourse: Language in the New Media. New York: Oxford University Press, van Dijk, Teun A., 2009. Society and Discourse: How Social Contexts Influence Text and Talk. Cambridge: Cambridge University Press. Vigouroux, Cecile B., 2010. 'Double-mouthed discourse: interpreting, framing and participant roles'. Journal of Sociolinguistics 14: 341-69. Watts, Richard)., 2003. Politeness. Cambridge: Cambridge University Press-West, Candace, and Don H. Zimmerman, 1983. 'Small insults: a study of interruptions in cross-sex conversations between unacquainted persons'. In Barrie Thorne, Cheris Kramarae and Nancy Henley (eds), Language, Gender and Society. Rowley, MA: Newbury House. 102-17. Wodak, Ruth, Barbara Johnstone and Paul Kerswill (eds), 2011. The Sage Handbook of Sociolinguistics. London: Sage. 7 variation in language This chapter introduces the field ofvariation research that has grown from the pioneering work of William Labov in the 1960s to become the most cohesive and dominant strand of sociolinguistics. The study of sociolinguistic variation had its roots in the tradition of dialectology practised in Europe and North America. This involved surveys of rural communities, concentrating on disappearing dialectal words and pronunciations, and mapping their distribution. Until the mid twentieth century, there had been no such work in urban areas but that changed with Labov's innovative research in New York City. The 'variationist' strand of sociolinguistics studies the nexus of both variation and change. This chapter concentrates on sociolinguistic variation, as far as possible reserving discussion of linguistic change for the next chapter. Variation is wider than change: a feature may vary withoulxhanging^but it does not change without varying. Here we focus on three of the five main social dimensions on which language varies between speakers: clMs^^brricity,.andgendej. Chapter 8 will cover another, age, together with change, and Chapter 9 deals with jregion, including geographical variation. Many of the examples here are from English, reflecting the preponderance of the research. Variation analysis deals in linguistic detail, and so technical terminology will be required in this and subsequent chapters to specify the phonetics ofvariation with the necessary precision. 7.1 FOUNDATIONS: NEW YORK CITY Doing sociolinguistic interviews Labov began his doctoral research in the LowerEast Side of ManhattanJn J962^aiming to investigate the English of New York City, map its social distribution and identify how it was changing. The Lower East Side was an economically and ethnically mixed area that The Guidebook to Sociolinguistics, First Edition. Allan Bell. © 2014 Allan Bell. Published 2014 by Blacfcwdl Publishing Ltd