1.2.1 Institutional discourse

Agar defined institutional discourse as “discourse where one person who represents an institution encounters another person seeking its services” (Reference Agar1985: 147). Drew and Heritage (Reference Drew, Drew and Heritage1992b) defined institutional talk as interaction used in diverse institutional contexts. The institutional interactions examined in Drew and Heritage’s volume are “task-related and they involve at least one participant who represents a formal organization of some kind” (p. 3, my emphasis). According to these authors, institutional interaction is characterized by the following elements: it is goal-oriented, that is, oriented toward a goal for at least one participant; it is constrained by what at least one of the participants considers an allowable contribution; it is asymmetric; and it may be related to inferential frameworks that are specific to the institution (pp. 22, 48–49). Further, Bardovi-Harlig and Hartford noted that institutional discourse is both comparable and consequential: comparable due to its goal-oriented nature, and consequential because real-world interactions have socio-affective consequences in that speakers take turns and negotiate their face (Reference Bardovi-Harlig2005: 12–13). And Hakulinen stated that institutional conversations “have as a rule an overall structure and are organized into phases: they have a recognizable beginning and an end, and the conversationalists – or at least one of them – know what to expect from the talk and what the topics to be dealt with are” (Reference Hakulinen, D'Hondt, Ostman and Verschueren2009: 58, emphasis in original).

The definition and characteristics presented here refer to institutional interactions in formal contexts where at least one of the participants is a representative of the institution, such as a judge, a doctor, or an advisor at a university. In these formal interactions, the organization of discourse is constrained by an institutionalized setting (e.g. court house) and is organized into fixed adjacency sequences, with an absence of overlap and interruption, and a fixed turn-taking structure.

However, because of the proliferation of studies in a wide variety of institutional settings, institutional discourse is often analyzed as a continuum of formal and non-formal settings. Following Heritage and Greatbatch (Reference Heritage, Greatbatch, Boden and Zimmerman1991), Hutchby and Wooffitt (Reference Hutchby and Wooffitt2008: 140–156) described two types of institutions, namely, formal and non-formal types. Formal institutional interactions, such as those carried out in courts of law, job interviews, or classrooms, for example, feature a rigorous question–answer format with a fixed turn-taking system. In contrast, non-formal types display a more flexible structure but are still task/goal oriented. In these non-formal settings, “much less uniformity in the patterning of conduct is evident,” and some part of the interaction can be characterized as “quasi-conversational” (Hutchby and Wooffitt Reference Hutchby and Wooffitt2008: 151). In view of their combination of institutional and “quasi-conversational” discourse features, non-formal institutional interactions can include business meetings, counseling sessions, or “service encounters in places such as shops” (Hutchby and Wooffitt Reference Hutchby and Wooffitt2008: 140; Lamoureux Reference Lamoureux1988/89).

It is important to note that in non-formal settings the issue of symmetry or asymmetry may be difficult to define. Institutional settings may be considered asymmetrical with regard to the positional roles and statuses of the participants, as in the case of service encounters between a clerk and a customer (I discuss the issue of symmetry and asymmetry in Chapter 2 [Section 2.3]). Given their goal-oriented nature and the mutual understanding of the participants with respect to their roles and statuses, I agree with Hutchby and Wooffitt (Reference Hutchby and Wooffitt2008) that service encounters, like the ones examined in this book which took place in small shops, delicatessens, open-air markets, and an information center, can be considered types of non-formal institutional settings. Service encounters represent a particular genre or discourse type with specific discourse strategies that distinguish them from formal institutional settings (in Chapter 2 I explain the organizational and sequential structure in different types of service encounters).

Since I examine service encounters in face-to-face interactions in naturalistic settings, I restrict my discussion below to perspectives on spoken discourse in formal and non-formal settings.

1.3.1 Speech Act Theory: a performative-intentional perspective

Although Speech Act Theory (Austin Reference Austin1962; Searle Reference Searle1969) was not designed to examine stretches of talk in social interaction, it provided the foundation for the analysis of social action. Austin proposed a three-way taxonomy of speech acts: (i) a locutionary act refers to the act of saying something meaningful, that is, the act of uttering a fragment or a sentence in the literal sense (referring and predicating); (ii) an illocutionary act is performed by saying something that has a conventional force such as informing, ordering, warning, complaining, requesting, or refusing; and (iii) a perlocutionary act refers to what we achieve “by saying something, such as convincing, persuading, deterring, and even, say, surprising or misleading” (Austin Reference Austin1962: 109, emphasis in original). Austin’s main interest was in utterances used to perform actions with words (e.g. “I pronounce you husband and wife”). For these actions to be accomplished, they must be executed under the appropriate conditions: (i) a conventional procedure and effect; (ii) the appropriate circumstances; (iii) the correct and complete execution of the procedure by all persons; and (iv) certain thoughts and feelings about the realization of the act on the part of persons involved (pp. 14–15). The notion of performative action is fundamental to the analysis of formal and non-formal institutional interactions because it considers both speaker and hearer co-constructing joint actions in specific sociocultural contexts.

Inspired by Austin’s original classification of illocutionary acts, Searle (Reference Searle1976) proposed a five-way taxonomy of speech acts: representatives, directives, commissives, expressives, and declarations. Representatives constitute assertions carrying true or false values (e.g. statements); in directive speech acts there is an effort on the part of the speaker to get the hearer to perform an action (e.g. requests, advice); commissive acts create an obligation on the part of the speaker, that is, they commit the speaker to performing an act (e.g. offers, promises, or refusals); expressive speech acts express an attitude or inner state of the speaker that says nothing about the world (e.g. apologies, congratulations, compliments); and declarations are speech acts in which declarative statements are successfully performed and no psychological state is expressed (e.g. an excommunication or baptism). As will be noted below, this classification had significant consequences for the conceptualization of social action in formal and non-formal contexts. It has been adopted to examine sequences of speech acts in cross-cultural, intra-lingual, and intercultural contexts (e.g. Barron and Schneider Reference Barron and Schneider2009; Blum-Kulka, House, and Kasper Reference Blum-Kulka, House and Kasper1989; Félix-Brasdefer and Koike Reference Félix-Brasdefer and Koike2012; Márquez Reiter and Placencia Reference Márquez Reiter2005), as well as in service encounters (e.g. Kerbrat-Orecchioni and Traverso Reference Kerbrat-Orecchioni and Traverso2008a; Ventola Reference Ventola1987 [in Chapter 2 I analyze service encounters as “sequences of speech acts”]).

Searle (Reference Searle1969) proposed a set of felicity conditions that must be met before a speech act can be said to have a particular illocutionary force: (i) content condition concerns the reference for and predication of an act of a certain type (propositional content); (ii) preparatory condition refers to the circumstances that must be met prior to the performance of the act; (iii) sincerity condition states a true intention of performing, or belief in having performed an illocutionary act on the part of the speaker; and (iv) the essential condition indicates how an utterance is considered; for example, a request is classified as an attempt to get the hearer to perform an action (essential condition). In sum, each condition highlights a different aspect of an utterance: the propositional content focuses only on the textual content; the preparatory conditions emphasize the background circumstances; the sincerity condition reflects the speaker’s psychological state; and, the essential condition centers on the illocutionary point of what is said (Schiffrin Reference Schiffrin1994).

The contributions of Speech Act Theory have influenced empirical research in both discourse analysis and cross-cultural pragmatics. Notions such as uptake, illocutionary force, conventionality, and context have been employed to examine a wide range of speech acts at the discourse level across languages (e.g. Golato Reference Golato2004; Labov and Fanshel Reference Labov and Fanshel1977; Schneider and Barron Reference Placencia, Schneider and Barron2008; Taleghani-Nikazm Reference Taleghani-Nikazm2006; Wierzbicka Reference Wierzbicka2003). Of these, the concepts of illocutionary force and conventional indirectness have received the most attention in the investigation of requests, apologies, compliments, and compliment responses (Blum-Kulka et al. Reference Blum-Kulka, House and Kasper1989; Holmes Reference Holmes1995; Márquez Reiter Reference Marquez-Reiter2000; Ogiermann Reference 269Ogiermann2009). Speech acts are analyzed at the discourse level through a micro-analytic examination of the courses of action implemented through talk (Kasper Reference Kasper, Bardovi-Harlig, Félix-Brasdefer and Omar2006). The focus is on the sequential analysis of speech acts performed for the accomplishment of actions (i.e. openings, closings, speech act sequences, repair sequences). The concept of speech act sequence has been used in studies of classroom discourse (Sinclair and Coulthard Reference Sinclair and Coulthard1975), therapeutic sessions (e.g. Labov and Fanshel Reference Labov and Fanshel1977), academic talk (e.g. Bardovi-Harlig and Hartford Reference Bardovi-Harlig and Hartford1993; Limberg Reference Limberg2010), and service encounters (e.g. Aston Reference Aston1988a; Kerbrat-Orecchioni and Traverso Reference Kerbrat-Orecchioni and Traverso2008a) (see Chapter 2 for an overview of existing research on service encounters from different theoretical and methodological perspectives).

1.3.2 Conversation Analysis: a micro-analytic talk-in-interaction perspective

Conversation Analysis (CA) offers a rigorous methodological framework for the analysis of talk-in-interaction in formal and non-formal contexts. CA was conceptualized in the1960s based on sociological principles from the 1950s. Specifically, CA was influenced by Goffman’s ideas of the presentation of the Self (Reference Goffman1955) and social interaction (Reference Goffman1971, Reference Goffman1983), as well as by Garfinkel’s (Reference Garfinkel1967) contributions on ethnomethodology. Goffman was concerned with the interaction order, that is, the activities or social actions that are accomplished in focused interactions or encounters (also called “face engagements”), while Garfinkel proposed a social theory to analyze the method of reasoning and the social order. The first contributions of CA focused on sequencing in conversational openings (Schegloff Reference Schegloff1968) and the organization of turn-taking in conversation (Sacks, Schegloff, and Jefferson Reference Sacks, Schegloff and Jefferson1974).

CA was designed using the following assumptions: the data are taken from natural (unedited) conversations which can be audio-recorded or video-recorded so that the researcher can access them multiple times; the goal is to arrive at accurate descriptions of specific aspects of social interaction, followed by rigorous transcription conventions (e.g. Jefferson Reference Jefferson and Lerner2004; Schegloff Reference Schegloff2007: 265–269); the data are mainly analyzed qualitatively, but CA does not completely dismiss the informal (crude) quantification on distributional grounds (e.g. Clayman and Gill Reference Clayman, Gill, Bryman and Hardy2004: 592; Hutchby and Wooffitt Reference Hutchby and Wooffitt2008: 108–112); CA focuses on the analysis of individual and deviant cases for rigorous examination, via repeated analysis. In general, CA is concerned with the endogenous organization of talk-in-interaction in naturalistic contexts.

The goal of CA is to explain how sequences are organized. Sequences are understood as “courses of action implemented through talk” (Schegloff Reference Schegloff2007: 3). The adjacency pair (e.g. offer–refusal, compliment–compliment response) is central to sequence construction. It should be noted that the two basic elements of the adjacency pair, “First” and “Second,” are understood as design features of these turn types (Schegloff Reference Schegloff2007: 20). Adjacency pairs can be organized in immediate turns (e.g. A: ‘I love your sweater’ [First-Pair Part]; B: ‘Thanks’ [Second-Pair Part]). Another notion in CA is conditional relevance (Schegloff Reference Schegloff1968; Schegloff Reference Schegloff2007: 20–21): the idea that the occurrence of a second-pair part is conditioned by the occurrence of the first-pair part. That is, given the initial condition that the first-pair part is being uttered, the second part of that pair becomes relevant. The absence of that second part is a “noticeable absence.” According to Hutchby and Wooffitt, “the speaker of the first pair part may infer a reason for that absence” (Reference Hutchby and Wooffitt2008: 45). The adjacency pair, as the minimal sequence, organizes social action and determines coherence in spoken discourse.

CA is also concerned with how turns are constructed. Turns-at-talk comprise turn-constructional units (TCUs) that are utilized to express a wide range of actions such as agreeing, disagreeing, and offering an opinion. TCUs are defined according to three organizational criteria: syntax, prosody, and pragmatics. Syntax provides projectable TCUs within a turn or across turns: for instance, according to Sacks et al. a TCU consists of “sentential, clausal, phrasal, and lexical constructions … Instances of the unit-types so usable allow a projection of the unit-type under way, and what, roughly, it will take for an instance of that unit-type to be completed” (Reference Sacks, Schegloff and Jefferson1974: 702 and footnote 12). Prosody represents the intonational packaging of a TCU: a TCU is “grounded in the phonetic realization of the talk” (Schegloff Reference Schegloff2007: 3–4). And pragmatics encompasses the realization of actions during the negotiation of talk in a particular context. It includes the speaker’s intention and possibly the recognition of that action by the interlocutor: a TCU “constitutes a recognizable action in context” (p. 4). It should be noted that a TCU may not necessarily satisfy these three criteria at once: a TCU may be syntactically and prosodically complete, but pragmatically incomplete (see Ford and Thompson Reference Ford, Thompson, Ochs, Schegloff and Thompson1996). Projection of the TCU is what allows turn-units to be expanded and organized sequentially to achieve a specific purpose in communication.

As noted by Schegloff, a speaker starting to talk has the right and obligation to produce at least one TCU, which may comprise one or more actions. When a speaker nears possible completion of the first TCU in a turn, transition to the next speaker can become relevant at transition-relevance places (TRP). A TRP represents possible completions (e.g. where the current speaker should exit) or places where the speaker signals a transition to the next speaker immediately after the necessary completion of the TCU-in-progress (Sacks et al. Reference Sacks, Schegloff and Jefferson1974; Schegloff Reference Schegloff2007).

CA has been concerned with sequential organization (e.g. openings, closings, speech act sequences, pre-/post-sequences), the organization of turn-taking (e.g. overlap), the organization of preference or dispreference, the concept of repair, the organization of laughter, and the organization of topic development (see Schegloff Reference Schegloff2007 and Sidnell Reference Sidnell2010). CA methods have been applied to other types of spoken discourse, such as institutional talk in formal and non-formal settings (Drew and Heritage Reference Drew and Heritage1992a), children’s talk, human-computer interaction, and disorderly talk in aphasic speakers (for an overview, see Hutchby and Wooffitt Reference Hutchby and Wooffitt2008: 189–205). However, CA has been criticized for qualitative analysis of single interactions taken from ordinary conversations. Further, CA’s notion of context is restricted to the endogenous organization of talk that is present in the text and in the local sequential contexts (what precedes and follows the main sequence) (Clayman and Gill Reference Clayman, Gill, Bryman and Hardy2004: 603).

This approach to discourse has been applied to research on service encounters (e.g. Aston Reference Aston1988a; Gavioli Reference Gavioli and Caliumi1993, Reference Gavioli1995, Reference Gavioli, Bargiela-Chiappini and Harris1997; Kerbrat-Orecchioni and Traverso Reference Kerbrat-Orecchioni and Traverso2008a) (in Chapter 2 I explain how CA has improved our understanding of social action in service encounter research). CA tools (e.g. adjacency pair, TCU, TRP, turn-taking patterns, repair, and preferred and dispreferred actions) are essential for the analysis of interactions in service encounters with regard to how individual actions are organized into sequences in openings, closings, request–response sequences, laughter episodes, and turn-taking mechanisms, for example. These concepts will be central to the analysis of transactional (Chapters 3–6) and non-transactional talk (Chapter 7) (see Introduction [Figure 1] and Section 1.4.6 for an integrated model of pragmatics and discourse employed in this book). I used a modified version of the transcription notations of CA (Jefferson Reference Jefferson and Lerner2004) for the examples presented in Chapters 3–8 (see Transcription conventions at the beginning of this book).

1.3.3 Systemic Functional Linguistics: a social-semiotic and genre perspective

Systemic Functional Linguistics (SFL) was influenced by Malinowski and Firth’s functional view of language as a mode of action in the context of situation (see Introduction), and later refined and extended by Halliday (Reference Halliday and Lyons1970, Reference Halliday1978), Halliday and Hasan (Reference Halliday and Hasan1980), and Ventola (Reference Ventola1987, Reference 273Ventola2005). Two important concepts include the context of situation and the generic structure of texts. The context of situation comprises three elements: field of discourse (what is happening), tenor of discourse (participants’ roles), and mode of discourse (the medium or channel in which the language is realized, namely, oral, written, visual). The generic structure of texts such as service encounters is analyzed with regard to three semiotic planes: genre, register, and language (Ventola Reference 273Ventola2005). Genre defines the necessary elements that characterize a particular text (e.g. service encounters at a store or at a post office) and refers to the “cultural expectations of appropriate unfolding of social activity in the society” (Ventola Reference 273Ventola2005: 29). Register addresses variation with regard to the similarities or differences in the choices of field, tenor, and mode. Both of these planes of discourse, genre and register, are realized through the third semiotic plane, language (i.e. semantics, lexicogrammar, phonology).

The theoretical concepts of this approach are useful because they allow the researcher to examine the structure of social interaction in service encounters. Notions such as the context of situation, the generic structure of texts (e.g. service encounters), and the planes of discourse will be used to examine similarities and differences across subgenres of service encounters (I pick up on this issue in Chapter 4, which analyzes variation in the generic structure of delicatessens and small shops. In Chapter 5 I look at market encounters that have a different generic structure from delicatessens and small shops). Finally, analysts look at the generic structure of service encounters with regard to transactional talk and its obligatory elements such as sale request, sale purchase, and its optional elements which include openings or greetings, for example (see Chapter 2, Section 2.5.1, for a discussion of the genre of service encounters from a systemic-functional perspective).

1.3.4 Comprehensive discourse analysis: a doctor–patient discourse perspective

Labov and Fanshel’s model (Reference Labov and Fanshel1977) represents a comprehensive approach to discourse that examines the sequential structure of therapeutic discourse. Owing to the complexity of social practices in this institutional setting, Labov and Fanshel’s approach included the following notions: Goffman’s (Reference Goffman1971) view of conversation as a form of interaction, Hymes’ (Reference Hymes, Gladwin and Sturtevant1962) notion of speech event, Austin’s (Reference Austin1962) and Searle’s (Reference Searle1969) Speech Act Theory (see Section 1.3.1), and Conversation Analysis (see Section 1.3.2). The data are taken from naturally occurring situations in therapeutic conversations between a patient and a therapist. Social actions such as request and refusal sequences across multiple turns are analyzed at the discourse level. Merritt’s (Reference Merritt1976a,Reference Merrittb, Reference Merritt1976b) analysis of service encounters was based on a modified version of Labov and Fanshel’s discourse approach (coherence) (see Section 1.3.6 below). Merritt’s work looks at social action between customers and servers who negotiate service in face-to-face interactions. It adopts a cognitive and a discursive approach to social interaction.

1.3.5 Interactional Sociolinguistics: an inferential approach to situated communicative practices

From an anthropological point of view, Gumperz (Reference Gumperz1982, Reference Gumperz, Schiffrin, Tannen and Hamilton2001) proposed Interactional Sociolinguistics (IS) as an approach to discourse analysis that analyzes the ability to interpret participants’ meaning in situated communicative practices. This inferential perspective was influenced by different research traditions: Hymes’ ethnography of speaking (in particular, situations of speaking) (Hymes Reference Hymes, Gladwin and Sturtevant1962), Goffman’s (Reference Goffman1981, Reference Goffman1983) interactional order and frames in situated settings, Garfinkel’s (Reference Garfinkel1967) ethnomethodological framework of social interaction, and Grice’s (Reference Grice, Cole and Morgan1975) inferential model of conversational implicature. According to Grice, communication is an intentional process and relies on the speaker’s meaning and the hearer’s ability to interpret the speaker’s intentions. Gumperz’s model is based on (shared) background knowledge that allows the participants to infer meaning through verbal or non-verbal information. He used the term contextualization cue to refer to verbal and non-verbal signs that can be inferred from the situation, in co-occurrence with other lexicogrammatical information. Examples include prosodic cues (e.g. intonation, stress), pronunciation, non-verbal signs (e.g. eye gaze, strong hand shake), and code-switching.

As shown in Chapter 2, several studies in service encounters selected IS to examine the complexities of social action in intercultural contexts (e.g. Kidwell Reference Kidwell2000). The analysis is based on the interpretation of the interlocutors’ utterances during a sales transaction. In Chapter 6 (Section 6.4.5) I examine the pragmatic function of prosodic elements (e.g. high and low terminals) in the visitors’ requests that function as contextualization cues.

Finally, this approach to discourse is fruitful for the analysis of pragmatic variation in pronominal forms during the negotiation of service. The selection of the appropriate pronominal form (e.g. “informal Tu vs. formal Vous’” by the same speaker in the same conversation) creates a pragmatic effect on the interlocutor to express solidarity or deference politeness (in Chapter 8 I use this approach to examine pragmatic variation of pronominal forms).

1.3.6 Discourse coherence: a discourse-analytic approach to language use in context

Schiffrin’s integrative model of discourse analysis is applied to the analysis of discourse markers (Reference Schiffrin1987) in natural conversations. It is a model of discourse coherence that echoes Gumperz’s (Reference Gumperz1982) integrated model of coherence and situated communicative meaning. In her 2006 revised approach, Schiffrin described five planes of discourse:

Participation framework refers to the various ways in which people organize and establish social interaction by “adopting and adapting roles, identities, and ways of acting and interacting” (Reference Schiffrin, Fasold and Connor-Linton2006: 195). This framework includes the relationship between the speaker and the hearer, and how the participants interpret what is said in social interaction. This approach echoes Goffman’s (Reference Goffman1981) notion of participation framework and footing (i.e. alignments in conversation), and Gumperz’s (Reference Gumperz1982, Reference Gumperz, Schiffrin, Tannen and Hamilton2001) communicative practices and inferential model to assess the speaker’s messages.

Exchange structure refers to how people take turns in two-party or multi-party interactions, and how they overlap. This plane of discourse alludes to the methodological framework of CA (see Section 1.3.2).

Act structure combines notions of Speech Act Theory (Austin Reference Austin1962; Searle Reference Searle1969) and CA (speech act sequences organized through adjacency pairs) with regard to the sequential organization of actions (speech acts) accomplished through talk.

Information state refers to the organization and distribution of new or old information in social interaction. It concerns what the speaker and hearer know from the message conveyed, and what common knowledge the interlocutors can assume in conversation. This plane highlights the cognitive or inferential capacity of the interlocutors to make appropriate assessments of what is said. This approach echoes Gumperz’s (Reference Gumperz, Schiffrin, Tannen and Hamilton2001) inferential view of discourse (IS) (see Section 1.3.5).

Idea structure concerns the manner in which information is organized or linked within and across sentences. It alludes to Halliday’s (Reference Halliday1978) notions of cohesion or genre structures (Halliday and Hasan Reference Halliday and Hasan1980).

Under this integrated model of discourse, every utterance has meaning and is part of several planes of discourse. That is, the speaker’s utterance is multifunctional and may interact simultaneously with different planes of discourse. This approach offers important analytical tools for the analysis of speech act sequences, the participants’ roles, and turn structure of service encounter interactions. As noted above (Sections 1.3.4 and 1.3.5), Merritt (Reference Merritt1976b) represents one of the early studies that used a model of discourse coherence to examine the structure of US service encounters.

1.3.7 Joint activity in face-to-face conversation: joint actions in language use

Clark (Reference Clark1996) offers an insightful approach for the analysis of language use in face-to-face interaction which adopts the conversation as the basic setting. He uses the concept of joint action, that is, action “carried out by an ensemble of people acting in coordination with each other” (p. 3). He emphasizes the need to analyze natural social interactions in face-to-face conversation, which has the following features: immediacy (copresence, visibility, audibility, and instantaneity), the medium of how meaning is expressed in interaction (evanescence, recordlessness, and simultaneity), and control of the interaction (extemporaneity, self-determination, and self-expression) (pp. 9–10). Following Levinson’s (Reference Levinson, Drew and Heritage1992) notion of activity type, Clark employs the interactional concept of joint activity to refer to the social setting or speech event in which social action takes place (e.g. a family dinner, a classroom, or a supermarket). A joint activity is defined by the setting (e.g. a service encounter), the participants’ roles (e.g. a clerk and a customer), and the allowable contributions for each activity (e.g. the rights and obligations of the interlocutors during the interaction). Clark adopts an extended version of Speech Act Theory (Austin Reference Austin1962; Searle Reference Searle1969) and key methodological concepts of CA (Sacks et al. Reference Sacks, Schegloff and Jefferson1974) to examine individual communicative acts performed by at least two participants who engage in joint actions. Participants share “common ground” (mutual beliefs or expectations) during a joint activity. According to Clark, there are two types of common ground: (i) communal common ground, which refers to information based on the “cultural communities” (Reference Clark1996: 121) to which people believe they belong (e.g. a clerk and a customer as members of a supermarket); and (ii) personal common ground, which concerns the information two people believe they share (e.g. same language and dialect).

Clark adopts a pragmatic-discourse perspective to examine language use in social interaction. The concepts of joint action in a greeting exchange and joint activity in a sales transaction can be applied to the analysis of interactions in service encounters. Clark used this approach to examine the structure (joint activity of a sales transaction) and the participants’ roles at a drug store (Reference Clark1996: 31–47). In this book, the pragmatic-discursive approach to face-to-face service encounters in two sociocultural settings, the United States and Mexico, is influenced by Clark’s notions of joint actions, joint activity, and common ground.

1.3.8 Multimodal Discourse Analysis: a verbal and non-verbal modal perspective

Multimodal Discourse Analysis examines language use in context that emerges from different verbal and non-verbal modes of discourse. In their introductory chapter, Scollon and LeVine (Reference Scollon, LeVine, LeVine and Scollon2004) argue that discourse is multimodal. It includes data from both verbal and non-verbal modes in order to offer a better understanding of the actions performed in social interaction. Although in this book I examine the linguistic resources of service encounters (based on audio-recorded transactions), I also comment on non-verbal actions: the presence of the customer at the designated setting, laughter episodes, and prosodic resources in the request for service (e.g. intonation, stress, duration).

The aforementioned integrative approaches to discourse (Sections 1.3.2–1.3.8) are influenced by Goffman’s (Reference Goffman1967, Reference Goffman1971, Reference 262Goffman1974, Reference Goffman1981, Reference Goffman1983) interaction order, Speech Act Theory (Austin Reference Austin1962; Searle, Reference Searle1969), Grice’s (Reference Grice, Cole and Morgan1975) intention-based model, Levinson’s (Reference Levinson, Drew and Heritage1992) notion of activity type, and CA’s methodological framework. While these approaches examine social interaction through the analysis of speech act sequences (or joint actions), turn-taking, and sequence organization (joint activities across discourse), variation at the pragmatic level is not given sufficient attention. The following approach takes a sociolinguistic perspective in order to examine variation at different levels of pragmatics and discourse.

1.3.9 Variational Pragmatics: intra-lingual variation in varieties of a language

Variational Pragmatics (VP) examines variation across varieties of a language (intra-lingual variation or Fried’s [Reference Fried, Fried, Östman and Verschueren2010] local perspective of linguistic variation [see Introduction]). It adopts an integrative model of spoken discourse (Barron and Schneider Reference Barron and Schneider2009: 429; Schneider Reference Schneider and Fried2010; Schneider and Barron Reference Placencia, Schneider and Barron2008: 19). The authors proposed a framework that consists of five levels of pragmatic analysis: (i) the formal level, which includes the analysis of linguistic forms like discourse markers (e.g. ‘well’); (ii) the actional level, which focuses on the pragmalinguistic strategies that comprise the speech act type; (iii) the interactional level, which examines sequential patterns of speech acts (e.g. request–response, greetings, terminal exchanges); (iv) the topic level, which refers to topic selection, topic development, or topic change in sequences of propositions; and finally, (v) the organizational level, which alludes to turn-taking, overlap, and interruptions. VP emerged as a result of previous research in cross-cultural pragmatics that had not focused on the effect of region (Blum-Kulka et al. Reference Blum-Kulka, House and Kasper1989). Unlike previous models, VP takes a pragmatic-discursive approach to the analysis of social action from a sociolinguistic perspective. For example, using VP and corpus linguistics, Jautz (Reference Jautz2013) analyzed intra-lingual pragmatic variation in expressions of gratitude in British and New Zealand English. The author found regional differences at the actional and discourse levels. At the actional level, differences were observed in the choice of different thanking formulas. Thank you was used frequently in British English, while thank you and thanks were employed almost equally in the New Zealand corpus. At the discourse level, the thanking expression varied with respect to the position in the conversation: in the British English data, the thanking expressions occurred most frequently in final position, while in the New Zealand data, it was found most in middle position. More recently, Terkourafi (Reference Terkourafi2011, Reference 272Terkourafi, Félix-Brasdefer and Koike2012) critically reviewed the model of VP by looking at the pragmatic variable and the ways in which variants can apply at different levels of discourse. (In Chapter 9 [Conclusions], I revisit the issue of pragmatic variation from a variationist perspective and propose future directions in service encounters research from a pragmatic variationist view.)

In their 2008 edited volume on variational pragmatics, Schneider and Barron included an article that examined regional variation in service encounters in two small shops in Ecuador (one in Quito and one in Manta). In their edited volume on pragmatic variation in first and second languages, Félix-Brasdefer and Koike (Reference Félix-Brasdefer and Koike2012) adopted a modified version of this approach in order to examine pragmatic variation across languages, and varieties of a language. A variationist perspective on service encounters advances our understanding of regional variation at the sociolinguistic–pragmatics interface (the issue of pragmatic variation in relation to service encounters is further analyzed in Chapter 2, Section 2.5).

1.3.10 Critical Discourse Analysis: analyzing service in media discourse

Critical Discourse Analysis (CDA, van Dijk Reference van Dijk1993) is an integrated approach to the analysis of relations between power and discourse. It represents a cognitive approach to the study of attitudes, knowledge, and ideologies. This approach is concerned with the role of discourse in both the production and the challenge of dominance by elite groups or institutions that result in social inequality (e.g. gender, racial, and political). From a discourse-analytic perspective, van Dijk (Reference van Dijk1993) examined the properties of the “text” of speech, analyzing speech acts at the local level, as well as macro-speech acts (Reference van Dijk1980) (global level), through which the speaker accomplished the speech act (e.g. an assertion). Macro-speech acts consist of speech act sequences taken as a whole, and function as global acts (e.g. a global request [‘I hope you can help me. I need to get to school and my car broke down. Can I borrow yours?’]). They comprise sequenced speech actions that are accomplished by one speaker in one turn (e.g. a political debate that includes assertions, complaints, and directives). This approach includes the negotiation of those speech acts by two or more speakers (e.g. defending, attacking, persuading) (interaction sequence).

Although this approach has not been explicitly used to examine interactions in service encounters, its discourse-analytical tools and the notions of power and discourse can be employed to do so. For example, the macro- and micro-structure of media service encounters can be analyzed with regard to the overall organization of a sales transaction (e.g. openings, sales request, sales compliance, payment, closing) and the sequential analysis of speech act sequences (e.g. greeting and request for service/compliance). And given the asymmetric nature of service encounters and shifts of alignment in the participants’ roles (Goffman Reference Goffman1981), issues of power and dominance can shed light upon the power roles of servers and customers (see Chapter 7, Section 7.3.3, for an analysis of customer dissatisfaction and alignment of social power between the server and customer).

1.3.11 Computer-Mediated Discourse: negotiating e-service encounters

Unlike the previous approaches, which examined discourse in either spoken or written texts, Computer-Mediated Discourse Analysis (CMDA) studies language use in online interaction. In these interactions participants interact through verbal language that is typed on a keyboard and read on a computer screen, as in the norm for e-mail, chat, social media such as Facebook, or negotiated through live video and audio (e.g. Skype). In her model of CMDA, Herring (Reference Herring, Barab, Kling and Gray2004) proposed four domains of language: structure, meaning, interaction, and social behavior. Of these, two, meaning and interaction, can be applied to the negotiation of online service. At the level of meaning, Herring includes one or more ordered speech acts (e.g. a clarification question or a series of speech acts in an e-mail message). At the interactive level, sequences of speech acts are negotiated by means of “interactive exchanges” by two or more participants (e.g. Skype [spoken discourse] or chat [written discourse]).

These levels of language can be applied to the analysis of the emerging genre of e-service encounters. Placencia (forthcoming) examines the negotiation process of online service among native speakers of Spanish with regard to the address forms used during the online transaction. Garcés-Conejos Blitvich (forthcoming) provides a programmatic agenda for the analysis of e-service encounters in online interaction. The author reviews some of the theoretical and methodological models that are commonly used to analyze the structure of e-service encounters within the framework of computer-mediated communication. Finally, Zahler (Reference Zahler2013) (using Herring’s [Reference Herring, Barab, Kling and Gray2004] CMDA model) offers a contrastive analysis of online personal ads with regard to the speech act strategies used to request a personal encounter among users in Mexico City and London. The negotiation of e-service represents a rich research desideratum, since many current consumers prefer sales transactions conducted through internet sites such as e-Bay, Vivastreet, and many other department stores that offer online sales.

1.3.12 Discourse approaches to the analysis of face and (im)politeness

This section provides an overview of selective approaches that examine the negotiation of face to express connectedness and dissociation, facework, and (im)politeness practices. It focuses on approaches to discourse that represent the postmodern view of politeness (Bargiela-Chiappini and Haugh Reference Bargiela-Chiappini and Haugh2009; Kádár and Haugh Reference Kádár and Haugh2013; Locher and Watts Reference 267Locher and Watts2005; Watts Reference Watts2003). These approaches to (im)politeness, face, and facework make important contributions to the analysis of transactional and non-transactional talk in service encounter interactions (see Introduction [Figure 1] and Chapter 7 for an analysis of non-transactional talk, face, and politeness).

Rapport management Spencer-Oatey (Reference Spencer-Oatey and Spencer-Oatey2000) revised Brown and Levinson’s (Reference Brown1987 [Reference Brown, Levinson and Goody1978]) model of linguistic politeness and proposed a modified framework for the analysis of face and rapport management. According to Spencer-Oatey, rapport management, or the management of harmony–disharmony among people, comprises two components: (i) face management, which refers to the management of face needs; and, (ii) the management of sociality rights, which concerns the management of social expectancies that are “fundamental personal/social entitlements that individuals effectively claim for themselves in their interaction with others” (Reference Spencer-Oatey and Spencer-Oatey2000: 14, emphasis in the original). Each of these components has two aspects. First, face management, which includes quality face (e.g. “we have a fundamental desire for people to evaluate us positively in terms of our personal qualities,” p. 14) and identity face (e.g. acknowledgment of social identities or group roles as a valued customer, close friend, or teacher). Second, sociality rights, which include equity rights (i.e. personal consideration from others/autonomy-imposition) and association rights (i.e. belief of association with others). Each of these components can be analyzed from a personal or independent perspective (quality face and equity rights) and from a social component (identity face and association rights).

Spencer-Oatey proposed five domains of discourse for the appropriate management of face (Reference Spencer-Oatey and Spencer-Oatey2000: 19–20):

1. (i) Illocutionary domain: This domain refers to the pragmalinguistic strategies used to perform speech acts: degrees of directness/indirectness, strategy type, and internal modification (similar to Barron and Schneider’s [Reference Barron and Schneider2009] actional level);

2. (ii) Discourse domain: This domain includes an analysis of the discourse content and discourse structure of an interchange. It concerns topic choice and topic management, and the sequencing of information (e.g. openings, speech act sequences) and is similar to Barron and Schneider’s (Reference Barron and Schneider2009) interactional and topic levels;

3. (iii) Participation domain: This domain concerns the analysis of procedural aspects of an interchange through the organization of turn-taking, as well as the rights and obligations of participants (similar to Barron and Schneider’s [Reference Barron and Schneider2009] organizational level);

4. (iv) Stylistic domain: This domain comprises stylistic aspects of an interchange such as choice of tone and the appropriate forms of address and honorifics; and,

5. (v) Non-verbal domain: This domain addresses the non-verbal aspects of an interchange such as gestures, body movement, eye gaze.

VP and rapport management show similarities and differences: while both propose various levels of analysis from a pragmatic and discursive perspective, Spencer-Oatey’s model shows a preference for natural (face-to-face) conversation, whereas VP adopts an eclectic approach using natural and experimental methods (e.g. role plays, written questionnaires). And while rapport management is largely qualitative in nature, VP adopts descriptive and inferential methods to quantify language use in context. Finally, VP focuses on intra-lingual pragmatic variation or variation within a language, while rapport management looks at cross-cultural pragmatic variation or variation across languages or in intercultural contexts.

An approach to politeness in service encounters Kerbrat-Orecchioni (Reference Kerbrat-Orecchioni2001, Reference Kerbrat-Orecchioni, Hickey and Stewart2004, Reference Kerbrat-Orecchioni2006) examined politeness in social interaction in small shops in France (Kerbrat-Orecchioni and Traverso Reference Kerbrat-Orecchioni and Traverso2008a). This author’s discourse model is based on a modified version of Brown and Levinson’s (Reference Brown, Levinson and Goody1978, Reference Brown1987) framework of linguistic politeness. In her model she looks at politeness practices and the sequential organization of small shop encounters, and includes contributions from Lakoff (Reference Lakoff, Corum, Smith-Stark and Weiser1973), Leech (Reference Leech1983), and Goffman (Reference 262Goffman1974). Kerbrat-Orecchioni proposed alternative concepts for the analysis of politeness practices, such as Face-Flattering Act (FFA) (instead of Brown and Levinson’s Face-Threatening Act [FTA]), politeness, impoliteness, apoliteness (normal absence of politeness markers), and hyperpoliteness, or politeness in excess. It is important to note that her approach is applied to discourse in small shops at various levels of analysis: actional (requests, offers), interactional (speech act sequences), and politeness practices (Kerbrat-Orecchioni and Traverso Reference Kerbrat-Orecchioni and Traverso2008a; see Chapter 2, Section 2.5.4).

Face, facework, and (im)politeness in social interaction Departing from Brown and Levinson’s (Reference Brown1987) universal model of linguistic politeness, post-modern approaches to politeness include relational work for the analysis of appropriate and polite behavior (Locher and Watts Reference 267Locher and Watts2005), face as relational and interactional, and facework as the product of an interactional achievement (Arundale Reference Arundale, Bargiela-Chiappini and Haugh2009). Following Goffman’s work, a face-to-face service encounter is considered a focused interaction that can be negotiated through verbal or non-verbal actions. Further, recent work on face and facework (Bargiela-Chiappini and Haugh Reference Bargiela-Chiappini and Haugh2009) has convincingly shown that face is an interactional and relational phenomenon. And Culpeper (Reference Coupland2011) proposed an approach to the examination of impoliteness at the discourse level that examines social action such as inappropriate, rude, or impolite behavior in formal and non-formal institutional contexts (in Chapter 7 I examine the negotiation of face and im/politeness at a visitor information center). In the context of service encounters, the notions of face (e.g. involvement or independence [Scollon and Scollon Reference Scollon and Scollon2001]), facework (i.e. the strategies we use to negotiate face in interaction), and polite or appropriate behavior are central to the negotiation of service. The issue of whether social actions such as greetings and direct or indirect requests, for example, are perceived as polite, impolite, or socially appropriate depends on the interlocutor, the communicative situation, and the culture (in Chapters 3–6 I will comment on polite practices or appropriate behavior in transactional talk, and in Chapters 7–8 I examine polite practices and the negotiation of face in non-transactional talk).

1.4.1 Social action in face-to-face interaction

The analytical framework for my analysis of service encounters is influenced by Goffman’s (Reference Goffman1963, Reference Goffman1967, Reference Goffman1971, Reference 262Goffman1974, Reference Goffman1981, Reference Goffman1983) ideas on interaction order and encounters as focused engagements. According to Goffman, face-engagements consist of “all those instances of two or more participants in a situation joining each other openly in maintaining a single focus of cognitive and visual attention – what is sensed as a single mutual activity, entailing preferential communication rights” (Reference Goffman1963: 88, emphasis in original). This definition echoes Clark’s (Reference Clark1996) concept of joint activity (see Section 1.3.7), which is realized in the form of coordinated joint actions. In my analysis I consider a service encounter a joint activity that is realized through a series of joint actions that make up the entire sales transaction. I will focus on speech act sequences that are mutually constructed: openings, closings, request-response sequences, offer-responses, etc. I focus on the analysis of social actions realized by two or more participants during the negotiation of meaning, including transactional and non-transactional talk (see Introduction, Figure 1).

1.4.2 Setting

Service encounters occur in specific social situations that consist of “any physical area anywhere within which two or more persons find themselves in visual and aural range of one another” (Goffman Reference Goffman1981: 84). Besides the physical area in which the transaction takes place, the setting includes the position of the participants (server and customer), the products, the counter, the place of the recording device, and whether or not customers have direct access to the products. Although a physical location is not necessarily a requirement for service encounters to take place, the encounters analyzed in this book took place in designated locations and in comparable settings: small shops, delicatessens, markets, and a visitor information center. (A visual description of each setting is provided at the beginning of each chapter [see Chapters 3–6].)

1.4.3 Changes of frame and footing

In this book service encounters are analyzed with regard to transactional and non-transactional talk (see Introduction, Figure 1). I examine changes in footing with regard to the alignment of participants’ roles from transactional talk (e.g. request for service) to relational talk (e.g. small talk or phatic exchanges). According to Goffman, a “change in footing implies a change in the alignment we take up to ourselves and the others present as expressed in the way we manage the production and reception of an utterance” (Reference Goffman1981: 128). During the negotiation of service, customers and servers constantly change their footing to express an orientation toward the task (sales purchase), as well as an orientation toward interpersonal relations that expresses involvement with, or independence from, the interlocutor. Since most service encounters involved changes of alignment in the participants’ roles, in this book I examine the discursive function of small talk, laughter, joking episodes, and phatic exchanges embedded in transactional talk (see Chapters 7 and 8 for additional information on changes of footing).

1.4.4 Service encounters in context

I assume that language use (e.g. face-to-face service encounters) occurs in sociocultural and cognitive contexts. Since the environment surrounding the utterance (co-text) is important for the understanding of meaning, in this book utterances are not divorced from context. Understanding of context is central to the contextualization of social actions that are produced and interpreted in cognitive contexts (contextual assumptions and factual assumptions), cultural contexts (shared meaning and perspectives of the world), and social contexts (interaction between self and interlocutor).¹ I assume that the language used to accomplish social action in service encounters is context sensitive, specifically, that “language reflects those contexts because it helps to constitute them” (Schiffrin Reference Schiffrin1987: 5). Finally, the role of inference in spoken discourse is widely accepted by various approaches to discourse analysis that examine language use in context (e.g. Labov and Fanshel Reference Labov and Fanshel1977; Schiffrin Reference Schiffrin1987; Scollon and Scollon Reference Scollon and Scollon2001). Merritt’s (Reference Merritt1976a, Reference Merritt1976b) work employed a discourse coherence model to examine the pragmatic interpretation of social actions in U.S. service encounters (see Chapter 2, Section 2.2).

1.4.5 Variation in service encounters

Pragmatic variation plays a central role in service encounters. Variation can be analyzed with regard to the sequential structure of interactions, the setting (i.e. genre variation), as well as variation that is influenced by macro-social factors (e.g. gender, age, and ethnicity). Since the data in this book were collected in two languages (English and Spanish) and across varieties of the same language, I analyze cross-cultural variation (Chapter 3) and intra-lingual variation (Chapter 4). Pragmalinguistic variation at the actional and interactional levels will be analyzed in Chapters 3–6. A revised version of the variational pragmatics approach will be used to examine the pragmatic variable and its recdizations (actional level) (Félix-Brasdefer and Koike Reference Félix-Brasdefer and Koike2012; Schneider Reference Bataller, Fernández-Amaya and Hernandez-Lopez2010) (see Chapter 5 and my conclusions for this book in Chapter 9).

1.4.6 Integrative model of discourse for the analysis of service encounters

The framework adopted in this book is an integrative model of spoken discourse. As shown in the discussion above, most approaches to social action adopt an integrative approach because a combination of congruent theoretical and methodological research traditions yields a broader understanding of language use in social interaction than just one approach. The proposed framework is influenced by two perspectives of social interaction discussed above, namely, Variational Pragmatics (Barron & Schneider Reference Barron and Schneider2009: 429; Schneider and Barron Reference Placencia, Schneider and Barron2008: 19) (Section 1.3.9) and rapport management (Spencer-Oatey Reference Spencer-Oatey and Spencer-Oatey2000: 19–20) (Section 1.3.12).

The discourse approach I adopt to examine service encounters is interdisciplinary. It includes theoretical and methodological notions from Goffman (Reference Goffman1961, Reference Goffman1971, Reference Goffman1981) (face-to-face encounters and changes in footing), Speech Act Theory, Discourse Analysis, and Conversation Analysis (see Section 1.3). The data in this book comprise face-to-face interactions in naturalistic settings. By natural data I mean authentic interactions that are recorded in natural settings (e.g. delicatessens, markets, small shops, a visitor information center). Methods used to gather elicited or experimental data in pragmatics (role plays or written questionnaires, for example) are not suitable for the analysis of social interaction in service encounters; in particular, these methods are unsuitable for the analysis of non-transactional talk.

The current framework includes eight levels. Although some of these levels closely resemble those of Variational Pragmatics (e.g. Schneider Reference Schneider and Fried2010; Schneider and Barron Reference Placencia, Schneider and Barron2008), the present framework differs methodologically with regard to the type of data selected and the unit of analysis. The data in Chapters 3–8 are analyzed with regard to one or more levels of pragmatic analysis:

Formal level This level concerns the formal analysis of linguistic expressions with regard to form, function, and force in specific contexts. It includes the analysis of discourse markers, epistemic expressions, and backchannels: for example, Schiffrin’s (Reference Schiffrin1987) analysis of discourse markers in American English, Wierzbicka’s (Reference Wierzbicka2006) analysis of epistemic expressions in English and other languages, or the contrastive analysis of backchannel expressions or response tokens in the United States and England (Tottie Reference Tottie, Aijmer and Altenberg1991) and in British and Irish discourse (O’Keeffe and Adolphs Reference O'Keeffe, Adolphs, Schneider and Barron2008). In these studies linguistic expressions are examined in naturalistic settings with regard to frequency and distribution. Moreover, discourse markers such as ‘OK,’ ‘oh,’ ‘yes,’ or ‘well’ were examined in bookshops (Mansfield Reference Mansfield and Aston1988: 229–230) and self-service stores (Merritt Reference Merritt, Baugh and Sherzer1984). The authors focused on the pragmatic and discourse functions of the discourse marker “OK” in retail service encounters.

Actional level This level concerns the analysis of the pragmalinguistic strategies used in speech acts (e.g. requests, offers, greetings). For example, a request consists of a “head act” (e.g. ‘can I have a half a pound of ham?’) and an internal modification (e.g. ‘could I please have a half a pound of ham?’). The aim at this level is to provide an analysis of the pragmalinguistic expressions (Leech Reference Leech1983; Thomas Reference Thomas1983) used to convey the illocutionary force of the action. Issues of directness or indirectness, or politeness or impoliteness can also be analyzed at this level. For instance, Placencia (Reference Placencia2005) examined the pragmalinguistic expressions used to express a request for action in service encounters in small shops in Spain and Ecuador, while Anderson, Guy, and Tucker (Reference Anderson, Aston, Tucker and Aston1988) analyzed the pragmalinguistic forms of customer requests in bookshop encounters. Further, Kerbrat-Orecchioni (Reference Kerbrat-Orecchioni2006) looked at politeness practices when making requests in small shops in France.

In this book the actional level is examined through an analysis of request types (the request for service), following Blum-Kulka et al. (Reference Blum-Kulka, House and Kasper1989) and Placencia (Reference Placencia, Schneider and Barron2008) (for the Spanish data). Table 1.1 presents ten request strategy types (request variants) that will be used to examine the data in Chapters 3–6.

All customer-initiated requests were analyzed according to request type and gender of the customer and vendor. In addition, the following elements of internal modification of the request were examined: diminutives, negation prefacing the pre-request or the main request (request head act), the conditional, the politeness form, por favor ‘please’ (Blum-Kulka et al. [Reference Blum-Kulka, House and Kasper1989: 281–283]), and final rising intonation (↑) (in assertions) used as a prosodic downgrader (Félix-Brasdefer Reference Félix-Brasdefer, García and Placencia2011; Placencia Reference Placencia, Schneider and Barron2008: 315).

Interactional level This level centers on the analysis of speech act sequences or joint social actions. It employs tools of Conversation Analysis (see Section 1.3.2) to examine the sequential patterns for the organization of speech actions jointly produced by the customer and the server. It consists of the analysis of openings, closings, and request–response sequences. Examples of this include Aston’s (Reference 257Aston1995a, Reference Aston1995b) analysis of openings and closings in bookshop service encounters, Merritt’s (Reference Merritt1976b) analysis of questions following questions, or elliptical questions–responses in small store transactions, and Filliettaz’s (Reference Filliettaz, Kerbrat-Orecchioni and Traverso2008) co-construction of request–response sequences.

Stylistic level This level focuses on the stylistic aspects of an interchange. It is influenced by Goffman’s (Reference Goffman1981) notions of frame and footing (Goodwin Reference Goodwin, Slobin, Gerhardt, Kyratzis and Guo1996); for example, shifting from business talk (e.g. buying and selling) to a friendly tone (e.g. joking or small talk). It also includes the choice of forms of address used to open, close, and negotiate a business transaction, or shifting from formal (V) to informal (T) styles to express involvement or camaraderie. In her work in service encounters in Spain and Ecuador, Placencia (Reference Placencia2005) showed how Spaniards shifted alignment from a business transaction to relational talk. Bailey (Reference Bailey1997) looked at interethnic negotiations in service encounters between US immigrant Korean retailers and African American customers. The author showed how they shifted from a business frame to socially expanded sequences to talk about the weather, current events, or jokes. And Virpi Ylänne-McEwen (Reference Ylänne-Mcewen2004) analyzed the shifts of alignment from business to interpersonal talk (phatic exchanges) during the negotiation of service in travel agencies. In this book I look at two dimensions of the stylistic level: how participants align (or not) according to their social roles to maintain appropriate levels of rapport and involvement with their interlocutor (Chapter 7); and pragmatic variation of address forms (Chapter 8).

Topic level This level is concerned with discourse content throughout the interaction. It includes topic selection, topic management, topic abandonment, topic shift, and reintroduction of topics. It is concerned with knowing which topics are appropriate or inappropriate to bring up in a conversation, or the choice and management of sensitive topics. For instance, Tannen (Reference Tannen, Fasold and Conner-Linton2006: 344) discussed cultural differences regarding topic choice among Americans in Germany. In intercultural interactions, the Germans introduced topics of politics and religion with American students who did not feel comfortable talking about sensitive topics with someone they did not know well. With regard to service encounters, James (Reference James1992) examined the role of topic control during the negotiation of service at a convenience store; specifically, topic selection, topic shifts, topic development, and whether the topic is initiated by the customer or the seller. This level is relevant to the analysis of service encounters because of its treatment of the presence or absence of small talk (In Chapter 7 I look at the topics selected for small talk exchanges).

Organizational level This level centers on the organization of turn-taking in conversation and is influenced by Conversation Analysis (e.g. Sacks et al. Reference Sacks, Schegloff and Jefferson1974; Schegloff Reference Schegloff2007). It addresses aspects of turn-taking, overlap, interruption, silence, and preference organization. For instance, Tannen (Reference Tannen, Fasold and Conner-Linton2006) associated instances of high-considerateness with a lack of overlap, and speakers who prefer one-at-a-time turn-taking. In contrast, a high-involvement culture was related to speakers who showed a preference for overlap or fast speech to show enthusiasm, solidarity, or interest in the interlocutor’s speech. With regard to service encounters, Placencia (Reference Placencia, Schneider and Barron2008) examined turn-taking procedures in openings and closings in small shop transactions in Ecuador, and showed a preference for a fixed turn-taking format with an absence of overlap in both the phatic and the transactional exchanges. And in an examination of dispreferred responses in bookshop encounters among Britons and Italians, Gavioli (Reference Gavioli1995) analyzed turn-taking procedures for the presence of laughter as a repair device in turn-initial and turn-final position. (In Chapter 6 I examine the structure of dispreferred responses at a US visitor information center.)

Non-verbal level This level consists of social actions performed through gesture. Kendon refers to gestures as “visible acts” when uttering actions (Reference Kendon, Esposito, Bratanic, Keller and Marinaro2010: 1), such as body movement, hand movement, or gaze direction. More specifically, we use gesture “to show what kinds of actions we are taking with our utterances: with gestures we can, among other things, show agreement or disagreement, affirmation or denial, we can show that we are asking a question or begging another’s indulgence, that we are doubtful of something or that what we are saying is hypothetical” (p. 1). The level of non-verbal behavior is relevant for the analysis of service encounters. For instance, Moore (Reference Moore and Robert2008) analyzed meaning expressed through hand signals and body movements to complement verbal messages during the requests for service in print shops. And Kidwell (Reference Kidwell2000) showed the primacy of non-verbal communication in accomplishing the action of requesting (in the absence of verbal actions) in front-desk encounters. To capture the non-verbal signals, this level requires video recording face-to-face interactions and knowledge of the methodological tools that analyze gesture in interaction. In this book, non-verbal information was recorded through detailed field notes during participant observation at the research sites (see Introduction for information on data collection procedures).

Prosodic level Prosody is a pragmatic resource used to express interpersonal or marked meaning. This level focuses on pragmatic meaning that is conveyed through prosodic information: intonation (e.g. low or high pitch), stress, loudness, duration, and timing (e.g. rhythm and rate of speech). For example, these prosodic resources function as “contextualization cues” (Gumperz Reference Gumperz1982; Reference Gumperz, Schiffrin, Tannen and Hamilton2001) (see Section 1.3.5): signals that allow the interlocutor to draw an inference from the speaker’s message. Under the CA view (see Section 1.3.2), prosodic completion is one component of the turn-constructional unit (TCU) (Schegloff Reference Schegloff2007: 3–4). For instance, Couper-Kuhlen (Reference Couper-Kuhlen, D'hondt, Östman and Verschueren2009) showed how an extra high pitch in a TCU (“by the way”) is used to signal that “an upcoming topic proffer is not motivated by anything in prior talk” (p. 184). In her corpus-based analysis of “please” in English, Wichmann (Reference Wichmann2004) noted that in response to an offer (e.g. ‘Yes, please’), a high terminal (rise ↑ or fall rise ᵔ) signals a hearer-oriented perspective, a level tone (→) may sound indifferent, and a falling to low contour (↓) may be perceived as discourteous by the interlocutor (p. 1546). And, in their introduction to the pragmatics–prosody interface, Wichmann and Blakemore state that the main function of prosody is expressive, that is, “to convey emotions and attitudes rather than any referential meaning” (Reference Wichmann and Blakemore2006: 1537). In bookshop service encounters, Mansfield (Reference Mansfield and Aston1988) examined different pragmatic functions of the intonation (rising or falling) of requests, offers, and minimal exchanges. For example, some customer requests for information were uttered with a falling tone (impersonal interpretation), while others ended with a rising intonation to signal polite requests. In service encounters, the interpretation of social action (e.g. request for service, offers, greetings, clarification requests, payment sequence, closing) is contingent upon the interlocutor’s understanding of the prosodic cues that accompany the utterances.

In most service encounters analyzed in this book in delicatessens, small shops, and markets (Chapters 3–5), background noise, such as music, television, bystanders engaged in conversation, or sellers offering their products in a loud voice, affected the clarity of the recordings. In these chapters I provide general comments about prosody (this analysis is based on my perceptual interpretation of the prosodic resources, and verified by one of my research assistants). Given the high quality of the recordings in the visitor information center, however, in Chapter 6 I analyze general aspects of prosody in visitor requests. The acoustic descriptions were realized using the software program Praat (version 5.3.77, Boersma and Weenink Reference Boersma and Weenink2013) (see Section 6.4.5).

These levels of pragmatic analysis aim at providing a comprehensive examination of social action in service encounters in formal and non-formal institutional interactions. Similar to Variational Pragmatics (Barron and Schneider Reference Barron and Schneider2009) or sociopragmatic variation (Márquez Reiter and Placencia Reference Márquez Reiter2005), the present model for the analysis of service encounters is contrastive and aims at examining one or more levels of pragmatic analysis in comparable sociocultural settings, or in one or more varieties of the same language.