285 41 12MB
English Pages 718 [720] Year 2011
Pragmatics of Society HoPs 5
Handbooks of Pragmatics
Editors
Wolfram Bublitz Andreas H. Jucker Klaus P. Schneider Volume 5
De Gruyter Mouton
Pragmatics of Society
Edited by
Gisle Andersen Karin Aijmer
De Gruyter Mouton
ISBN 978-3-11-021441-3 e-ISBN 978-3-11-021442-0 Library of Congress Cataloging-in-Publication Data Pragmatics of society / edited by Gisle Andersen, Karin Aijmer. p. cm. ⫺ (Handbook of pragmatics; 5) Includes bibliographical references and index. ISBN 978-3-11-021441-3 (alk. paper) 1. Pragmatics. 2. Sociolinguistics. I. Andersen, Gisle. II. Aijmer, Karin. P99.4.P72P748 2012 306.44⫺dc23 2011040683
Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.d-nb.de. ” 2011 Walter de Gruyter GmbH & Co. KG, Berlin/Boston Cover image: letty17/iStockphoto Typesetting: Dörlemann Satz GmbH & Co. KG, Lemförde Printing: Hubert & Co. GmbH & Co. KG, Göttingen ⬁ Printed on acid-free paper Printed in Germany www.degruyter.com
Preface to the handbook series Wolfram Bublitz, Andreas H. Jucker and Klaus P. Schneider The series Handbooks of Pragmatics, which comprises nine self-contained volumes, provides a comprehensive overview of the entire field of pragmatics. It is meant to reflect the substantial and wide-ranging significance of pragmatics as a genuinely multi- and transdisciplinary field for nearly all areas of language description, and also to account for its remarkable and continuously rising popularity in linguistics and adjoining disciplines. All nine handbooks share the same wide understanding of pragmatics as the scientific study of all aspects of linguistic behaviour. Its purview includes patterns of linguistic actions, language functions, types of inferences, principles of communication, frames of knowledge, attitude and belief, as well as organisational principles of text and discourse. Pragmatics deals with meaning-in-context, which for analytical purposes can be viewed from different perspectives (that of the speaker, the recipient, the analyst, etc.). It bridges the gap between the system side of language and the use side, and relates both of them at the same time. Unlike syntax, semantics, sociolinguistics and other linguistic disciplines, pragmatics is defined by its point of view more than by its objects of investigation. The former precedes (actually creates) the latter. Researchers in pragmatics work in all areas of linguistics (and beyond), but from a distinctive perspective that makes their work pragmatic and leads to new findings and to reinterpretations of old findings. The focal point of pragmatics (from the Greek prãgma ‘act’) is linguistic action (and inter-action): it is the hub around which all accounts in these handbooks revolve. Despite its roots in philosophy, classical rhetorical tradition and stylistics, pragmatics is a relatively recent discipline within linguistics. C.S. Peirce and C. Morris introduced pragmatics into semiotics early in the twentieth century. But it was not until the late 1960s and early 1970s that linguists took note of the term and began referring to performance phenomena and, subsequently, to ideas developed and advanced by Wittgenstein, Ryle, Austin and other ordinary language philosophers. Since the ensuing pragmatic turn, pragmatics has developed more rapidly and diversely than any other linguistic discipline. The series is characertised by two general objectives. Firstly, it sets out to reflect the field by presenting in-depth articles covering the central and multifarious theories and methodological approaches as well as core concepts and topics characteristic of pragmatics as the analysis of language use in social contexts. All articles are both state of the art reviews and critical evaluations of their topic in the light of recent developments. Secondly, while we accept its extraordinary complexity and diversity (which we consider a decided asset), we suggest a definite structure, which gives coherence to the entire field of pragmatics and provides
vi
Wolfram Bublitz, Andreas H. Jucker and Klaus P. Schneider
orientation to the user of these handbooks. The series specifically pursues the following aims: – it operates with a wide conception of pragmatics, dealing with approaches that are traditional and contemporary, linguistic and philosophical, social and cultural, text- and context-based, as well as diachronic and synchronic; – it views pragmatics from both theoretical and applied perspectives; – it reflects the state of the art in a comprehensive and coherent way, providing a systematic overview of past, present and possible future developments; – it describes theoretical paradigms, methodological accounts and a large number and variety of topical areas comprehensively yet concisely; – it is organised in a principled fashion reflecting our understanding of the structure of the field, with entries appearing in conceptually related groups; – it serves as a comprehensive, reliable, authoritative guide to the central issues in pragmatics; – it is internationally oriented, meeting the needs of the international pragmatic community; – it is interdisciplinary, including pragmatically relevant entries from adjacent fields such as philosophy, anthropology and sociology, neuroscience and psychology, semantics, grammar and discourse analysis; – it provides reliable orientational overviews useful both to students and more advanced scholars and teachers. The nine volumes are arranged according to the following principles. The first three volumes are dedicated to the foundations of pragmatics with a focus on micro and macro units: Foundations must be at the beginning (volume 1), followed by the core concepts in pragmatics, speech actions (micro level in volume 2) and discourse (macro level in volume 3). The following three volumes provide cognitive (volume 4), societal (volume 5) and interactional (volume 6) perspectives. The remaining three volumes discuss variability from a cultural and contrastive (volume 7), a diachronic (volume 8) and a medial perspective (volume 9): 1. Foundations of pragmatics Wolfram Bublitz and Neal R. Norrick 2. Pragmatics of speech actions Marina Sbisà and Ken Turner 3. Pragmatics of discourse Klaus P. Schneider and Anne Barron 4. Cognitive pragmatics Hans-Jörg Schmid 5. Pragmatics of society Gisle Andersen and Karin Aijmer
Preface to the handbook series
6. Interpersonal pragmatics Miriam A. Locher and Sage L. Graham 7. Pragmatics across languages and cultures Anna Trosborg 8. Historical pragmatics Andreas H. Jucker and Irma Taavitsainen 9. Pragmatics of computer-mediated communication Susan Herring, Dieter Stein and Tuija Virtanen
vii
Table of contents
Preface to the handbook series . . . . . . . . . . . . . . . . . . . . . . . .
v
Introducing the pragmatics of society Karin Aijmer and Gisle Andersen . . . . . . . . . . . . . . . . . . . . . .
1
I.
Social, regional and situational factors
1.
Doing age and ageing: language, discourse and social interaction Alexandra Georgakopoulou and Anna Charalambidou . . . . . . . .
31
Gender identities and discourse Bróna Murphy . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
53
Regional pragmatic variation María Elena Placencia . . . . . . . . . . . . . . . . . . . . . . . . .
79
2.
3.
4.
Pragmatics in multilingual language situations Susanne Mühleisen . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
II.
The language system and pragmalinguistic features
5.
Speech and writing: linguistic styles enabled by the technology of literacy Douglas Biber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
6.
Phonetics and the management of talk-in-interaction Gareth Walker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
7.
Prosody and pragmatic effects Anne Wichmann . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
x
Table of contents
III. Pragmatic markers and the notion of speaker attitude 8.
9.
10.
Pragmatic markers in a sociopragmatic perspective Ad Foolen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
217
Interjections Neal R. Norrick . . . . . . . . . . . . . . . . . . . . . . . . . . . .
243
Vagueness and hedging Maryann Overstreet . . . . . . . . . . . . . . . . . . . . . . . . . .
293
IV.
Different interpretational levels – speech acts, politeness and beyond
11.
Requests and orders: a cross-linguistic study of their linguistic construction and interactional organization Carmen Taleghani-Nikazm . . . . . . . . . . . . . . . . . . . . . .
321
Appreciatory sounds and expressions of embodied pleasure used as compliments Andrea Golato . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
361
Politeness and impoliteness Jonathan Culpeper . . . . . . . . . . . . . . . . . . . . . . . . . . .
393
Honorifics and address terms Sachiko Ide and Kishiko Ueno . . . . . . . . . . . . . . . . . . . . .
439
12.
13.
14.
V.
Sequential patterns and activities
15.
Social and pragmatic variation in the sequential organization of talk Jan Svennevig and Ronny Johansen . . . . . . . . . . . . . . . . . .
473
Turn-taking in conversation Jakob Steensig . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
501
Pauses and hesitations Anna-Brita Stenström . . . . . . . . . . . . . . . . . . . . . . . . .
537
16.
17.
Table of contents
xi
VI.
Pragmatics and the notion of culture
18.
Cultural variation in language use Anna Gladkova . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 571
19.
Intercultural rhetoric and language of healthcare Ulla Connor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 593
VII. Pragmatics and the larger societal context 20.
Global and intercultural communication Juliane House . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607
21.
Critical discourse analysis: overview, challenges, and perspectives Ruth Wodak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 627
22.
Pragmatics, linguistic anthropology and history Leila Monaghan . . . . . . . . . . . . . . . . . . . . . . . . . . . . 651
About the authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 689 Author index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695 Subject index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 701
Introducing the pragmatics of society1 Karin Aijmer and Gisle Andersen
Introduction We endorse the view of pragmatics as “the cognitive, social and cultural science of language and communication” (Verschueren 2009: 1). Within this perspective, we refer to the study of language and communication in its social and cultural context as sociopragmatics. Sociopragmatics in a broad sense aims to show how social and cultural factors are brought to bear in language practices, and how they influence pragmatic strategies which are manifested by linguistic forms in particular communicative contexts. Some of the basic notions of pragmatics are speech acts, pragmatic markers and speaker attitude, relevance and implicature, discourse structure and coherence, presupposition and deixis, etc. Sociopragmatics does not approach these pragmatic phenomena as theoretical constructs or as cognitive phenomena per se, but aims to account for their instantiations in empirical socio-cultural contexts and to present cultural, social and situational differences in their manifestation. Among the concerns of this field is how linguistic forms can convey social or cultural meanings because of their close association with particular situations or situational dimensions. Although the roots of sociopragmatics go further back (e.g. Lakoff 1975; House and Kasper 1981), the initial use of the term ‘sociopragmatic(s)’ stems from Thomas (1981, 1983) and Leech (1983), who, in fact, seem to attribute the term to each other. Thomas (1981, 1983) introduces the dichotomy between two different types of ‘pragmatic failure’, that is, observable breaches in pragmatic competence as seen for example in second language learners. Pragmalinguistic failure is the result of systematic pragmatic differences between the source and target language, “occurring when the pragmatic force mapped onto a given linguistic token is different from that normally assigned to it by native speakers” (Thomas 1981: 35). Pertinent examples would be when a non-native speaker fails to recognise the pragmatic force of an utterance such as the (in)appropriateness of the expression good day, or fails to recognise the indirectness of a certain speech act in the target language, such as It’s cold in here meant as a request to close the window. Sociopragmatic failure, on the other hand, is due to culturally different judgements about what counts as imposition, about power and social distance, or about the relative rights and obligations of speaker and hearer. Thus, while pragmalinguistics concerns the appropriate linguistic means used for thanking, congratulating, requesting, etc., sociopragmatics concerns the appropriateness of these speech acts in different situations. Although this is certainly a useful distinction, we do not see it as
2
Karin Aijmer and Gisle Andersen
constitutive of the research field of sociopragmatics, which we perceive to be wide enough to include both concerns. In fact, we wish to maintain that any type of breach in pragmatic competence that can be explained by social, cultural or situational factors would belong to this research field. This is in line with the ‘Continental’ approach to pragmatics represented for example by the Journal of Pragmatics, the proceedings of the International Pragmatics Assocociation (IPrA) conferences, and Jacob Mey’s (1993) social theory of pragmatics. Sociopragmatics in this wide sense is distinct from theories of pragmatics based in philosophical logic or cognition, such as relevance theory (Sperber and Wilson 1986/1995), which focuses on the cognitive processes involved in the interpretation of utterances, in strict isolation from their social context (dealt extensively with in Volume 4 of the Handbooks of Pragmatics series; cf. Schmid forthcoming). Notwithstanding the fact that cognitive or psychological realities have no place in Leech’s model, we find his distinction between ‘general pragmatics’ and sociopragmatics useful; in fact we are concerned precisely with the “specific ‘local’ conditions on language usage” (Leech 1983: 10) that are excluded from general pragmatics but captured by his sociopragmatics notion, and with pragmalinguistics, “the more linguistic end of pragmatics – where we consider the particular resources which a given language provides for a particular illocution” (Leech 1983: 11). In more recent years, sociopragmatics has developed primarily within the confines of historical pragmatics; a particularly notable contribution is the special issue of the Journal of Historical Pragmatics (2009, 10.2) and its introduction by Culpeper (2009), who, in line with our approach, stresses that “sociopragmatics is not simply concerned with mapping regular patterns of usage in interaction, as might characterize much work in sociolinguistics, but with understanding how those regular patterns are used and exploited in particular interactions” (Culpeper 2009: 180f). Nevertheless, Culpeper (2010) rightly acknowledges that “sociopragmatics is not a wellrecognised and agreed research area even within synchronic research” (2010: 69 in Volume 8 of the Handbooks of Pragmatics series; cf. Jucker and Taavitsainen 2010). Our focus in this volume is exclusively synchronic. The subfield of sociopragmatics includes different strands of research which will be comprehensively explored in this chapter. As we see it, sociopragmatics encompasses interactional sociolinguistics (Gumperz 1982; Duranti and Goodwin 1992), linguistic anthropology (Hymes 1964; Duranti 2009) and variational pragmatics (Barron and Schneider 2005; Schneider and Barron 2008), and indeed several of the contributions in this volume take a sociolinguistic perspective to pragmatics, accounting for a number of social and situational dimensions along which pragmatic language features vary. Such studies focus on variability according to well-known macro-sociological variables such as age, gender, culture and regional provenance. However, the scope of sociopragmatics is wider, and the subfield also encompasses studies which take a stricter pragmalinguistic perspective. This research focuses
Introducing the pragmatics of society
3
on specific pragmatic phenomena that are intrinsically sociopragmatic, a cardinal example being honorific expressions (Ide and Ueno, this volume), which cannot be accounted for without reference to the social roles of speaker and hearer. This phenomenon can be described in terms of its formal and functional manifestations and social significance. Moreover, sociopragmatics includes conversation analytical studies which take an ethnomethodological perspective. These studies seek explanations for variation by scrutinising the activity engaged in and the methods used by the participants to manage problems in the conversation, as well as the identities displayed as relevant in interaction. Conversational activities are often described in terms of the sequential organisation of talk rather than the external social characteristics of its participants. Finally, the subfield encompasses studies which take a societal perspective to pragmatics, investigating pragmatics with a more general socio-cultural focus that includes cross-linguistic comparisons of communicative behaviour, critical discourse analysis (CDA) and global and intercultural aspects of communication, such as multilingualism and lingua franca usage. The contributors to this volume are thus pragmaticians working on socio-cultural aspects of discourse and communication, as well as sociolinguists working on pragmatic and discourse aspects, applying a variety of complementary qualitative and quantitative methods and drawing from research traditions such as variational sociolinguistics, interactional phonetics and sociophonetics, corpus linguistics, conversation analysis (CA) and (critical) discourse analysis. The study of pragmatics as the relationship between language and the social and cultural context goes hand in hand with a revival of interest in linguistic relativity, the idea that ‘culture through language, affects the way we think, especially perhaps our classification of the experienced world’ (Gumperz and Levinson 1996: 3). This idea, which can be traced back to Humboldt, Sapir and Whorf, was difficult to test empirically and was discredited by the emergence of the cognitive sciences in the 1960s. This resulted in a shift of research focus to the universality of language processes and to the innate human capacity for language acquisition (e.g. Chomsky 1965). However the hypothesis of language universality came to be questioned when a number of empirical studies by anthropologists showed that there were variations in language use which could only be explained by the cultural and social context. More recently it has been shown that cultural and social factors affect the use of language on different levels and in particular that contextual factors are associated with language use. The idea that the meaning, structure and use of language are socially and culturally relative is basic to interactional sociolinguistics, a hybrid discipline with a background in anthropology, sociolinguistics and pragmatics (see e.g. Gumperz 1982). What is needed is a “general theory of verbal communication which integrates what we know about grammar, culture and interactive conventions into a single overall framework of concepts and analytical procedures” (Gumperz 1982: 4; quoted from Schiffrin 1994: 99). Pragmatics, sociopragmatics, and indeed the current volume can be seen as contributions towards
4
Karin Aijmer and Gisle Andersen
this aim. In the following we describe different strands of sociopragmatic research and how the individual chapters of this book contribute towards this aim. The sections that follow correspond to the distinct parts of the book and to the thematic groupings of the individual chapters.
Social, regional and situational factors The influence of sociolinguistics is seen from the focus on how socio-economic class, gender, age and ethnicity can explain linguistic variation. These categories have been shown in variational sociolinguistics to have a systematic impact on grammar, vocabulary and pronunciation. However, sociolinguists have paid less attention to discoursal and pragmatic features associated with language in interaction. As a result of the growth of interest in pragmatic issues in general linguistics, there has also been more attention given to how pragmatic differences between languages can be explained by the impact of social factors. The budding field of variational pragmatics (Barron and Schneider 2005; Schneider and Barron 2008) studies the influence of region on pragmatic phenomena at several levels. Not only dialects but pluricentric languages with several standards such as English, Spanish, German or French have subvarieties which may be distinguished by differences in usage. For example native speakers of Irish English may respond differently to thanks from British or American speakers (Schneider 2005). García’s study (2008) of invitations in Peruvian and Venezuelan Spanish shows different intralingual varieties of Spanish use, and Muhr (1994) noticed pragmatic differences in apologising between Austrian Standard German and German Standard. Another situation is when language is used differently by speakers in post-colonial societies (cf. Barron and Schneider 2009: 425 on ‘post-colonial pragmatics’). For example, in the Caribbean, pragmatic variation is due to the presence of both European and African traditions in the same society. The notion of (social) macro-category continues to play an important role as shown in several contributions to this volume. However, the focus has shifted from the major demographic categories to how speakers use language to signal who they are and who they affiliate with through group membership. From a sociolinguistic perspective speakers are members of social networks and peer groups sharing much of their socio-cultural knowledge. The notion of communities of practice allows for a fine-tuned view of a peer group as an aggregate of people who come together around a shared purpose and who develop a shared understanding and a repertoire of semiotic resources through regular interaction. The contribution by Alexandra Georgakopoulou and Anna Charalambidou explores this notion through a focus on age as a distinguishing factor in discourse and social interaction. Their study is innovative in that it brings together two strands of research which have developed in considerable isolation, namely research on youth lan-
Introducing the pragmatics of society
5
guage and research on ageing. It documents a shift in focus from normative accounts, in which the language of young or old speakers is seen as deviations from an adult norm – or even as ‘deficient’ – to more practice-based accounts which show how age-related identities are negotiated in discourse. They propose that dynamic view of language use and the context can be combined with an interest in social variation and suggest that we need an activity-based approach to language and age research which also considers the sequential organization of the communication. This interest in activities and settings in sociolinguistic studies of age and gender is to be welcomed and points out one direction in which sociopragmatics is moving. Further, the chapter by Bróna Murphy deals with the role of gender as an identity marker and explores how males and females use and interpret linguistic resources in interaction. She gives an overview of an unbroken tradition of language and gender research over four decades, pointing out significant shifts or ‘waves’ in the research, characterised as the deficit, dominance, difference and dynamic approaches. The account shows how language and gender research has gradually moved away from a reliance on binary oppositions and global statements about the behaviour of men and women, to more nuanced and mitigated statements about certain groups of women or men in particular circumstances. Murphy points out the clear bias in research on gender and pragmatics towards studies of female language use. Her own research contributes to remedying this by focusing on abusives (swear words) in the context of casual conversation between male friends. Her work looks beyond gender and compares speakers belonging to different age groups in a corpus of Irish English. The results show differences between early, middle and late adulthood in terms of abusive forms and their functional properties, reflecting how men in different life stages construct distinct gender identities, e.g. as fathers, professionals and mature men. Abusives are clearly most commonly used with social functions linked to camaraderie, social bonding, humour etc., and are rarely intended as insults. In contrast with most previous literature in corpusbased pragmatics, Murphy supplements her corpus-based findings with observations made in interviews with the same speakers. María Elena Placencia is concerned with regional differences which are explored from a pragmatic variational perspective. She sets out to define and characterise the notion of regional pragmatic variation as a subfield of variational pragmatics in the tradition of Schneider and Barron (2008) and provides an overview of earlier work on a wide range of pragmatic phenomena, including strategies for face-threat mitigation, intensification, discourse organisation, opening and closing, address forms, discourse markers, laughter, etc. These phenomena are related to the analytical domain which they belong to, specifically the illocutionary, discourse, participation, stylistic and non-verbal domains (Spencer-Oatey 2000), and they are explored with reference to a variety of languages and national varieties, with a main focus on Spanish, English and German.
6
Karin Aijmer and Gisle Andersen
The final chapter in this part of the book, by Susanne Mühleisen, also deals with region as a distinguishing factor, accounting for pragmatic verbal behaviour in a number of language situations that are characterised by multilingualism and the existence of parallel or competing norms of communication. She explicitly challenges the implicit equation “one language = one culture = one community”, warning that pragmatic studies are in danger of relying on assumed universality in patterns of verbal behaviour (e.g. Brown and Levinson 1987). Even cross-cultural studies do not necessarily do justice to many language situations where more than one language is used by all members of the community or by part of the group, but may actually contribute to a production or reproduction of national stereotypes. She reviews the pragmatics literature on European multilingual countries (Switzerland, Belgium and Luxemburg), differences between Pakeha and Maori speakers in New Zealand and pragmatic variation in the Anglophone Caribbean. Her work shows, for instance, that in the complex diglossic Creole/English language situation of Caribbean countries, pragmatic variation does not correspond straightforwardly with certain linguistic or ethnic groups but is part of a complex negotiation of one‘s alignment with Creole versus English/European values and norms of communication.
The language system and pragmalinguistic features The effects of pragmatic differences are observable at all levels of language. For example, honorifics in Japanese are indexically associated with social identity (Ide and Ueno, this volume). Other well-attested examples are pragmatic markers, conversational routines, interjections and taboo words, turn-taking and address forms. But pragmatic effects can also be witnessed in systemic language phenomena such as syntactic complexity and morphological variation, vocabulary, and a range of phonetic and prosodic phenomena, including phonatory, articulatory, intonational and rhythmic details. Several of the contributions to this volume are specifically concerned with how the communicative context systematically constrains linguistic output. Douglas Biber investigates variation in language use according to the mode of communication by comparing a variety of spoken and written modes. He presents a synthesis of earlier work on a wide range of grammatical characteristics in seven academic and non-academic registers, including clausal structure, vocabulary, nominalisation, use of adjectives and adverbs, etc. His work shows that in spoken registers, communicative purpose is actually secondary to mode; whether a speech event is interactive and interpersonal (e.g. conversation) or monologic and informational (e.g. classroom teaching), it is characterised by the same set of linguistic features. Thus, classroom teaching has more in common with other spoken modes than with textbooks, although these two pedagogical modes share the same
Introducing the pragmatics of society
7
informational communicative purpose. In writing, on the other hand, there is much greater variability and communicative purpose and audience have a major influence on grammatical features. In fact, popular written registers with non-informational purposes share many of the characteristics of speech. This overall difference between speech and writing can be attributed to differences in production circumstances; for example, the dense use of complex noun phrase structures which is typical of some kinds of written (e.g. academic) prose is simply not feasible in speech due to the production constraints or the spoken mode. While Biber deals with grammatical and lexical language features, the contributions by Walker and Wichmann consider phonetic and prosodic aspects. Gareth Walker’s contribution is placed within interactional phonetics (Local and Walker 2005), a distinct analytical approach that has adapted its methodological principles from conversation analysis and which studies the social-interactional import of phonetic variability. This distinguishes it from sociophonetics (Foulkes and Docherty 2006), which studies phonetic variation which can be attributed to the social macro-categories such as age, gender etc., discussed above. Interactional phonetics shows how phonetic features or clusters thereof are not simply present in the interaction, but how they are treated as significant – or ‘oriented to’ – by the participants themselves. Equipped with CA’s analytical toolbox of adjacency pairs, turn-constructional units and transition relevance points, Walker inspects how the phonetic design of utterances – including pitch, loudness, articulation rate, cutoffs, silences and creak phonation – has bearings on the management of turntaking, the marking of relationships between and within turns, the marking of competitive vs. non-competitive overlaps, increments, self-interrupting talk, etc. Among the literature that he reviews, we find work that shows that the phonetic details which mark transition relevance differ across varieties of English, and that phonation and glottal stops may be significant in the management of turn-taking, as in Finnish, where creak phonation is associated with turn-yielding and glottal stops with turn-holding (Ogden 2001). In comparison with Walker’s approach, Anne Wichmann is specifically concerned with one aspect of prosody, namely intonation, and she deals more explicitly with sociopragmatics in the variational sense and with the sociopragmatic meanings generated by paradigmatic speaker choices in prosody. This includes work which shows how prosody reflects and constructs power and social distance. She outlines the prosodic resources available to speakers and conveys how the components of prosody can be linked with the expression of paralinguistic meaning, such as aggression, solidarity, affection and stance. Further, she points at gender differences and speaking styles such as the so-called ‘uptalk’ (also known as the high-rising terminal) used in some varieties of English. She explores both the semiotic value of pitch and the sound symbolism associated with pitch differences; e.g. of low pitch as more dominant and assertive while high pitch is submissive or lacking in confidence. Wichmann’s own research looks into how the
8
Karin Aijmer and Gisle Andersen
relative rights and obligations of participants are reflected in requests containing the marker please, documenting how prosody can be used to express the speaker’s stance by means of accommodating or not to the prosody of an interlocutor.
Pragmatic markers and the notion of speaker attitude Sociopragmatics is also concerned with how speakers modify the propositional content of utterances or the speech act information by various means including hedging, interjections and markers. Typically, speakers convey speaker attitude and other types of higher-level information which guide the hearer in the interpretation process. Pragmatic markers (discourse markers) present an interesting in case in terms of sociopragmatic variation and have been a fertile area of research the past few decades. They have been analysed in the tradition of sociolinguistic variation analysis for example in the work by Sali Tagliamonte (see e.g. Tagliamonte 2005). Tagliamonte studied the pragmatic markers just, so and like on the basis of a quantitative analysis of a corpus of the spoken language of Canadian adolescents. The findings suggested that at least just and so are innovative features in youth language on their way to becoming part of the general language. Similarly, Andersen (2001) showed that there are observable age and gender differences in the use of like as a pragmatic marker. Several studies have compared the use of pragmatic markers by men and women. Janet Holmes found interesting contrasts between men and women in New Zealand in the use of pragmatic markers such as you know (Holmes 1986) and sort of (Holmes 1988). The choice of marker can be associated with the needs of speakers belonging to a particular social class. Huspek (1989) conducted a study of the selection of you know or I think in American industrial workers’ speech which showed that you know was more frequent than I think in worker speech. In many cases the high frequency of you know could be associated with powerlessness and the rejection of group solidarity. However it could also help or embolden the worker to be ‘positively solidary and open to criticism and thought’ (Huspek 1989: 682). In comparison with the large number of studies of speech acts in variational pragmatics there are comparatively few studies of pragmatic markers in different varieties of languages. However we can expect there to be differences across varieties of pluricentric languages like English or Spanish. Sure is for example used more frequently as a response to offers and requests and as a feedback item in American than in British English (Aijmer 2009). Cuenca (2008) found differences between the Spanish and Catalan correspondences of well in translations from the film Four Weddings and a Funeral. Pragmatic markers can often be associated with the social identities of speakers (for instance as members of a peer group or in social roles as doctor or teacher), social relationships (friends, acquaintances), activities (debating, interviewing), so-
Introducing the pragmatics of society
9
cial acts (requests, offers), attitudes and feelings (see Ochs 1996 on the parameters affecting sociolinguistic meaning). Moreover, pragmatic markers are multifunctional and some functions may be associated with a particular activity (text type). De Fina (1997) has for instance found that Spanish bien (overlapping with well) was used by the teacher in the classroom for functions associated with the teacher role such as marking a new stage in the classroom lesson. Bien functions as a cue to the hearer to interpret the turn (or turn sequence) in terms of common or shared knowledge about the activity (cf. the discussion of pragmatic markers as contextualization cues by Foolen, this volume). In addition to sociolinguistic differences, there are other, principally different ways of looking at pragmatic markers from a variationist perspective. One may take the individual form as a starting point and study its functional variability by surveying its range of functions in a discourse variety, such as Stenström’s (1986, 2002) accounts of really in British English. On the other hand, one may take a specific pragmatic function as a starting point and study the range of forms that map onto this particular function. This is done, for instance by Stenström, Andersen and Hasund (2002) in their account of formal variation in the marking of reported speech by either the quotative verbs say or go, the grammaticalised BE + like or zero-quotations and mimickry. As Romero-Trillo (2006) observes, functions are more stable than forms, and variability in the use of pragmatic markers can be construed as “discourse slots” which are “filled by elements that may vary according to regional, idiolectal, or sociolinguistic features within one and the same language” (Romero-Trillo 2006: 640). Ad Foolen takes a sociopragmatic perspective on pragmatic markers and gives an overview of various theoretical perspectives, including relevance theory, which continues to be a productive framework for the study or pragmatic markers (e.g. Andersen and Fretheim 2000; Blakemore 2002). Acknowledging that different theories highlight different aspects of the functioning of pragmatic markers, Foolen devotes considerable attention to the sociolinguistic notion of contextualisation cues, a signal which “serves to construct the contextual ground for situated interpretation and thereby affects how constituent messages are understood” (Gumperz 2001: 221). Alongside non-verbal cues such as gesture, posture and facial expressions, pragmatic markers are among the metapragmatic signs which due to their implicit nature are distinct from grammatical and lexical signs. In accordance with the claims concerning pragmatic markers made by relevance theorists, Foolen stresses that contextualisation cues have context-evoking capabilities; thus that the relevant context for the interpretation of an utterance is not fixed in advance but depends on the specifics of the utterance and on the contextualisation cues themselves. He devotes considerable attention to variation, referring to studies which represent a variety of languages and showing that the use of pragmatic markers varies according to regional and national variety, age, gender, cultural background and spoken and written register. His account adds an applied perspective in that it
10
Karin Aijmer and Gisle Andersen
also covers the use of pragmatic markers in foreign language teaching and translation, as well as in computer-mediated communication. Another salient way of expressing speaker attitude is by means of interjections, which is the topic of Neal Norrick’s contribution. Basing his account on Ameka’s (1992) distinction between primary and secondary interjections depending on whether or not the forms are used with other functions, he shows that interjections are an indefinitely extendable class of items, unlike the relatively circumscribed and closed class of other pragmatic markers. A very general observation is that the functions of interjections seem to follow from their status as expressions of shifts in cognitive states of various kinds. Norrick reports sociopragmatic differences in the use of interjections, with a special focus on regional varieties of English, such as the fact that yeah is more common in American English than in British English, and that it is more common among younger speakers in both varieties. He also notes distributional differences and sociolinguistically constrained uses of interjections in British, Australian and New Zealand varieties. A crucial component of Norrick’s contribution is his account of collocations of interjections, such as oh no/ yes/well and phrasal interjections like I tell you. Although interjections are generally free-standing elements, they are not necessarily independent of other constructions when integrated into a single intonation unit with another word or clause, as with hell yes or shit no. Contrary to what is often claimed about interjections, many primary interjections do not express emotions but information states, but when they occur together, it is often up to the secondary interjections, with their common roots in religion, sex or scatology, to provide the emotional content. Finally, Maryann Overstreet considers hedging and vague language, yet another prolific theme in the sociopragmatics literature. She argues in favour of an approach to vagueness and hedging based on a broader sociopragmatic analysis rather than a more limited propositional analysis, as is applied especially in explorations of these phenomena in philosophy. Vague language includes a variety of structures commonly occurring in everyday speech: from inherently vague expressions like placeholders (thing, stuff), to expressions that are vague by implicature (Channell 1994). The previous literature documents differences between standard and non-standard varieties of English, as well as differences between adult and teenage speakers. However, contrary to expectation, it has been shown that the differences concern the inventory of forms more than overall frequency of vague expressions (Stenström, Andersen and Hasund 2002; Cheshire 2007); where adults have a preference for and so forth and other formal markers, teenagers prefer and stuff like that as well as other highly informal expressions. There is also by now a rich literature that reports on the use of like as pragmatic (vagueness) marker, showing that this is clearly an age-specific phenomenon and an effect of influence of American English on other varieties (Andersen 2001). Moreover, social class can be seen to have an effect on vague language, as certain expressions (and stuff/ things) are more indicative of middle class speech while others are indicative of
Introducing the pragmatics of society
11
working class speech (and that; Cheshire 2007). Another perspective is differences between native and non-native usage, a case in point being the discrepancy between the English/German formally equivalent vague expressions or so, as in sixty or so pubs, vs. oder so, which has a wider functional range. Hence, the markers are not necessarily functionally equivalent but may be false friends used inappropriately by EFL speakers with German language backgrounds (Terraschke and Holmes 2007).
Different interpretational levels – speech acts, politeness and beyond The study of speech acts started out as a philosophical concern spearheaded by philosophers such as Austin and Searle in the 1960s. Austin opened the way by showing ‘how we can do things with words’ (cf. Austin 1962) and Searle (1969) formulated conditions or rules for the appropriate use of various speech acts such as promises or requests which he claimed to be universal. However Searle’s analysis of speech acts is based on isolated examples and no attention was given to the embeddedness of speech acts in different social contexts. Brown and Levinson’s (1978) face-based theory of politeness sparked interest in studying how speakers use certain general politeness strategies oriented to saving their own or the hearer’s ‘face’ for example when asking someone to do something. The relationship between politeness and indirectness which was a cornerstone for speech act research based on Brown and Levinson’s theories has been shown to be problematic because of its culture-specificity (e.g. Ide 1989; Eelen 2001). Jointly, speech act theory and politeness theory have been catalysts for the emergent field of emancipatory pragmatics, which acknowledges that “pragmatics as an analytic enterprise has been dominated by views of language derived from Euro-American languages and ways of speaking” (Hanks, Ide and Katagiri 2009: 1) and therefore aims to free the analysis “from the confines of theoretical orthodoxies grounded in dominant thought and practice” (Hanks, Ide and Katagiri 2009: 2). However, Brown and Levinson’s ideas have influenced a number of cross-cultural studies of speech acts. The best known of the early cross-cultural studies testing Brown and Levinson’s ideas cross-culturally is probably the influential CCSARP project (Cross-Cultural Speech Act Realization Project), which had the aim to compare the (politeness) strategies used by speakers of different nationalities (Danish, German, Hebrew, American English) for requesting and apologising (Blum-Kulka et al. 1989). A number of interesting findings have come out of the project and subsequent studies of speech acts using the same methodology. The cross-cultural differences in the use of strategies for requesting and apologising were shown to involve the use of (in)directness and attitudinal modifications of the speech act. Blum-Kulka (1989: 58) has for instance argued that Hebrew speakers would use a more direct formula (corresponding to perhaps and future) to ask for information where English speak-
12
Karin Aijmer and Gisle Andersen
ers would use an indirect form such as Can you tell me …. However there is no general consensus about the categories needed to analyse speech acts and to compare them across different cultures. Wierzbicka (2003[1991]) regards speech act labels such as ‘compliment’ or ‘warning’ as theoretically unjustified ‘folk categories’ and proposes instead a ‘natural semantic metalanguage’ in which meanings can be encoded and compared. The semantic metalanguage which contains components such as ‘say’ or ‘think’ makes it possible to show both similarities and differences between speech acts (cf. Gladkova this volume, described below). (See also Volume 6 in the Handbooks of Pragmatics series on Interpersonal Pragmatics; cf. Locher and Graham 2010). A number of different approaches have been attempted to study speech acts empirically. The study of speech acts in the CCSARP project was based on socalled discourse completion tasks (DCTs). In such a task the subjects are presented with a number of situations in which they are expected to make a request or to apologize. They are then asked to write down what they would say. The use of DCTs makes it possible for the researcher to control for certain situational variables such as power, social distance and ranking of imposition (when analysing requests) and to collect large quantities of data from many languages. The project has inspired research on many different language pairs and on other speech acts (e.g. leave-taking and invitation refusals). The methodology has however been criticized because it produces data which differs from authentic spoken language (Golato 2005). As Golato points out, “DCTs are better suited to the study of ‘what people think they would say’ than to the study of ‘what people actually do say’ in a given situation” (2005: 14, author’s emphasis). If one is interested in what people actually say and how they use speech acts in the wider context of the discourse it follows that we should use naturally occurring data. A number of scholars have used CA in order to study how requests are used in their sequential environment based on audio or video recordings of naturally occurring talk-in-interaction (e.g. Golato 2005). CA analyses speech acts as actions positioned as first and second parts in their sequential environment. The perspective is that of the speakers themselves who design their talk so that they can display an understanding of previous talk. For example a factual response will show an understanding of the previous turn as a question. In this approach the patterns of requesting are generally not associated with social factors such as age or gender but related to the speakers’ understanding of the social situation and the interactional environment in which the speech act is performed. If there are other factors affecting the request, they would be beyond what is usually studied in CA. Although based on (often elaborate) transcriptions of authentic data, this approach does not explore the potential of quantitative methods. On the other hand, the use of quantitative data and frequencies is more closely associated with corpus linguistics. In corpus-linguistic studies based on large amounts of spoken language it is possible to make generalisations about the rela-
Introducing the pragmatics of society
13
tionship between speech acts and the context in which a speech act occurs. Adolphs (2008) for instance has studied the relationship between suggestions, function and contextual factors such as speaker role and situation on the basis of the large CANCODE Corpus. In this volume, the contributions by Taleghani-Nikazm and Golato focus on the realisation of specific types of speech acts, namely requests/orders and compliments, respectively. With a theoretical basis in CA, Carmen Taleghani-Nikazm studies the design of requests, focusing mostly on how they are realised by imperative, interrogative or declarative syntax. She argues that the design is tightly connected to speakers’ interpretation of the interactional environment in which the request occurs, that is, the contingencies of the speech situation, while social factors such as age, gender and social status are not necessarily attended to. She reports several studies which show how design is governed systematically by contextual factors; for instance in the choice between common request forms in English, would/could you vs. I wonder if, there are systematic distributional patterns such that the former is used in ordinary conversation while the latter occurs in institutional interactions. Moreover, the two forms differ in that, in wonder-prefaced requests speakers avoid displays of entitlement to the requested object or action, while in modal-prefaced requests, speakers display their assessment of their entitlement to the object of the request. Thus the design of requests is related to the speakers’ assessment of their entitlement to the requested object or activity and of the contingencies surrounding its granting (Curl and Drew 2008). Similar systematicity is observed in other languages and for other types of structural variation; for instance it is shown that non-granting responses to requests are shaped by the ways the requester (customer) produced the request originally. Taleghani-Nikazm takes this line of research further in an empirical study of Persian and German everyday conversation. This cross-linguistic study shows that the particular design of requests is closely connected to the sequential position in which they occur, and that the selection of the can I versus can you request format has equivalent effects in these two very different languages. Andrea Golato looks into compliments and compliment responses, both of which are important vehicles for communicating cultural norms and values. Her main focus is on German dyadic and multi-party conversation and cross-cultural comparison to American English. Compliments, or more precisely complimenting turns, can be accounted for in terms of their design, their interactional function and how the reference to the assessable (the element they make the compliment about) is expressed. Many compliment turns have no overt reference to the assessable, but are instead expressed through appreciatory sounds, stand-alone adjectives (lovely!), a combination of the two, or more elaborate clausal constructions, usually so-called verb-first constructions (Auer 1993) like schmeckt gut ‘tastes good’, in which the topic slot is empty. Previous studies of English and German have also shown that compliments are often formulaic in nature. Golato’s specific
14
Karin Aijmer and Gisle Andersen
focus is on the use of appreciatory sounds and expressions of embodied pleasure in compliment sequences. Representative of these categories in German are the forms oh and mmmh. The token mmmh is a gustatory marker and oh is a token that expresses an emotional change-of-state, in the sense of Heritage (1984). Golato points to previous work which shows differences between English and German oh, specifically that German oh is only used in instances where the change of state is of an emotional kind, such as joy, pleasure, physical pain, unhappiness, disgust etc. She also documents how mmmh is used in order to embody pleasure in the activity which speakers are engaged in, and how this usage is phonetically distinct from the response token mhm. It is clear from the study that compliment responses are shaped differently according to the function of the compliment, depending for instance on whether or not the compliment is produced in dispreferred first pair parts, such as a criticism. The current volume also accounts for politeness from different perspectives. Jonathan Culpeper reviews theories of politeness and impoliteness as well as socio-culturally based critiques. More specifically, he presents a detailed account of two waves in the literature on politeness, expressed through the dichotomy between the ‘classic’, pragmatic view to politeness, represented by Brown and Levinson (1978, 1987) and subsequent criticisms, and the socio-cultural view of politeness, represented by scholars who argued for the rejection or radical revision of the classic models on a variety of grounds, such as their presumed universality or their failure to treat politeness as a discursive phenomenon (e.g. Eelen 2001, Locher 2004). Citing corpus-linguistic work such as Aijmer (1996), Culpeper shows how the use of a word such as please, which ‘looms large in the British psyche’, is constrained by grammatical and situational differences and is characterised by differences between American and British usage. Acknowledging that social norms are naturally sensitive to context, Culpeper also points at social situations in which behaviours are not subject to politeness prescriptions, for example due to a gross power imbalance, associated for instance with army recruit training, as well as the sanctioned impoliteness of the UK’s House of Commons (cf. Harris 2001). Another notable part of this contribution is the focus on the relatively new research field of impoliteness. Culpeper problematises the relation between models of politeness and impoliteness, as well as presents new research on the latter. Specifically he reports a study of 100 impoliteness events reported by British undergraduates. He also presents a timely specification of what it means to be impolite, proposing a taxonomy for how impoliteness can be described in terms of attitudes that are (in order of predominance) patronising, inconsiderate, rude, aggressive, inappropriate or hurtful (Culpeper 2011). Finally, Sachiko Ide and Kishiko Ueno’s account of honorifics throws yet more light on the shortcomings of politeness theories which are based on Western cultures only, and deals with the neglected aspects of linguistic politeness from the perspective of Japanese. They argue that what is crucially needed is a description
Introducing the pragmatics of society
15
and explanation of what it means for native speakers to use honorific devices which either raise a referent, express a polite attitude towards an addressee or humble the speaker herself. The authors review the considerable literature on Japanese honorifics – most of which is virtually unknown to the Western world – and account for the practice of honorifics from the inside perspective of the context of speech situations. Japanese speakers must situate themselves in the speech context in ways which may be startling to westerners. Factors which come into play include the sociological nature of nominal referents in the utterance, the interpersonal relation between the nominal referents, as well as the formality of the setting. Honorifics are expressed linguistically by a variety of means, including the use of particles and the ‘honorification’ (morphological indexing) of nominal elements as polite address terms. Thus, first, second or third person referents (e.g. last names or social roles) may be marked with a variety of honorific pre- or suffixes to index the speaker’s perception of the other as an honourable professional or the like, and the system varies according to speaker’s gender. Moreover, the absence of specific items may also have an honorific effect; for instance the absence of a copula (desu or da) contributes to avoiding an overly strong assertion and expressing a congenial relationship between speaker and hearer. The different modes of expression used in for instance Japanese versus English must be considered a difference in the speakers’ perspective on speech events. At the core of Japanese pragmatics is the need to show the speaker’s discernment and sentiment about the contextual elements of the speech event. That is, the use of polite forms is not a matter of volitional choice; in fact, these forms are constitutive of the pragmatic well-formedness of utterances.
Sequential patterns and activities Sociopragmatics also concerns the structural and organisational level of discourse. Episodes, adjacency pairs, sequences, openings and closings of conversation are realised differently depending on factors such as speech community, medium of communication and activity type. Sinclair and Coulthard’s well-known study of classroom discourse (1975) shows for example that a typical organizational unit in the classroom consists of Initiative-Response-Feedback. The pupil responds to a conversational move initiated by the teacher which is a response followed by feedback from the teacher. The absence of feedback can also be interpreted in social terms. Sequences of Initiative-Response not followed by a feedback move from the interviewer in a news interview can be regarded as typical of journalistic speech. Both openings and closings in conversation lend themselves to sociopragmatic variation with regard to the age, gender and class of the speakers and are sensitive to the medium of communication, for example the distinction between informal conversation and chat interaction.
16
Karin Aijmer and Gisle Andersen
The importance of the medium of communication (and norms associated with the national culture) has been shown in several studies of openings in telephone conversation. Sifianou (1989) compared the sequential patterns in informal telephone conversations in American English and Greek. She showed for instance that the variations in Greek and English telephone behaviour are due to socio-cultural differences between the two communities and can lead to communicative misunderstandings. The functions of pauses are similar to those of pragmatic markers like well or you know discussed above, and pauses are used differently depending on who the speakers are and the situation. Speakers’ social background may have an impact on their use of pauses, and it has been shown that male and female speakers use pauses differently. The cultural stereotype of ‘the silent Finn’ is based on a view of Finns as comfortable with silence and less talkative than people of other nationalities (Stenström, this volume). (Related issues are also discussed in Volume 3 of the Handbooks of Pragmatics series; cf. Schneider and Barron forthcoming.) Jan Svennevig and Ronny Johansen study how sequential organisation varies according to several parameters including speech community, medium of communication and activity type. For instance, what counts as a relevant response to an opening sequence such as how are you? is crucially dependent on the sequential context it appears in. They discuss the value of conversation analysis as an approach to sociopragmatic variation, and their discussion is focused on some of the fundamental characteristics of CA. An important observation they make is that CA studies – often narrowly focussed on the particular details of highly specific datasets – seldom discuss explicitly the scope of their claims and the potential applicability to different situations or communities. Therefore it is not always clear whether the mechanisms described are meant to be understood as generic conversational rules or a feature associated with the specific activity engaged in or the socio-cultural identity of the speakers, and there may a risk of committing an ‘ethnocentric fallacy’ (Moerman 1988; Peräkylä 1997). Further they describe closing sequences of chat interaction in Norwegian, where participants can be seen to orient to gender as a feature of the context with recognisable consequences for the interaction. They point out clear qualitative and quantitative differences in displays of affection in same-gender or mixed-gender chat interaction, showing that among the teenage users, gender is a procedurally consequential aspect of context. The contribution by Jakob Steensig focuses on turn-taking in more general terms, presenting several complementary studies and perspectives. On the basis of findings from several languages Steensig, like the previous authors, problematises the universality of some of the basic claims in CA. He presents a comprehensive survey of a range of resources that are available to speakers in the construction and allocation of turns. This includes syntactic, morphological, prosodic and phonetic resources by which speakers may signal such features as turn completion and non-
Introducing the pragmatics of society
17
completion, speaker selection, self selection, speaker continuation and recursion. It also includes semantic and pragmatic resources, as well as gestures and gaze. Importantly, Steensig points out that turn-taking distribution works differently in different cultures, admitting though that the large body of evidence for this is ‘more or less anecdotal’. There is clear evidence to suggest that the cultural tolerance for very long silences varies considerably, but also that all languages seem to adhere to the principle of ‘minimal-gap, minimal-overlap’ (Sacks, Schegloff and Jefferson 1974; Stivers et al. 2009). However, more systematic empirical research is needed in order to establish that the basic principles of Sacks, Schegloff and Jefferson’s (1974) ‘simplest systematics’ are universal. While Steensig is concerned with the explicit verbal and non-verbal resources deployed in turn-taking, Anna-Brita Stenström considers the specific role of pauses and hesitations in discourse organisation. She provides an overview of the field of pausology, accounting for several perspectives including discourse and conversation analysis as well as psychology and anthropology. Pauses and hesitation devices are distinguished according to their formal characteristics, and Stenström’s taxonomy includes silent (unfilled) pauses, filled pauses, verbal fillers such as you know, sort of, like and well, lengthening, repetitions, repairs, reformulations and restarts. Functionally, pauses and hesitations can be distinguished as demarcating, hesitating or interacting, according to their structural or interactional characteristics. However, Stenström’s main focus is on the sociopragmatic characteristics of pauses and hesitations. She reports studies which reveal differences between languages and cultures with regard to the frequency and phonetic characteristics of filled pauses, as well as differences in tolerance for silence between turns and preference for fast talk and short pauses and turns. Moreover, she presents corpusbased findings which show how the use of pauses and hesitations in English varies according to age, gender and social group, in terms of nasalisation versus non-nasalisation of filled pauses (er vs. erm) and use of specific verbal fillers. She also points out differences in attitude towards such items and evaluates their role from a language learning perspective.
Pragmatics and the notion of culture Brown and Levinson (1987) showed the way in which sociolinguistics should be moving in their slogan “Sociolinguistics … ought to be applied pragmatics” (Brown and Levinson 1987: 6). Recent work in sociolinguistics and linguistic anthropology has shown that in order to account for the pragmatic meaning of linguistic phenomena in their social and cultural context, we need to consider the social interaction itself and the type of activity involved (e.g. whether the speaker is debating or joking), as well as speaker identity. As Gumperz and Levinson argue, “[t]he kind of contextual information that is actually needed turns out to be deeply
18
Karin Aijmer and Gisle Andersen
embedded in practices of speaking, the local conduct of social life, and the social distribution of shared understandings” (Gumperz and Levinson 1996: 8). We are thus witnessing a shift in many disciplines towards locating the social actors’ pragmatic choices in the activities they are engaged in. “The fields of pragmatics, linguistic anthropology, sociolinguistics, conversation analysis, and ethnomethodology all articulate ways in which the meaning of cultural forms, including language, is a function of how members engage these forms in the course of their social conduct” (Ochs 1996: 410). It is generally appreciated in these disciplines that language varies depending on factors such as participant roles and identities, setting, activities (e.g. courtroom proceedings or a parliamentary debate), social act (request, apology), power relations, as well as medium. For example in the analysis of speech acts one needs to go beyond traditional speech act theory and describe speech acts both cross-culturally and in different social situations. If languages have different social and cultural practices and norms this may result in misunderstandings. For example, in some cultures speakers show their modesty and regard for the addressee by using an indirect form to make a request. However, indirectness may be interpreted differently and regarded as inappropriate in the same situation in other cultures. In the current volume, Anna Gladkova explores cultural variation in language use from a specific methodological viewpoint. She considers a range of different cultural practices through the notion of cultural values expressed through cultural keywords (Wierzbicka 1997). Using the methodology of cultural scripts (Goddard and Wierzbicka 2004), Gladkova performs an analysis of how culture-specific linguistic practises and cultural values are embedded in culture-specific ways of speaking. Her work draws on research which explores notions such as truthfulness and sincerity in Russian culture, differences between Singapore English and other varieties of English in ways of expressing personal autonomy, as well as cursing in Yiddish as a culture-specific speech act that can be seen to have a therapeutic function. Ulla Connor’s contribution focuses on the use of language in health care settings and has a basis in intercultural rhetoric, a research area within second language acquisition that addresses the intersections of language, culture, and communication. An important influence on intercultural rhetoric has come from speech accommodation theory (e.g. Giles, Coupland and Coupland 1991). It is argued that the study of intercultural encounters requires a two-way model of communication, in which both parties – native speakers and non-native speakers – need to accommodate and negotiate to understand and be understood. Connor reviews research which suggests that better linguistic and cultural issues need to be considered in the planning of instruction for intercultural health care personnel, reporting specifically on a study of international medical graduates with African, Asian, Latin American or Middle Eastern backgrounds. Covering issues from eye contact and body language via humour and rapport to the management of sensitive issues or unfamiliar situations, the reported research shows that in addition to conventional
Introducing the pragmatics of society
19
language training, international medical graduates can benefit greatly from instruction in intercultural health care communication. See also the contribution by Juliane House, described below.
Pragmatics and the larger societal context Sociopragmatics can also be seen as a perspective on pragmatics from the point of view of society itself, that is, how the characteristics of societies and their members have bearings on language use at more local levels. We include the perspectives of global and intercultural communication, critical discourse analysis and linguistic anthropology under this societal approach. Rather than focus on isolated words and sentences, such chiefly interdisciplinary studies focus on larger discoursal units and their functions in communities and organisations. The topics investigated often concern the macro-effects of globalisation on individual languages and varieties, how language is used by more or less powerful groups, for instance to suppress minority groups in society, and the interplay between linguistic behaviour and non-verbal gestures, images, film, the internet and multimedia. Due to globalisation people of different nationalities increasingly come into contact with each other. Because of their different cultural backgrounds they will often have difficulties in communicating. Communicative misunderstandings are frequent when sociopragmatic conventions are not the same in the languages spoken. However there has been a recent shift from a concern with cultural differences and misunderstandings to describing successful interaction between speakers of different native languages with many cultural backgrounds. English can be used as a lingua franca (ELF) variety by speakers of different native languages. Varieties of ELF have in common that they use non-standard grammatical forms and have certain pronunciation features which are not found in the standard language. ELF can be studied as a variety in its own right, for instance according to the special style used by speakers when making a request or a compliment, or the strategic use of pragmatic markers for other functions and with different frequencies than those applied by native speakers. Juliane House accounts for global and intercultural communication from a general perspective, focusing on English as a lingua franca, but also presenting alternative means of global communication, specifically via translation and the European model of ‘intercomprehension’. She observes an emergent rethinking of the norms against which speaker behaviour in intercultural settings is to be matched. This norm should not be the monolingual and monocultural native speaker, since in a globalised community such speakers are the exception and not the norm. Rather, speakers in intercultural communicative settings ought to be viewed as expert bi-, tri- or multilingual speakers, which is usually a more correct picture. There is growing evidence to suggest that in the realisation of pragmatic
20
Karin Aijmer and Gisle Andersen
phenomena such as speech acts, bilingual speakers tend to opt for a decidedly “intercultural style” which lies in between and thus differs from both the languages they master. This tendency is not the result of lack of competence but due to a specific preference for a “third way” or a hybrid communicative style. The inherent variability of ELF, she argues, is not to be equated with the speakers’ failure to fulfil native norms but a result of creative exploitation of forms and functions, items and collocations available to them. This takes place in communities of practice that are governed by a joint purpose to communicate efficiently in English as the agreed and chosen language of communication. This is illustrated with reference to a range of language pairs and phenomena, including transfer, code switching and the re-interpretations of discourse markers among ELF speakers. Ruth Wodak describes Critical Discourse Analysis, which is an interdisciplinary research programme within ‘societal pragmatics’ combining different approaches with a shared interest in “the semiotic dimensions of power, injustice and political-economic, social or cultural change in society”. CDA does not study linguistic units per se but complex social phenomena that are the effects of different aspects of language use. For instance, an effect of the language of academic disciplines, with its common use of nominalisation and passivisation, may be to exclude outsiders and mystify the functions and intentions of the research. Thus, concepts such as power, ideology and critique are constitutive of most of the work in CDA. Wodak explores each of these concepts and describes how a notion such as “critical” is construed differently in Western cultures from the conceptualisations in Eastern cultures including Chinese. The notion of power is a complex one which partly relates to discourse struggles over different interpretations of meaning. This can be illustrated by work in social semiotics, which explores ways of analysing the intersection of language, images, colour, and spatial arrangement which occurs for instance in multimedia settings. Work such as van Leeuwen (2009) has shown that the semiotic practices in society are currently changing as a result of the fact that it is dominated by global corporations and technology providers who regulate the production and consumption of multimodal texts. In the final chapter of the book, Leila Monaghan explores the relationship between pragmatics, linguistic anthropology and history in an account of a specific speech event in a community of deaf speakers. She gives an overview of the research traditions of linguistic anthropology – the general study of language within anthropology – and the ethnography of communication – the study of linguistic competence and genres in specific settings. Her particular focus on deaf communities stems from the acknowledgement that “deaf communities often exist despite great opposition to them”. Despite this opposition, deaf people congregate and tell stories about themselves, often performed as narratives about past events that construct their culture. The question of what language should be used – signing or lipreading/speaking – is at the heart of the struggles over deaf culture. Monaghan’s study is based on video-recorded data of New Zealand deaf speakers, where a
Introducing the pragmatics of society
21
group of friends reject the ideological framework of the school authorities and created their own new community. The data is annotated with linguistic (spoken and signed) information as well as non-linguistic finger, hand and body movements. A historical approach is necessary in order to capture the full significance of the events described, because they become part of an – essentially historical – discussion of how this group breaks away from an entire system. From this she generalises to claiming that “to understand the pragmatics of any situation, we need both an in-depth investigation of the event itself and a larger historical context to be able to judge not only the importance of the event but where that event has come from and is pointing towards”.
Challenges for the future Sociopragmatics is an approach to pragmatics which is inspired by recent advances in sociolinguistics, variational pragmatics, linguistic anthropology, critical discourse analysis and related disciplines. It also raises a number of challenges for the future. There are both theoretical issues and methodological issues to be addressed. There is some consensus that we need a theory of meaning which goes beyond the narrow truth-conditional account in semantics and relates meaning to the linguistic, social and cultural context. Thinking on these issues is also inspired by theories based on the role of indexicality in linking linguistic form to the context. However, formulating such as theory presupposes that we know what we mean by context or by culture. The vitality of the field is seen in the contributions to this volume. Where do we go next? The way forward is empirical research cross-linguistically and within languages to find out more about how the social and cultural situation or context affects language and the interpretation of meaning. The context must be taken in a broad sense to include speakers, speaker identities and roles, social relationships, aspects of the social activity as well as temporary social networks kept together by shared practices and values associated for example with age or gender. Although many theoretical issues are not resolved, we have witnessed a broadening of the field to new speech acts, pragmatic markers and sequential patterns and in a broad sense in the extent of deixis in languages. Such studies need to be continued including more languages and more contexts (activity types, social acts, with different speaker roles). New areas are studied in a sociopragmatic perspective vitalising historical pragmatics as well as applied linguistics. Sociolinguistics and pragmatics have for instance influenced the study of speech acts and pragmatic markers in earlier stages of the language. We can also expect to see more cross-linguistic studies comparing how non-native and native speakers use pragmatic markers or make requests or apologies resulting in a revival of contrastive linguistics for language teaching.
22
Karin Aijmer and Gisle Andersen
New methodologies need to be tested. Bilingual and multilingual corpora have recently entered the corpus scene and now make it possible to systematically compare similarities and differences between languages on the lexical, grammatical and pragmatic level. Corpora need to be combined with ethnographic data and methodologies and experimental methods such as the discourse completion test used in speech act studies need to be refined.
Notes 1. We are greatful to Geoffrey Leech and Jonathan Culpeper for their careful reading and valuable comments to this introductory chapter.
References Adolphs, Svenja 2008 Corpus and context. Investigating pragmatic functions in spoken discourse. Amsterdam/Philadelphia: John Benjamins. Aijmer, Karin 1996 Conversational routines in English: Convention and creativity. London and New York: Longman. Aijmer, Karin 2009 The pragmatics of adverbs. In: Günter Rohdenburg and Julia Schlüter (eds.), One language, two grammars? Differences between British and American English, 324–340. Cambridge: Cambridge University Press. Ameka, Felix 1992 Interjections: The universal yet neglected part of speech. Journal of Pragmatics 18: 101–118. Andersen, Gisle 2001 Pragmatic markers and sociolinguistic variation. Amsterdam/Philadelphia: John Benjamins. Andersen, Gisle and Thorstein Fretheim (eds.) 2000 Pragmatic markers and propositional attitude. Amsterdam/Philadelphia: John Benjamins. Auer, Peter 1993 Zur Verbspitzenstellung im gesprochenen Deutsch. [The verb in first position in spoken German]. Deutsche Sprache 3: 193–222. Austin, J. L. 1962 How to do things with words. Oxford: Clarendon Press. Barron, Anne and Klaus Schneider (eds.) 2005 The pragmatics of Irish English. Berlin/New York: Mouton de Gruyter. Barron, Anne and Klaus Schneider 2009 Variational pragmatics: Studying the impact of social factors on language. Intercultural Pragmatics 6(4): 425–442.
Introducing the pragmatics of society
23
Blakemore, Diane 2002 Relevance and linguistic meaning. The semantics and pragmatics of discourse markers. Cambridge: Cambridge University Press. Blum-Kulka, Shoshana 1989 Playing it safe: The role of conventionality in indirectness. In: Blum-Kulka et al. (eds.), 37–70. Blum-Kulka, Shoshana, Juliane House and Gabriele Kasper (eds.) 1989 Cross-cultural pragmatics: Requests and apologies. Norwood, NJ: Ablex. Brown, Penelope and Stephen C. Levinson 1978 Universals in language usage: Politeness phenomena. In: Esther Goody (ed.), Questions and politeness, 56–289. Cambridge: Cambridge University Press. Brown, Penelope and Stephen C. Levinson 1987 Politeness. Cambridge: Cambridge University Press. Channell, Joanna 1994 Vague language. Oxford: Oxford University Press. Cheshire, Jenny 2007 Discourse variation, grammaticalisation and stuff like that. Journal of Sociolinguistics 11: 155–193. Chomsky, Noam 1965 Aspects of the theory of syntax. Cambridge, Mass.: The MIT Press. Cuenca, Maria-Josep 2008 Pragmatic markers in contrast: The case of well. Journal of Pragmatics 40: 1273–1391. Culpeper, Jonathan 2009 Historical sociopragmatics: an introduction. Journal of Historical Pragmatics 10(2): 179–186. Culpeper, Jonathan 2010 Historical sociopragmatics. In: Andreas H. Jucker and Irma Taavitsainen (eds.), Historical pragmatics, 69–96. Berlin/New York: Mouton de Gruyter. Culpeper, Jonathan 2011 Impoliteness: Using language to cause offence. Cambridge: Cambridge University Press. Curl, Traci and Paul Drew 2008 Contingency and action: A comparison of two forms of requesting. Research on Language and Social Interaction 41(2): 129–153. De Fina, Anna 1997 An analysis of Spanish bien as a marker of classroom management in teacherstudent interaction. Journal of Pragmatics 28: 337–354. Duranti, Alessandro and Charles Goodwin 1992 Rethinking context. Cambridge: Cambridge University Press. Duranti, Alessandro (ed.) 2009 Linguistic anthropology: a reader. Malden, Mass.: Wiley-Blackwell. Eelen, Gino 2001 A critique of politeness theories. Manchester: St. Jerome. Foulkes, Paul and Gerard Docherty 2006 The social life of phonetics and phonology. Journal of Phonetics 34:409–438.
24
Karin Aijmer and Gisle Andersen
García, Carmen 2008 Different realizations of solidarity politeness: Comparing Venezulean and Argentinean invitations. In: Klaus Schneider and Anne Barron (eds.), Variational pragmatics: A focus on regional variation in pluricentric languages, 267–302. Amsterdam/Philadelphia: John Benjamins. Giles, Howard, Nikolas Coupland and Justine Coupland 1991 Accommodation theory: Communication, content, and consequence. In: Howard Giles, Nikolas Coupland and Justine Coupland (eds.), Contexts of accommodation: Developments in applied sociolinguistics, 1–68. New York: Cambridge University Press. Goddard, Cliff and Anna Wierzbicka 2004 Cultural scripts: What are they and what are they good for? Intercultural Pragmatics 1–2: 153–166. Golato, Andrea 2005 Compliments and compliment responses. Grammatical Structure and sequential organization. Amsterdam/Philadelphia: John Benjamins. Gumperz, John and Stephen C. Levinson 1996 Introduction: linguistic relativity re-examined. In: John J. Gumperz and Stephen C. Levinson (eds.), Rethinking linguistic relativity, 1–18. Cambridge: Cambridge University Press. Gumperz, John 1982 Discourse strategies. Cambridge: Cambridge University Press. Gumperz, John 2001 Interactional sociolinguistics: A personal perspective. In: Deborah Schiffrin, Deborah Tannen and Heidi E. Hamilton (eds.), The Handbook of Discourse Analysis, 215–228. Oxford: Blackwell. Hanks, William F., Sachiko Ide and Yasuhiro Katagiri 2009 Introduction: Towards an emancipatory pragmatics. Journal of Pragmatics 41(1): 1–9. Harris, Sandra 2001 Being politically impolite: Extending politeness theory to adversarial political discourse. Discourse and Society 12(4): 451–472. Heritage, John 1984 A change-of-state token and aspects of its sequential placement. In: J. Maxwell Atkinson and John Heritage (eds.), Structures of Social Action. Studies in Conversation Analysis, 299–345. Cambridge: Cambridge University Press. Holmes, Janet 1986 Functions of you know in women’s and men’s speech. Language in Society 15: 1–22. Holmes, Janet 1988 Sort of in New Zealand women’s and men’s speech. Studia Linguistica 42(2): 85–121. House, Juliane and Gabriele Kasper 1981 Politeness markers in English and German. In Coulmas, Florian (ed.), Conversational Routine, 157–186. The Hague: Mouton. Huspek, Michael 1989 Linguistic variability and power: An analysis of you know/ I think variation in working-class speech. Journal of Pragmatics 35: 661–683
Introducing the pragmatics of society
25
Hymes, Dell. (ed.) 1964 Language in Culture and Society. New York/London: Harper and Row. Ide, Sachiko 1989 Formal forms and discernment: Two neglected aspects of linguistic politeness. Multilingua 8(2–3): 223–248. Jucker, Andreas H. and Irma Taavitsainen (eds.) 2010 Historical Pragmatics. Berlin/New York: Mouton de Gruyter. Lakoff, Robin 1975 Language and woman’s place. Language in Society 2(1): 45–80. Leech, Geoffrey 1983 Principles of Pragmatics. London and New York: Longman. Local, John and Gareth Walker 2005 Methodological imperatives for investigating the phonetic organisation and phonological structures of spontaneous speech. Phonetica 62(2–4):120–130. Locher, Miriam A. 2004 Power and Politeness in Action: Disagreements in Oral Communication. Berlin/New York: Mouton de Gruyter. Locher, Miriam A. and Sage L. Graham (eds.) 2010 Interpersonal Pragmatics. Berlin/New York: Mouton de Gruyter. Mey, Jakob. L. 1993 Pragmatics: an introduction. Oxford: Blackwell. Moerman, Michael 1988 Talking Culture: Ethnography and Conversation Analysis. Philadelphia, PA: University of Pennsylvania Press. Muhr, Rudolph 1994 “The pragmatics of a pluricentric language: A comparison between Austrian German and German German. In: Klaus Schneider and Anne Barron (eds.), Variational pragmatics: A focus on regional variation in pluricentric languages, 209–242. Amsterdam/Philadelphia: John Benjamins. Ochs, Elinor 1996 Linguistic resources for socializing humanity. In: John J. Gumperz and Stephen C. Levinson (eds.), Rethinking linguistic relativity, 407–437. Cambridge: Cambridge University Press. Ogden, Richard 2001 Turn transition, creak and glottal stop in Finnish talk-in-interaction. Journal of the International Phonetic Association 31(1):139–152. Peräkylä, Anssi 1997 Reliability and validity in research based on transcripts. In: David Silverman (ed.), Qualitative Research. Theory, Method and Practice, 201–220. London: Sage. Romero-Trillo, Jesús 2006 Discourse markers. In: Keith Brown (ed.), Encyclopedia of Language and Linguistics, 639–641. 2nd edition. Oxford: Elsevier. Sacks, Harvey, Emanuel A. Schegloff and Gail Jefferson 1974 A simplest systematics for the organization of turn-taking for conversation. Language 50(4): 696–735.
26
Karin Aijmer and Gisle Andersen
Schiffrin, Deborah 1994 Approaches to discourse. Oxford, UK/Cambridge, Mass: Cambridge University Press. Schmid, Hans-Jörg (ed.) Forthcoming Cognitive Pragmatics. Berlin/New York: Mouton de Gruyter. Schneider, Klaus 2005 ‘No problem, you’re welcome, anytime’: Responding to thanks in Ireland, England, and the U.S.A. In: Anne Barron and Klaus Schneider (eds.),. The pragmatics of Irish English, 101–139. Berlin/New York: Mouton de Gruyter. Schneider, Klaus and Anne Barron (eds.) 2008 Variational pragmatics: A focus on regional variation in pluricentric languages. Amsterdam/Philadelphia: John Benjamins. Schneider, Klaus P. and Anne Barron (eds.) Forthcoming Pragmatics of Discourse. Berlin/New York: Mouton de Gruyter. Searle, John R 1969 Speech acts: an essay in the philosophy of language. Cambridge: Cambridge University Press. Sifianou, Maria 1989 On the telephone again! Differences in telephone behaviour: England vs. Greece. Language in Society 18: 524–544. Sinclair, John and Malcolm Coulthard 1975 Towards an analysis of discourse. Oxford: Oxford University Press. Spencer-Oatey, Helen 2000 Rapport management: A framework for analysis. In: Helen Spencer-Oatey (ed.), Culturally Speaking: Managing Rapport through Talk across Cultures, 11–46. London: Continuum. Sperber, Dan and Deirdre Wilson 1986/1995 Relevance: Communication and Cognition. 1st/2nd edition. Oxford: Blackwell. Stenström, Anna-Brita 1986 What does really really do? Strategies in speech and writing. In: Gunnel Tottie and Ingegerd Bäcklund (eds.), English in speech and writing, 149–163. Uppsala: Acta Universitatis Upsaliensis. Stenström, Anna-Brita 2002 Taking another look at really. In: Sybil Scholz, Monika Klages, Evelyn Hantson and Ute Römer (eds.), Language: Context and cognition, 301–308. München: Langenscheidt-Longmann. Stenström, Anna-Brita, Gisle Andersen and Ingrid Kristine Hasund 2002 Trends in Teenage Talk: Corpus compilation, analysis and findings. Amsterdam: John Benjamins. Stivers, Tanya, Nick J. Enfield, Penelope Brown, Christina Englert, Makoto Hayashi, Trine Heinemann, Gertie Hoymann, Federico Rossano, Jan Peter de Ruiter, KyungEun Yoon and Stephen C. Levinson 2009 Universality and cultural specificity in turn-taking in conversation. Proceedings of the National Academy of Science 106(26): 10587–92. Tagliamonte, Sali 2005 So who? Like how? Just what? Discourse markers in the conversations of English speaking youth. Journal of Pragmatics 37(11): 1896–1915.
Introducing the pragmatics of society
27
Terraschke, Agnes and Janet Holmes 2007 ‘Und Tralala’: Vagueness and general extenders in German and New Zealand English. In: Joan Cutting (ed.), Vague Language Explored, 198–222. Basingstoke, UK: Palgrave MacMillan. Thomas, Jennifer A. 1981 Pragmatic failure. University of Lancaster, unpublished M.A. dissertation. Thomas, Jennifer A. 1983 Cross-cultural pragmatic failure. Applied Linguistics 4(2): 91–112. Van Leeuwen, Theo 2009 Discourse as the recontextualisation of social practice: a guide. In: Ruth Wodak and Michael Meyer (eds.), Methods of CDA, 144–161. London: Sag. Verschueren, Jef 1995 The pragmatic perspective. In: Jef Verscheuren, Jan-Ola Östman and Jan Blommaert, Handbook of pragmatics: Manual. Amsterdam: John Benjamins. Verschueren, Jef 2009 Introduction: The pragmatic perspective. In Jef Verschueren and Jan-Ola Östman (eds.), Key notions for pragmatics, 1–27. Amsterdam: John Benjamins. Wierzbicka, Anna 1997 Understanding cultures through their key words: English, Russian, Polish, German, and Japanese. Oxford: Oxford University Press. Wierzbicka, Anna 2003[1991] Cross-Cultural Pragmatics. 2nd ed. Berlin: Mouton de Gruyter.
I.
Social, regional and situational factors
1.
Doing age and ageing: language, discourse and social interaction Alexandra Georgakopoulou and Anna Charalambidou
1.
Introduction
Sociolinguistic research on language and identities has by and large privileged certain foci of inquiry, such as social class and, following that, gender, leaving age as an under-represented category. As has been observed before (Androutsopoulos and Georgakopoulou 2003b:6), age has remained a hidden or backgrounded category and this is particularly evident in influential studies that have documented how language shapes and is shaped by for instance, gender or ethnicity, whilst focusing exclusively on data from adolescents. In turn, within the category of ‘age’, there is also a pecking order of level of attention with research having traditionally concentrated on the early parts of the life course (N. Coupland 2004: 69), leaving language in later life (even more) under-researched. In this chapter, we wish to redress the balance on both counts. We will foreground the category of age in our overview of language and identity studies at the same time as bringing together ‘young’ and ‘old’ age studies. In pursuing both aims, we are building on recent collections that have brought together socio-pragmatic and sociolinguistic research on youth language and identities (e.g. Stenström, Andersen and Hasund 2002; Androutsopoulos and Georgakopoulou 2003a). However, we are in a far more uncharted territory as regards the co-examining of studies of young and old age and communication. We do see it as a worthwhile enterprise though that will allow us to establish commonalities or equally rifts between the two strands and that will pave the way towards a much needed inclusive approach. To pursue the above aims, we will provide an overview of young age and old age in language studies, as they have developed separately. We will subsequently draw synthesizing or comparative conclusions. Below, we will document what we see as a generalizeable shift with regard to both young and old age studies from earlier normative accounts that treated age as a static and homogeneous social variable (section 3) and their initial scrutiny (section 4) towards recent, practice-based studies, which locate language choices in specific sites and activities that social actors are engaged in (section 5).
32 2.
Alexandra Georgakopoulou and Anna Charalambidou
Defining young and old age
The difficulty in defining young and old age is demonstrated by the proliferation of terms which are themselves part of competing definitions and diverging traditions. For instance, within youth language studies, the term adolescence has monopolized earlier research. As we will see below, the use of the term adolescence has been linked with the assumptions and aims of research on youth language. Adolescence is also a biological and developmental criterion used as the guiding principle in definitions. Youth in earlier research was seen as a transitional, developmental phase with fixed boundaries and in between childhood and adulthood. In Förnas’ terms, it was determined “in relation to that which is interpreted as respectively childish or adult” (1995: 3). Definitions of youth using rigid chronological boundaries and the accompanying emphasis on adolescence have come under fire (e.g. Wyn and White 1997; Bucholtz 2002) not only for their one-dimensional rendering of what is a complex and socio-culturally defined experience, but also for their homogenizing of the diversity of individuals with different biographies and trajectories that engender different positionings and self-perceptions in relation to age. Youth is currently defined on the basis of a host of biological, social, psychological and cultural criteria, and it is widely accepted that it presents fuzzy boundaries with other life phases and with how social actors perceive those boundaries (e.g. see Androutsopoulos and Georgakopoulou 2003a for examples). It thus includes a whole variety of “cultural actors whose experiences are best understood from their own point of view” (Bucholtz 2002: 533) and not in relation to adult and child categories. Similarly, the issue of who can be defined as an old person has been raised quite early on in gerontology. The argument has been that there is nothing intrinsic about being old that differentiates older from younger populations in an easily categorizable way. Also the heterogeneity of people categorized as ‘elderly’ or ‘older adults’ and the difficulties this imposes on defining old age have been widely attested.1 In addition, heterogeneity in social, physical and mental functioning of over 60s and in social perceptions about them has given rise to the need to define different ‘stages’ of old age: e.g. ‘young-old’ or ‘third agers’ and ‘old-old’ or ‘4th agers’.2 At a different level, as is the case in youth research, it has been argued that chronological age is a poor predictor of ageing-related behaviours. Other criteria, such as contextual age (i.e. levels of interpersonal interaction, mobility, life satisfaction, socio-economic status) and social age identity (one’s subjective perception of self as member of an age group), in addition to biological/functional definitions of old age, have been proposed (Butler 1975: 18; Pecchioni, Ota and Sparks 2004: 170; Rawlins 2004: 279). A more radical conceptualisation of age proposes that ageing is in fact exclusively culturally constructed. Ageing is either wholly disassociated from physical decline (over-constructivism) or it is associated with the body but the body and physical change do not exist prior to or outside language (critical paradigm). In
Doing age and ageing: language, discourse and social interaction
33
fact, ageing in critical research is a symbolic construct and its meaning is shifting in relation to economic, political, technological and historical context, and language as well as theory and policies construct age identities (Taylor 1994). Although such constructivist and critical approaches have influenced studies about age stereotypes in areas such as literature, paintings, print advertisements, policies and theories (Taylor 1994; J. Coupland 2003; Hepworth 2004), this radical deconstruction of age has not fed into sociolinguistic studies of age and ageing. It is interesting to note that even in practice-based studies, as we will discuss below, the chronological criterion has not altogether been abolished. For instance, researchers overwhelmingly categorize informants as older adults on the basis of chronological age (often over 65).
3.
Normative approaches and the deficit paradigm
Within the social sciences, one of the dominant approaches to social identities, including age, has viewed them as homogeneous, pre-existing and relatively static social categories that can be demographically assigned to individuals. In sociolinguistic research, this view was implicated in earlier accounts that attempted to map linguistic elements (mainly phonological variables) onto the social identity under investigation (be it gender, social class or age) in clear-cut ways. What we here refer to as the deficit paradigm is characterized by a developmental outlook on age that posits the two endpoints of youth and old age as aberrations from norms. Therefore, there is a more or less explicit bias to seeing the language of the young (mostly teenagers) as a divergent variety, an anti-language and by implication, a ‘deficient’ variety, if compared with that of the adults. Even though it emanates from different assumptions about old age as to those about youth, this deficit paradigm runs across studies on old age, too. To look at this paradigm in more detail below, we provide an overview on studies of youth language (mainly variationist), and then we examine studies on language in later life, focusing on the phenomenon of old-age loquacity or verbosity. Although, as we have already suggested, age has not been focal to many studies, the emphasis on linguistic variation and how it is conditioned by social identities is evident in the few existing ones. Phonological variants, particularly non-standard ones, have been the ones mostly put under scrutiny within sociolinguistic studies that are often described as ‘variationist’ (e.g. Rickford et al. 1991, Dubois 1992, Lee 1995, Kerswill 1996, Stenström et al. 2002). The underlying assumption in such studies has been that the speech of adolescents would somehow mirror their propensity for defying anything that is regulated, proper, standardized and by extension characteristic of ‘adults’, particularly those acting as figures of authority in their lives (e.g. parents, teachers). Cheshire’s study (1982) of adolescents in Reading, England, is a case in point. The study certainly went beyond pre-
34
Alexandra Georgakopoulou and Anna Charalambidou
vious studies of sociolinguistic variation in that it linked language variation not just with age but with age-specific features of socialization (e.g. the peer group) and with differentiated participation roles within the peer group. In addition, gender proved to shape language use, thus providing evidence for the now widely held idea that it is impossible to talk about adolescence as an undifferentiated whole. In particular, adolescent boys were found to use more non-standard forms than girls and this was explained on the basis of the forms’ association with a model of masculinity (i.e. based on machismo and acting tough) that was valued locally. The notable feature about such an approach is the belief that language choices at the micro-level can be mapped onto social categories such as age in relatively straightforward ways. A common finding in studies such as the above is that adolescents are heavy vernacular users with a propensity for non-standard or regional (dialectal) language use (e.g. see Labov 1972, Romaine 1984, Radtke 1990, Chambers 1995, Stenström et al. 2002). Although the processes of slang formation in the lexis of many languages have been amply described and many similarities have been documented (for a discussion, see Androutsopoulos and Scholz 1998), concrete interactional analyses of how such slang is used are lagging behind.3 In ageing research, the deficit paradigm4 stems from a longstanding view of human ageing in the western cultures that sees it in biomedical terms, that is, as progressive, inevitable and unwanted decline (Hepworth 2003: 89; see also Westerhof and Tulle 2007: 235). Consequently, elderly speech is not seen as a matter of choice but as a result of decline (Coulmas 2005: 62). Studies of language and ageing focus on decrement in linguistic and communicative ability, with a special emphasis on the language of communicatively and cognitively impaired older adults, i.e. with conditions such as Parkinson’s (Obler 1980), Alzheimer (De Bot and Makoni 2005), aphasia (Korpijaakko-Huuhka and Klippi 2008) or the like. Research within this paradigm typically employs quantitative tools to study talk elicited in experimental conditions (e.g. surveys or research interviews) and detailed scales to measure psychosocial, cognitive and linguistic functioning of the research subjects. A classic example of this paradigm is found in the volume edited by Obler and Albert (1980). In Obler’s chapter (1980) on the narrative discourse style of older adults, a group of older adults, diagnosed with Parkinson’s, are compared with a group of healthy older adults and the latter are in turn compared with a group of younger adults. The older (compared to younger) and the parkinsonian elderly (compared to healthy elderly) participants were found to use a more elaborate narrative style with fewer and more complex sentences and greater ‘loquacity’ (greater number of words per theme, Obler 1980: 78). Although the author contemplates as possible explanation for this more elaborate discourse style of the elderly, not only the loss of primary cognitive functions but also the cultural/educational differences between cohorts, it is obvious that what is deemed researchable with regard to language in later life is deviations from the norm, i.e. the language of
Doing age and ageing: language, discourse and social interaction
35
healthy middle-aged adults. At the same time, age is taken as a fixed identity that exists prior to and outside of discourse. The issue of elderly verbosity is also addressed in a later study that also falls within the decrement paradigm (Gold, Arbuckle and Andres 1994). The frequency and extent of off-target verbosity (i.e. talk that diverges from the nominal topic of the conversation) is measured in four corpora of research interviews with older adults. Detailed scale measurements are also used to measure psychosocial variables (e.g. extroversion, levels of stress) and cognitive functioning, while demographic categories, such as gender and socio-economic and marital status, are also employed to explain verbosity. Leaving aside the problems of defining ‘off-target verbosity’, the results show that only a small percentage of it is explained by agerelated factors (1–12 %). Statistical results of controlled, experimental studies lend themselves to generalisation and can influence social policies (Wilson 1991). However, the validity of the results is open to challenge, not least because language is dislocated from the context in which it naturally occurs (see e.g. N. Coupland and Coupland 2001). For example, in Gold et al.’s (1994) research, the high occurrence of off-target verbosity could be encouraged by the situational context (interview), where the interviewee is given the floor without interruption. Also the study does not address the issue that what is regarded as relevant or appropriate by the interlocutors is also context-specific. On the whole, what is investigated in this paradigm is the agerelated deterioration of language use or ‘an appraisal of residual competence’ (N. Coupland and Coupland 1990: 453; another example is Cohen 1994 which focuses on problems in the retrieval of proper names in elderly communication). Yet, as Coupland and Coupland point out, the expectation of decline in speech and cognition that underlines this line of inquiry legitimizes and naturalizes the deficit model of ageing (N. Coupland and Coupland 1990:455).
4.
The deficit paradigm under scrutiny
The deficit paradigm, as outlined above, characterizes earlier work on age identities. The recent shift towards contextualized views of language and identities has rendered many of the assumptions of this paradigm problematic. To be specific, in both young and old-age research, contextual variables, other than chronological age, as well as the local situation of the interaction have been employed in explaining the linguistic behaviour of the participants. This has led, on the one hand, to the re-conceptualisation of ‘peer group’ in youth language research and, on the other hand, to problematizing the relation between language in old age and decline in linguistic performance. Below, we will flesh out the main characteristics of studies that have scrutinized the deficit paradigm starting with research on youth, which has more radical methodological implications.
36
Alexandra Georgakopoulou and Anna Charalambidou
First of all, it has become apparent that the data should not be collected exclusively in research interviews but in situations that seem typical of the participants’ lives. In this respect, there has been an increasing tendency towards ethnographic methods which include extensive fieldwork and this is even the case in studies which are still interested in how linguistic variation is conditioned by social identities. For instance, Eckert’s study (2002) of linguistic variation amongst female students of an American high school has linked language choices with the local cultures of the students and their affiliations with them, producing a much more nuanced picture of ‘adolescent’ language usage, one that connects meaning-making of acting and talking with which particular group one’s allegiances are in. Second, there is currently a realization that the emphasis on variables at the micro-linguistic level (e.g. phonological, lexical) produces a restrictive view of communication, one that overlooks the role of larger units (e.g. utterances, episodes, types of text) in communication. The recognition of the importance of the type of sequence and activity for how language choices are shaped is one of the hallmarks of more recent approaches to language and identities (e.g. see Benwell and Stokoe 2006). This is coupled with an interactional view which sees all language choices as collaboratively produced between speaker and audience and thus as saying as much about the speaker’s identities as about the audience’s (e.g. see chapters in Antaki and Widdicombe 1998). There are many positions within both these shifts towards activity-based and interactional views of communication which this discussion cannot do justice to. For our purposes, it suffices to say that what they converge on is attention to the context of language use, even if the concept of context can be variously defined. In the case of language and age research, this means that emphasis is placed on: – what the participants do, where, when and why – what their roles and relations are and what their conventionalized expectations are about what is to be done in given settings – how socially and culturally recognized events that may be dedicated to specific communicative purposes may shape their ways of speaking – how these situationally and socio-culturally constrained ways of speaking relate to the speakers’ identities, including age. This emphasis on ‘the sites of lived experience in which locally motivated linguistic choices can be creatively related to extra-situational social categories and meanings’ (Androutsopoulos and Georgakopoulou 2003b: 7) has, in the case of youth language research, re-conceptualized the influence of the peer group in adolescent language use. The peer group is no longer seen as a generalized concept for friendship-based groupings of young people but as a situated set of practices. In Bucholtz’s ethnographic study (1999), for instance, the notion of peer group was qualified in terms of distinct communities of practice, to which different female students of a high school in California belonged. Originating in Wenger’s
Doing age and ageing: language, discourse and social interaction
37
work (1998; see also Lave and Wenger 1991), the notion of community of practice allows for a more fine-tuned view of a peer group as an aggregate of people who come together around a shared purpose and who through regular interaction over time develop a shared understanding of their bonds and a shared repertoire of semiotic resources. In Bucholtz’s data, these resources included ways of using language, but also shared ideologies regarding language, shared attitudes to and participation in the school life, dating practices and many semiotic practices, including dress code. The language use characteristic of each community of practice was located at different levels, e.g. phonological, lexical, interactional, etc. Such contextually sensitive studies have brought to the fore the heterogeneity and multiplicity of youth language practices even within the same site, e.g. a school (cf. Eckert 2000). Developments in ageing research that have challenged the deficit paradigm have been slower. In the first instance, research introduced additional demographic categories that could account for changes in older adults’ language as opposed to exclusively age-related decline in language ability. For example, De Bot and Makoni (2005) examine the effects of age and education on narrative and syntactic complexity in older Chinese in the U.S.A., using formalized tests and scales. The assumption that individuals will continuously decline as they age is challenged, and other categories such as bilingualism, gender, space of interaction, ethnoracial grouping, and education are found to be consequential for language production and comprehension in later life. Another study that belongs to the same paradigm is Korpijaakko-Huuhka and Klippi’s paper (2008). This follows the ‘International Classification of Functioning, Disability and Health’, a new bio-psychosocial model which describes a person’s condition from various perspectives. The study focuses ‘on functioning instead of disorders or deficits’ (Korpijaakko-Huuhka and Klippi 2008: 482) and on this basis it provides an overview of changes in language and discourse skills in later life, comparing them to changes in older aphasic or demented adults (e.g. suffering from Alzheimer). Unlike research in the deficit paradigm, this study does not necessarily evaluate change as decline. It also challenges the widely held view that older adults exhibit off-target verbosity (cf. p. 5). Overall, the study documents the effect of variables such as communication disorders, education, and lifestyle choices on elderly communication which were overlooked in previous studies that classified language change as decline and primarily age-driven. The above studies have succeeded in bringing into the discussion many variables that affect linguistic competence in later life. However, as with the deficit paradigm, the data remain largely elicited, the local, interactional organisation of talk is neglected, and peer group conversations are exceptionally scarce. Moreover, age, albeit intersected with other identities, continues to be seen as a more or less static, pre-discursive entity. Thus, the focus is still on how health conditions render elderly communication as deficient.
38
Alexandra Georgakopoulou and Anna Charalambidou
The influence of Communication Accommodation Theory (CAT) in the study of intergenerational communication has also been instrumental in challenging some de-contextualized conceptualisations of identities and stereotypical assumptions in ageing research. The CAT framework explains modifications in speech, based on various accommodation strategies of the social actors (e.g. convergence to or divergence from the interactional partner’s speech style), while it also acknowledges the complex relations between various factors of the communicative event, such as the socio-historical context, the participants’ socio-psychological state, the immediate situation, the goals of interaction etc. (Shepard, Giles and Le Poire 2001). Much of the research on age stereotyping in interaction has been carried out within this framework (for an outline of such research see Hummert et al. 2004). The application of CAT to intergenerational interaction has shown that younger adults perceive older adults’ talk as under-accommodating, i.e. extensively diverging from the interlocutor’s style, topics introduced, and their own as over-accommodating, that is, as converging to the interlocutor’s style and to viewpoint that the recipient will presumably share (cf. e.g. Ylänne-McEwen 1999). The elderly have also been found to use miscommunication (e.g. with under-accommodation) and self-handicapping strategies towards the young, by excusing themselves from performing adequately in interaction, saying things such as ‘I forget everything these days’. These strategies have been shown to elicit the recipient’s sympathy, mitigate criticism, encourage praise (for unexpected good performance) and thus be ultimately face-enhancing for the elderly speaker (N. Coupland et al. 1991a: 49). Overall, even though age identities are still treated as pre-existing categories in this paradigm, ageist assumptions regarding elderly discourse have been exposed; in fact, the linguistic characteristics of third agers are necessarily connected to decrement in language production and comprehension. Research on youth and elderly language focuses now on how age, along with other contextual factors (e.g. peer group, level of education), affects the interlocutors’ discursive behaviour. More dynamic approaches that document social ageing in earnest are to be found within the practice-based framework, which we will discuss below.
5.
Practice-based approaches to age and ageing
What we call here ‘practice-based approaches’ to language and identities can be brought together by their emphasis on social interaction, on what language does in specific contexts and for specific purposes. They are also characterized by a view of identities as flexible and fleeting categories, agentively constructed and de-constructed in situ by the interlocutors (for a discussion, see De Fina and Georgakopoulou 2008). The data used are usually naturally occurring conversations in informal everyday situations or institutional settings. Such approaches are increasingly gaining currency in age/aging research thus appearing as the most serious chal-
Doing age and ageing: language, discourse and social interaction
39
lenge to deficit-based accounts. The meanings of age are therefore viewed as socio-culturally and interactionally constructed and only partially driven by biology (cf. Pecchioni et al. 2004: 171). As a result, notions of adolescent rebellion and old-age decline are no longer taken for granted. Instead, the aim is to explore how social actors may accept, resist and negotiate age identities that may be ascribed to them in specific social interactions; also, how age-related positionings are shaped by the context at hand. In addition, emphasis is placed on how individuals make specific life-style choices and style themselves in ways that ‘construct’ them as young or old, regardless of their biological age. Although there is variation within the practice-based paradigm, certain key assumptions underlie research. First, linguistic devices do not straightforwardly point to a specific category; rather they can simultaneously perform different functions. Second, there is scepticism about generalizations of the kind ‘this is how young or old people talk’, not least because there is a great deal of heterogeneity within each age group. In addition, it is acknowledged that age identities are often co-articulated (i.e. co-constructed or inter-related in interactions) with other extrasituational (e.g. gender) and situational (e.g. institutional) identities (cf. Eckert and McConnell-Ginet 1999). Lastly, emphasis is placed on how identities are collaboratively produced in local contexts. A main tendency within practice-based approaches to youth language has been to document the use of specific linguistic devices in certain contexts of peer group communication without however wishing or equally being able to draw generalizations on their basis about ‘how young people speak’. That said, it is notable that the semiotic phenomena that have been proposed by numerous studies as markers of peer group membership form a relatively closed list (e.g. Corsaro and Eder 1990; Eder 1993; Eble 1996; Tertilt 1996; Andersen 1997; Ito and Tagliamonte 2003; Kallmeyer and Keim 2003). Such markers comprise: – –
– –
Playful and creative language, (e.g. coinage of new terms, nicknames), teasing and ritual insults or conflicts Innovative and unexpected or incongruous mixings of language registers (e.g. formal/informal), styles and different languages (e.g. code-switching), mainly to humorous effects Use of insulting or taboo lexis as terms of endearment and increased use of expressive language, including ‘intensifiers’ (e.g. just, really) Increased and innovative use of certain discourse markers (e.g. like), interjections (e.g. sort of, you know, I mean), extenders (e.g. and stuff like that), invariant tags (right? innit? yeah?), etc.
The heavy reliance on ingroup markers, shared assumptions and collaborative forms of communication in youth language (e.g. joint telling of stories, ‘mirroring’ or ‘echoing’ of previous turns) has also been noted (e.g. Eder 1998; Deppermann and Schmidt 2003; Georgakopoulou 2007). The frequency of more or less cryptic
40
Alexandra Georgakopoulou and Anna Charalambidou
references to shared cultural experiences (e.g. TV programmes, songs, films) is a similar phenomenon that has been described in terms of young people reaffirming intimacy and familiarity within their peer group (e.g. see Georgakopoulou 2008). At the same time, the importance of peer group remains unchallenged, but as noted above, the peer group is now seen as a much more nuanced concept, organically linked with settings, participant roles and relations and purposes of communication. Nevertheless, even within practice-based approaches, peer group language in later life remains a scarcely researched area. Instead, emphasis is placed on cross-generational communication, especially in the study of age stereotypes in interaction. However, it has been shown that older adults perform certain age identities in intergenerational interactions and others when they interact with peers (N. Coupland, Coupland and Grainger 1991b). Also, there is no agreement in literature as to whether age-adapted communication (see above, section 4) is used by older participants in peer conversations (see e.g. Hummert et al. 2004). Hence, all findings about elderly discourse need to be (re)assessed by studies of peer group interactions amongst elderly people. Research on language in later life has also yielded a list of phenomena (e.g. tellings of painful events and age, reception of patronizing talk etc.) that function as markers of old age. Specifically, devices that have been found to be pivotal in older adults’ talk are painful self disclosures (PSD- telling of ill health, bereavement, loneliness etc.) as well as disclosures of chronological age and self-association with the past (N. Coupland et al. 1991a). These devices are interactionally negotiated, sequentially consequential and ultimately face promoting and central to the performance of elderly identity. Thus age identification is found to be explicitly oriented to in talk, more so than other social categories (N. Coupland 2004:84). Coupland et al.’s work on PSD (1991a) has influenced other research on elderly communication in different contexts. For example, Poulios (2004; 2005; 2008; 2009) examines the notion of PSD in the context of naturally occurring, Greek intergenerational and peer-elderly conversations. Poulios shows a preference for instances of PSD or troubles-telling in the context of first encounters (like N. Coupland et al. 1991a) and provides a thorough investigation of the construction of elderly identity through PSD and the interactional management of stereotypes. Also in dialogue with Coupland et al.’s work on PSD is Matsumoto’s (2009) study, which examines PSD and humour in everyday conversations of older Japanese women. It is found that the humorous construction of tellings of spousal death enables the participants to subvert the stereotype that older women perceive their situation as depressing. The inextricable link of humour and laughter with PSD is absent from both Coupland et al. (1991a) and Poulios’ work (2005; 2008). This could be attributed to the fact the participants of Matsumoto’s study have a long interactional history. Patronizing in intergenerational (often institutional) interactions has been one of the main foci of this paradigm. Communication perceived as patronizing by the
Doing age and ageing: language, discourse and social interaction
41
recipient often consists of over-simplified talk, similar to that used to address young children. It entails verbal modifications (e.g. simplified vocabulary, grammar and syntax, repetition, diminutives and nicknames), paralinguistic features (e.g. loud, slow speech, exaggerated intonation and enunciation, high pitch) and non-verbal behaviour (Hummert and Ryan 2001: 254). The results show patronizing communication towards elderly ranging from less extreme ‘elderspeak’ (which is perceived as inducing dependence; see e.g. De Bot and Makoni 2005) to more extreme ‘secondary baby talk’ (perceived as controlling and severely condescending talk; Hummert and Ryan 2001: 254). Yet, unlike previous research, patronizing is seen as a local interactional accomplishment of both the interlocutors that is very much context-dependent. Although there are findings about lexical, syntactic and pragmatic aspects of youth and old-age language, one of the main conclusions of such studies has been that the same discourse phenomenon can serve as a multi-valent choice that serves different purposes in the same context and even in the same interaction. For instance, code-switching, which has frequently been associated with young people, has been found to be one of the several language choices that in certain contexts serves as a marker of intimacy and in others is used to create boundaries for instance between different peer groups or within a peer group and figures of authority such as teachers (for a discussion see Androutsopoulos and Georgakopoulou 2008: 462, 465). This widely documented multi-functionality of linguistic devices makes it very difficult both to attribute a certain function to one particular choice and to relate it with some certainty to one aspect of social identity (e.g. age) as opposed to another. As Ochs (1992) rightly points out, the discursive construction of an identity may be mediated by various other social factors, such as institutionally defined power, etc. If we carry this over to research on language and age identities, we can for instance claim that the use of code-switching may be more related to informal and friendly relationships than to age. As already suggested, one of the main contributions of practice-based approaches is the documentation of the multiplicity and heterogeneity of young people’s communication practices, thus definitively casting doubt on previous homogenizing accounts. Research on how youth peer groups linguistically demarcate themselves from other groups has shown this very clearly. For instance, the teenage male group studied by Deppermann and Schmidt (2003) set its own culture and language apart from other male youth groups, perceived as having ‘deviant’ lifestyles, and from female groups. Gendered differences in peer groups’ communication have also been amply demonstrated alongside distinct interactional ways in which different groups, be they male or female, build and maintain groupinternal hierarchies (e.g. Goodwin 1990; 2006). The relationships between age identities and other identities have also been explored. The methodological assumption has been that identities should not be studied in isolation in communication, as in fact they seldom work in that way. In
42
Alexandra Georgakopoulou and Anna Charalambidou
other words, speakers communicate not just as members of one particular social category but as members of multiple categories, some of which are age-based, others gender- or ethnicity-based, and so on. This is often described as the co-articulation of identities (Ochs 1992; also see chapters in De Fina, Bamberg and Schiffrin 2006) and acknowledging this co-articulation forces the analysts to explore how identities interrelate and are mediated by one another as opposed to singling out one. It is fair to say that as far as youth age identities are concerned, there is a volume of influential work that has nonetheless not explicitly focused on the role that the actual age of the participants plays in certain linguistic choices. In that respect, only tenuous conclusions can be drawn about their possible influence. A case in point is work on ‘crossing’ that has shown how the use of linguistic resources for which the speaker cannot claim either ‘ownership’ (e.g. in terms of belonging to the specific ethnic group which employs these resources) or full mastery are pivotal in processes both of boundary-setting and of alignment in the secondary school students’ interactions. In particular, Rampton (1995) found that the adolescent students in the London school he researched used Stylized Asian English to mark certain behaviours as inappropriate, improper or marginal as well as to disrupt adult authority (in this case, that of their teachers). Such takings on of ‘other voices’ have been found to complicate issues of linguistic and cultural membership authenticity and to level ethnically inflected divisions in favour of local cultures that are premised on factors other than traditional lines of ethnic belonging. Stylized representations of others and other voices for affiliative or disaffiliative purposes have been amply documented in relation to interactional data from adolescents (e.g. see Kallmeyer and Keim 2003; Georgakopoulou 2006; Spreckels 2006, etc.). In the spirit of practice-based approaches, this line of inquiry has prioritized the social practices that the participants are engaged in and the roles associated with them as factors shaping such interactional phenomena rather than their age. Most of this work has also been closely aligned with late modernity theorizing that has stressed the agency, creativity and power of social actors in interpersonal and self-management, as we have already argued. However, there is still scope for addressing the factor of age (as a socio-culturally shaped and dynamic identity) in more explicit terms and an accompanying need to shed light on how the creative and order-disrupting language activities that have been reported fare in the communication of other social actors, who are, among others, not adolescents. Examples of co-articulation of identities can also be found in ageing research, where age is intersected with gender but also institutional roles and sexuality. Paoletti’s (1998) study on older women, tackles issues of peer group interaction in institutional contexts, such as committee meetings and theatre workshops. Through the participants’ situational enactment or distancing from socially available age and gender identities, the study investigates how institutions affect older women’s identity production and how institutions are affected by them. The conclusion is that although there is stability in what is perceived as shared social knowledge as-
Doing age and ageing: language, discourse and social interaction
43
sociated with old age categories (i.e. ‘old people’ are bound with negative attributions such as loneliness, dependence, frailty), in fact there is variability in the actual use of such categorisations. The interlocutors negotiate positive self-appraisals as an older person through juxtaposition with stereotypical, decline constructions of old people. Paoletti’s ethnomethodological/constructivist approach to identities and the co-articulation of age and gender identifications, is also followed in Tainio (2002), where the sequential organisation of research interviews with elderly couples is analysed and focus is placed on syntactic and semantic elements of language to show actors’ strategic performance of gender and sexual identities against ageist stereotypes.
6.
Conclusions
In this chapter, we set out to bring together young and old-age sociolinguistic studies as two foci of inquiry that have been developed in considerable isolation from one another. Our aim has been to take a step toward a more inclusive approach to age identities. In our overview of studies, we showed that we can talk about a shift from earlier normative accounts to more recent practice-based approaches. The former in fact conceptualized the language of young and old-age population in highly similar ways as a variety that was more or less ‘deficient’ and ‘deviant’ compared to the adult language norms. The challenge of these normative accounts was initially marked with a move towards naturally occurring data, an emphasis on the local and socio-cultural context of the interaction and the recognition of the co-articulation of age with other identities. We then showed how research within the practice-based paradigm focuses on social ageing and shows how age-related identities are negotiated, enhanced or contested in the contingencies of situated activity. As we argued, youth peer group communication and intergenerational communication have been two major concentrations of young and old-age research respectively, resulting in a list of ingroup and old-age markers in each case. In both cases, however, there is a lingering association with assumptions linked to the deficit paradigm: that young people speak in a certain way, distinct from the norm, with their peers, and that the notion of peer group cannot be applied to later life. This last assumption is associated with stereotypical expectations of loneliness, dependence and lack of sociability as prevalent in old age, even though it has been documented in the literature that there is a lot to be said about same-age socialisation of older adults, especially women (e.g. Rawlins 2004). In youth research, deficit assumptions have been, to a certain degree, problematized with the acknowledgement of the multi-functionality and context-dependence of the semiotic phenomena put forward as indexing youth or peer group identities. In ageing studies, on the other hand, the scrutiny of the lasting impact of the deficit paradigm assumptions within practice-based research has
44
Alexandra Georgakopoulou and Anna Charalambidou
begun only recently with more nuanced accounts of same-age interactions in various settings (e.g. Paoletti 1998; Matsumoto 2009; Poulios 2009). However there is still ample scope for researching the notion of peer group in old age. Finally, we indicated how the heterogeneity within each age category has also been explored and how age has been intersected with other larger identities as well as institutional roles and local discourse identities that are inextricably connected to the sequential organisation of talk. Our overview of the literature on language, discourse and social interaction in youth and later life has shown that there are comparable paradigm shifts (from deficit to practice-based accounts) to be found in the disciplinary trajectories of the two areas. Changes in ageing research have, nonetheless, been more gradual, partly because there are significantly fewer studies that focus on language from a socio-pragmatic, sociolinguistic or discourse-analytic perspective than in youth research. However, the theoretical challenges that the two areas have been faced with are similar. Therefore, old-age research has a lot to learn from young-age research, not least by extending its remit to the study of peer group communication. Research on language and age identities is currently taking new directions with the expansion of digital information and communication technologies transforming the ways in which identities and relationships are formed (for a discussion of youth identities research in relation to the new media, see Androutsopoulos and Georgakopoulou 2008: 468–471). It is expected that future studies will provide further evidence about the relevance of all age identities for communication in a wide variety of social and cultural settings.
Notes 1. Cf. Butler (1975); Andrews (1991: 4); N. Coupland, Coupland and Giles (1991a: 58); Nussbaum and Coupland (2004: xii); Settersten (2005); Westerhof and Tulle (2007) for a discussion on different terms used for older adults. 2. There is discrepancy in the literature about the cut-off point between the two stages; ‘young-old’ are defined as individuals from 60 to 74 or from 65 to 75 or from 70 to 84 and ‘old-old’ can be adults above 75 or 80 or 85 (Hummert, Garstka, Ryan and Bonnesen 2004; O’ Hanlow and Coleman 2004; Pecchioni, Ota and Sparks 2004: 173; Estes 2005: 552; Westerhof and Tulle 2007). 3. For a notable exception, see Scwitalla and Streeck 1989 which argue for the double function of slang lexis as a sign of belonging and ingroup re-affirmation as well as a boundary-setting device for ‘outsiders’. 4. Also described as ‘the diachrony and decrement predicament’ in N. Coupland and Coupland 1990; N. Coupland et al. (1991a: 8–14); or ‘functionalist research’ in Taylor (1994: 186).
Doing age and ageing: language, discourse and social interaction
45
References Andersen, Gisle 1997 They gave us these yeah, and they like wanna see like how we talk and all that: The use of like and other pragmatic markers in London teenage speech. In: Ulla-Britt Kotsinas, Anna-Brita Stenström and Anna-Malin Karlsson (eds.), Ungdomsspråk i Norden, 82–95. Stockholm: MINS 43. Andrews, Molly 1991 Lifetimes of commitment: aging, politics, psychology. Cambridge/New York: Cambridge University Press. Androutsopoulos, Jannis K. and Alexandra Georgakopoulou 2003a Introduction: Discourse constructions of youth identities. In: Jannis K. Androutsopoulos and Alexandra Georgakopoulou (eds.), Discourse constructions of youth identities, 1–25. Amsterdam/Philadelphia: J. Benjamins. Androutsopoulos, Jannis K. and Alexandra Georgakopoulou 2008 Youth, discourse, and interpersonal management. In: Gerd Antos and Eija Ventola (eds.), Handbook of interpersonal communication, 457–500. Berlin/New York: Mouton de Gruyter. Androutsopoulos, Jannis K. and Alexandra Georgakopoulou (eds.) 2003b Discourse constructions of youth identities. Amsterdam/Philadelphia: J. Benjamins. Androutsopoulos, Jannis K. and Arno Scholz 1998 Jugendsprache, langue des jeunes, youth language: linguistische und soziolinguistische Perspektiven. Frankfurt am Main/New York: Peter Lang. Antaki, Charles and Sue Widdicombe (eds.) 1998 Identities in talk. London: Sage. Benwell, Bethan and Elizabeth Stokoe 2006 Discourse and identity. Edinburgh: Edinburgh University Press. Bucholtz, Mary 1999 Why be normal?: Language and identity practices in a group of nerd girls. Language in Society 28: 203–223. Bucholtz, Mary 2002 Youth and Cultural Practice. Annual review of Anthropology 31: 525–552. Butler, Robert N. 1975 Why survive? Being old in America. New York: Harper and Row. Chambers, Jack 1995 Sociolinguistic theory: linguistic variation and its social significance. Oxford: Blackwell. Cheshire, Jenny 1982 Variations in an English dialect: a sociolinguistic study. Cambridge: Cambridge University Press. Cohen, Gillian 1994 Age related problems in the use of proper names in communication. In: Mary Lee Hummert, John M. Wiemann and Jon F. Nussbaum (eds.), Interpersonal communication in older adulthood: Interdisciplinary theory and research, 40–57. Thousand Oaks, California: Sage.
46
Alexandra Georgakopoulou and Anna Charalambidou
Corsaro, William and Donna Eder 1990 Children’s peer cultures. Annual Review of Sociology 16: 197–220. Coulmas, Florian 2005 Sociolinguistics: the study of speakers’ choices. Cambridge/New York: Cambridge University Press. Coupland, Justine 2003 Ageist ideology and discourses of control in skincare product marketing. In: Justine Coupland and Richard Gwyn (eds.), Discourse, the body, and identity 127–150. Basingstoke: Palgrave Macmillan. Coupland, Nikolas 2004 Age in social and sociolinguistic theory. In: Jon F. Nussbaum and Justine Coupland (eds.), Handbook of communication and aging research, 69–90. Mahwah, N.J.: Lawrence Erlbaum Associates. Coupland, Nikolas and Justine Coupland 1990 Language and later life: the diachrony and decrement predicament. In: Howard Giles and W. Peter Robinson (eds.), Handbook of language and social psychology, 451–468. Chichester: Wiley. Coupland, Nikolas and Justine Coupland 2001 Language, ageing and ageism. In: W. Peter Robinson and Howard Giles (eds.), The new handbook of language and social psychology, 465–486. Chichester: Wiley. Coupland, Nikolas, Justine Coupland and Howard Giles 1991a Language, society, and the elderly: discourse, identity, and ageing. Oxford, UK/Cambridge, Mass.: Blackwell. Coupland, Nikolas, Justine Coupland and Karen Grainger 1991b Intergenerational discourse: contextual versions of ageing and elderliness. Ageing and Society 11: 189–208. De Bot, Kees and Sinfree Makoni 2005 Language and aging in multilingual contexts. Clevedon, Buffalo: Multilingual Matters. De Fina, Anna and Alexandra Georgakopoulou 2008 Analysing narratives as practices. Qualitative Research 8: 379–387. De Fina, Anna, Deborah Schiffrin and Michael Bamberg (eds.) 2006 Discourse and identity. Cambridge: Cambridge University Press. Deppermann, Arnulf and Axel Schmidt 2003 Vom Nutzen des Fremden für das Eigene. Interaktive Praktiken der sozialen Abgrenzung in einer jugendlichen peer group. In: Hans Merkens and Jürgen Zinnecker (eds.), Jahrbuch Jugendforschung 3, 25–56. Opladen: Leske and Budrich. Dubois, Sylvie 1992 Extension particles, etc. Language Variation and Change 4: 179–203. Eble, Connie 1996 Slang and sociability: In-group language among college students. Chapel Hill, North Carolina: The University of North Carolina Press. Eckert, Penelope 2000 Linguistic variation as social practice. Malden, MA: Blackwell.
Doing age and ageing: language, discourse and social interaction
47
Eckert, Penelope 2002 Demystifying sexuality and desire. In: Kathryn Campbell-Kibler, Robert Podesva, Sarah Roberts and Andrew Wong (eds.), Language and sexuality: contesting meaning in theory and practice, 99–110. Stanford: CSLI Publications. Eckert, Penelope and Sally McConnell-Ginet 1999 New generalizations and explanations in language and gender research. Language in Society 28: 185–201. Eder, Donna 1993 Go get ya a French!’: Romantic and sexual teasing among adolescent girls. In: Deborah Tannen (ed.), Gender and conversational interaction, 17–31. Oxford: Oxford University Press. Eder, Donna 1998 Developing adolescent peer culture through collaborative narration. In: Susan Hoyle and Carolyn Temple Adger (eds.), Kids talk. Strategic language use in later childhood, 82–94. Oxford; New York: Oxford University Press. Estes, Carroll 2005 Women, ageing and inequality: a feminist perspective. In: Malcolm Lewis Johnson, Vern L. Bengtson, Peter G. Coleman and T. B. L. Kirkwood (eds.), The Cambridge handbook of age and ageing, 552–559. Cambridge; New York: Cambridge University Press. Fornäs, Johan 1995 Youth, culture and modernity. In: Johan Fornäs and Göran Bolin (eds.), Youth culture in late modernity, 1–11. London: Sage. Georgakopoulou, Alexandra 2006 Small and large identities in narrative (inter)action. In: Anna De Fina, Deborah Schiffrin and Michael Bamberg (eds.), Discourse and identity, 83–102. Cambridge: Cambridge University Press. Georgakopoulou, Alexandra 2007 Small stories, interaction and identities Amsterdam; Philadelphia: J. Benjamins. Georgakopoulou, Alexandra 2008 ‘On MSN with buff boys’: Self- and other-identity claims in the context of small stories. Journal of Sociolinguistics 12: 597–626. Gold, Dolores Pushkar, Tannis Y. Arbuckle and David Andres 1994 Verbosity in older adults. In: Mary Lee Hummert, John M. Wiemann and Jon F. Nussbaum (eds.), Interpersonal communication in older adulthood: Interdisciplinary theory and research, 107–129. Thousand Oaks, California: Sage. Goodwin, Marjorie 1990 He-said-she-said: Talk as social organization among black children. Bloomington: Indiana University Press. Goodwin, Marjorie 2006 The hidden life of girls. Games of stance, status and exclusion. Oxford: B. Blackwell. Halliday, Michael 1978 Language as social semiotic: the social interpretation of language and meaning. London: Edward Arnold.
48
Alexandra Georgakopoulou and Anna Charalambidou
Hepworth, Mike 2003 Ageing bodies: aged by culture. In: Justine Coupland and Richard Gwyn (eds.), Discourse, the body, and identity, 89–106. Basingstoke: Palgrave Macmillan. Hepworth, Mike 2004 Images of old age. In: Jon F. Nussbaum and Justine Coupland (eds.), Handbook of communication and aging research, 3–29. Mahwah, N.J.: Lawrence Erlbaum Associates. Hummert, Mary Lee and Ellen Bouchard Ryan 2001 Patronizing. In: Peter Robinson and Howard Giles (eds.), The new handbook of language and social psychology, 253–269. Chichester: Wiley. Hummert, Mary Lee, Teri Garstka, Ellen Ryan and Jaye Bonnesen 2004 The role of age stereotypes in interpersonal communication. In: Jon F. Nussbaum and Justine Coupland (eds.), Handbook of communication and aging research, 91–114. Mahwah, N.J: Lawrence Erlbaum Associates. Ito, Rika and Sali Tagliamonte 2003 Well weird, right dodgy, very strange, really cool: layering and recycling in English intensifiers. Language in Society 32: 257–279. Kallmeyer, Werner and Inken Keim 2003 Linguistic variation and the construction of social identity in a German-Turkish setting. In: Jannis K. Androutsopoulos and Alexandra Georgakopoulou (eds.), Discourse constructions of youth identities, 29–46. Amsterdam; Philadelphia: J. Benjamins. Kerswill, Paul 1996 Children, adolescents and language change. Language Variation and Change 8: 177–202. Korpijaakko-Huuhka, Anna-Maija and Anu Klippi 2008 Language and discourse skills of elderly people. In: Gerd Antos and Eija Ventola (eds.), Handbook of interpersonal communication, 501–528. Berlin, New York: Mouton de Gruyter. Labov, William 1972 Language in the inner city: studies in the Black English vernacular. Philadelphia: University of Pennsylvania Press. Lave, Jean and Etienne Wenger 1991 Situated learning: Legitimate peripheral participation. Cambridge: Cambridge University Press. Lee, David 1995 Sociolinguistic variation in the speech of Brisbane adolescents. In: Ivar Werlen (ed.), Verbale Kommunikation in der Stadt, 175–196. Tübingen: Narr. Matsumoto, Yoshiko 2009 Dealing with life changes: humour in painful self-disclosures by elderly Japanese women. Ageing and Society 29: 929–952. Nussbaum, Jon F. and Justine Coupland (eds.) 2004 Handbook of communication and aging research. Mahwah, N.J.: Lawrence Erlbaum Associates. O’Hanlow, Ann and Peter Coleman 2004 Attitudes towards aging: adaptation, development and growth into later years. In: Jon F. Nussbaum and Justine Coupland (eds.), Handbook of communication and aging research, 31–67. Mahwah, N.J.: Lawrence Erlbaum Associates.
Doing age and ageing: language, discourse and social interaction
49
Obler, Loraine K. 1980 Narrative discourse style in the elderly. In: Loraine K. Obler and Martin L. Albert (eds.), Language and communication in the elderly: clinical, therapeutic, and experimental issues, 75–90. Lexington: Lexington Books. Obler, Loraine K. and Martin L. Albert (eds.) 1980 Language and communication in the elderly: clinical, therapeutic, and experimental issues. Lexington: Lexington Books. Ochs, Elinor 1992 Indexing gender. In: Alessandro Duranti and Charles Goodwin (eds.), Rethinking context: language as an interactive phenomenon, 335–358. Cambridge; New York: Cambridge University Press. Paoletti, Isabella 1998 Being an older woman: a study in the social production of identity. Mahwah, N.J.; London: Lawrence Erlbaum Associates. Pecchioni, Loretta, Hiroshi Ota and Lisa Sparks 2004 Cultural issues in communication and aging. In: Jon F. Nussbaum and Justine Coupland (eds.), Handbook of communication and aging research, 167–207. Mahwah, N.J.: Lawrence Erlbaum Associates. Poulios, Apostolos 2004 Constructing identities in the “periphery”: the case of the elderly. In: Christina Dokou, Efterpi Mitsi and Bessie Matsopoulou (eds.), The periphery viewing the world: selected papers from the fourth International Cconference of the Hellenic Association for the Study of English, 50–56. Athens: Parousia Pablications 60. Poulios, Apostolos 2005 The construction of elderly identity through conversational narrative. In: Marina Mattheoudakis and Angeliki Psaltou-Joycey (eds.), Selected papers on theoretical and applied linguistics from the16th International Symposium, April 11–13, 2003. 279–295. Thessaloniki: University Studio Press. Poulios, Apostolos 2008 Pain and joy in elderly discourse. In: Chryssoula Lascaratou, Anna Despotopoulou and Elly Ifantidou (eds.), Reconstructing pain and joy: linguistic, literary and cultural perspectives, 159–178. Newcastle: Cambridge Scholars Publishing. Poulios, Apostolos 2009 The flesh made text or the text made flesh? The construction of elderly identity through talk. In: Zoe Detsi-Diamanti, Katerina Kitsi-Mitakou and Effie Yiannopoulou (eds.), Bodies, theories, culture in the post-millennial era: selected proceedings, 149–161. Thessaloniki: University Studio Press. Radtke, Edgar 1990 Substandardsprachliche Entwicklungstendenzen im Sprachverhalten von Jugendlichen im heutigen Italien. In: Günter Holtus and Edgar Radtke (eds.), Sprachlicher Substandard III: Standard, Substandard und varietätenlinguistik, 128–171. Tübingen: M. Niemeyer. Rampton, Ben 1995 Crossing: language and ethnicity among adolescents. London: Longman.
50
Alexandra Georgakopoulou and Anna Charalambidou
Rawlins, William 2004 Friendship in later life. In: Jon F. Nussbaum and Justine Coupland (eds.), Handbook of communication and aging research, 273–303. Mahwah, N.J.: Lawrence Erlbaum Associates. Rickford, John, Arnetha Ball, Renee Blake, Raina Jackson and Nomi Martin 1991 Rappin on the copula coffin: Theoretical and methodological issues in the analysis of copula variation in African-American Vernacular English. Language Variation and Change 3: 103–132. Romaine, Suzanne 1984 The language of children and adolescents: the acquisition of communicative competence. Oxford: Blackwell. Schwitalla, Johannes and Juergen Streeck 1989 Subversive Interaktionen: Sprachliche Verfahren der sozialen Abgrenzung in einer Jugendlichengruppe. In: Volker Hinnenkamp and Margret Selting (eds.), Stil und Stilizierung, 229–252. Tübingen: Niemeyer. Settersten, Richard 2005 Linking the two ends of life: What gerontology can learn from childhood studies. Journal of Gerontology: Social Sciences 60B: 173–180. Shepard, Carolyn A., Howard Giles and Beth A. Le Poire 2001 Communication accommodation theory. In: W. Peter Robinson and Howard Giles (eds.), The new handbook of language and social psychology, 33–56. Chichester: Wiley. Spreckels, Janet 2006 “Britneys, Fritten, Gangschta und wir”: Identitätskonstitution in einer Mädchengruppe. Eine ethnographisch-gesprächsanalytische Untersuchung. Frankfurt am Main: Peter Lang. Stenström, Anna-Brita, Gisle Andersen and Ingrid Kristine Hasund 2002 Trends in teenage talk: corpus compilation, analysis, and findings. Amsterdam/Philadelphia: J. Benjamins. Tainio, Liisa 2002 Negotiating gender identities and sexual agency in elderly couples’ talk. In: Paul McIlveny (ed.), Talking gender and sexuality, 181–206. Amsterdam/Philadelphia: J. Benjamins Taylor, Bryan C. 1994 Frailty, language and elderly identity: interpretive and critical perspectives on the aging subject. In: Mary Lee Hummert, John M. Wiemann and Jon F. Nussbaum (eds.), Interpersonal communication in older adulthood: Interdisciplinary theory and research, 185–208. Thousand Oaks, California: Sage. Tertilt, Hermann 1996 Turkish Power Boys: Ethnographie einer Jugendbande. Frankfurt am Main: Suhrkamp. Wenger, Etienne 1998 Communities of practice: learning, meaning, and identity. Cambridge: Cambridge University Press. Westerhof, Gerben and Emmanuelle Tulle 2007 Meaning of ageing and old age: discursive contexts, social attitudes and personal identities. In: John Bond, Sheila Peace, Freya Dittmann-Kohli and
Doing age and ageing: language, discourse and social interaction
51
Gerben Westerhof (eds.), Ageing in society: European perspectives on gerontology, 235–254. London: Sage. Wilson, Gail 1991 Models of ageing and their relation to policy formation and service provizion. Policy and Politics 19: 37–47. Wyn, Johanna and Robert Douglas White 1997 Rethinking youth. New South Wales, Australia: Allen and Unwin. Ylänne-McEwen, Virpi 1999 “Young at heart”: discourses of age identity in travel agency interaction. Ageing and Society 19: 417–440.
2.
Gender identities and discourse Bróna Murphy
1.
Exploring gender and pragmatics
Although research on gender and language has been associated with pragmatics since the 1970s, it is, in fact, only since around the late 1990s (Cameron 1998) that explicit reference has been made to the word pragmatics in gender research. Such work reflects scholars’ growing preoccupation with how gender and context interrelate (Christie 2000, 2010; Mills 2003; Cameron 2005; Holmes 2006) and how notions of gender identity can be enhanced, for instance, by taking into account pragmatic notions (Mills 2002, 2003; Cameron 2005). Since the 1980s, approaches to language and gender studies have become more diverse and sophisticated (Coates 2003; Mullany 2010) and it is understood that insights from pragmatics have played a role in contributing to this, adding layers of complexity and sophistication. Christie (2010: 171) highlights that scholarship which has been carried out on pragmatics and gender, to-date, overlaps in two ways. Firstly, she states that conceptual tools developed within pragmatics have traditionally informed two key trajectories of gender and language research, a) that which asks how males and females use and interpret linguistic resources, which I will be concerned with in this chapter, and b) that which asks how males and females are represented through linguistic resources. However, she adds, if we take the central question that pragmatics asks to be “[w]hy has this utterance been produced?” (Haberland and Mey 2002: 1672) then gender and language scholarship, she writes, has, in turn, informed pragmatics in that it has consistently provided evidence of the many different ways in which any explanation of an utterance requires a systematic engagement with socio-cultural phenomena. In this chapter, I will explore the inter-relationship between pragmatics and gender, as mentioned by Christie (2010). Sections 1.1 to 1.3 will provide a brief overview of the role of pragmatics across the four waves of gender research since the 1970s (Deficit, Lakoff 1975; Dominance, Zimmerman and West 1975; Difference, Tannen 1992 and Dynamic, Mills 2003) while section 2 will highlight the role of Corpus Linguistics, as a methodology, in research on pragmatics. Both of these sections will provide a backdrop for the empirical corpus-based study which will be carried out in section 3 and which will explore the pragmatics of insults, in particular, across three age-differentiated sub-corpora of all-male discourse.
54 1.1.
Bróna Murphy
Pragmatics in gender and language research
Insights into gender and pragmatics date back to the time of Lakoff’s (1975) pioneering work. Christie (2010: 171) states that central to Lakoff’s (1975) argument that men and women use linguistic resources differently is an engagement with the function and context of utterances, which is central to pragmatics. She states that Lakoff (1975) draws, in particular, on Grice’s maxims (Grice 1968) as well as her own account of politeness (Lakoff 1973). Christie (2010: 171) explains that by conforming to politeness norms which are conceptualised as a set of culture-specific rules and learning “to speak like a lady”, women are, in effect, learning to use speech styles that lead them to be perceived by others and by themselves as powerless. Women’s language was portrayed, in this work, as powerless and was referred to as “lacking” (Lakoff 1975). Such an approach to viewing female discourse, which draws on intuitive insights based on knowledge of pragmatics, in the 1970s, gave rise to what is referred to as the deficit approach, which was a pioneering starting-point for gender and language studies and one of the first approaches which aimed at defining the relationship between gender and language. Lakoff’s (1975) work as well as early studies by Dubois and Crouch (1975) and Fishman (1983), among others, show that a focus on pragmatics, in order to understand why and how forms, such as tag questions and response tokens, for instance, are used across gender, was valued and influential. The following phase in gender and language research shifted to what was referred to as the dominance approach towards the late 1970s and the early 1980s and again makes reference to work in pragmatics, although, it seems to a lesser extent than what had gone before and what came after (the difference approach). This approach highlighted linguistic differences in women’s and men’s speech in terms of men’s dominance and women’s subordination (Zimmerman and West 1975). In this approach, issues of inequalities of power were at the forefront and the superior role of men in interaction was reflected in works such as Spender (1980) which was entitled “Man Made Language”. Zimmerman and West’s (1975) study took a more Conversation Analytical approach to isolating interruptions and counting them, but it did not take context into account. While they established that men interrupt more often and are therefore more dominant in interaction, calls from others such as Beattie (1982), for instance, highlighted the need to look beyond mechanical statistical evidence to a more pragmatics-driven analysis which would take into account the other possible functions underlying the interruptions. Again definitions and portrayals of gender looked to take into account the importance of context as a central factor if more accurate findings were to be produced (see Tannen 1993 as well as section 1.2) and misinterpretations were to be avoided. Overlapping from the dominance approach which explored gender differences in terms of inequality across males and females, the difference approach which began in the late 1980s and early 1990s, took a more expansive view of what “dif-
Gender identities and discourse
55
ferences” mean and looked beyond differences which were simply defined by the notion of dominance. The approach recognises the role of language as social practice (Coates 1986) and highlights how explorations of pragmatics and gender are mutually beneficial. While insights from pragmatics can enlighten gender and language studies (Coates 1986, 2003), theories of language and gender can also inform and explain pragmatics (Christie 2010). This third wave of research brought together more fully the areas of pragmatics and gender research than had been achieved before (see Tannen 1992, 1993). Tannen (1992: 24–25), for instance, made reference to insights from context and implicature stating that for women, conversations are negotiations for closeness in which they seek and give confirmation and support while men’s conversations are ones in which they try to achieve and maintain the upper hand if they can and avoid others’ attempts to put them down and push them around. Explanations for these findings drew on theories from language and gender such as Maltz and Borker’s work (1982: 199; see also Gumperz 1982), which claims that cultural differences exist between men and women and highlights the view that males and females are socialised into different subcultures throughout childhood. The view of gender which emerges in the difference approach, and which clearly has insights from pragmatics at the centre of its thesis, is a view of gender as a stable attribute and a part of males’ and females’ core identity which is normally acquired in the course of their early socialisation (Cameron 2005: 322). Insights from pragmatics as well as insights from language and gender research, it can be said, have shaped and influenced the view of gender put forward in the difference approach. Indeed, Cameron (1998) writes that although differences in the way that men and women use language, as reviewed earlier, can, indeed, lead to miscommunication between sexes, a focus on pragmatics, through the foregrounding of the inferential nature of communication can better explain the differences between male and female talk in utterance interpretation. Christie (2010: 172) writes that Cameron’s (1998) work is one of the first to signal a move away from conceptualising gender in terms of differences and positing the existence of a stable set of qualities that pre-exist and indeed generate differences in the use and interpretation of language. More recently, research on gender and language has given rise to another approach – the dynamic approach, which is more social in nature and which provides social constructionist perspectives of gender as a dynamic, fluid notion within language and gender studies (Mullany 2010: 225). Mullany (2010: 225) highlights, in particular, how Butler’s (1990, 2004) conceptualisation of gender as a performative social construct has come to dominate the field. She adds that the application of social constructionist models where identities are conceived as pluralised, negotiated and contested categories has resulted in the production of more intricate understandings of the interplay between [interpersonal] pragmatics, and the negotiation and performance of gender identities within interaction (Mills
56
Bróna Murphy
1997; Sunderland 2004; Crawford 2000; Mullany 2007). This direction has, in turn, shaped the current definition of what gender means and has enhanced our knowledge of gender identity. Mills (2002, 2003), for instance, reinforces the social constructionist approach and highlights the important role of pragmatics in providing a view of gender that moves away from a reliance on binary oppositions and global statements about the behaviour of all men and all women, to more nuanced and mitigated statements about certain groups of women or men in particular circumstances (Coates and Cameron 1988; Johnson and Meinhof 1997), who negotiate within certain parameters of permissible or socially sanctioned behaviour. Mullany (2010: 226) adds that while gender is a highly significant aspect of our social identity construction and performance, there is also a range of other crucial facets of our sociolinguistic identities which interact with gender construction, such as class, sexuality, ethnicity and age. This chapter will, in particular, explore the role of age as a variable alongside gender in later sections and discuss how it may shed light on the construction of gender identity. Although this discussion presents a clear-cut chronological theoretical account of gender and language, it does so in order to illustrate the role of context in shaping definitions of gender over the past forty years. It is not suggesting the importance or usefulness of one approach over another, as it is often not possible to separate them with two or even more of the approaches occurring simultaneously (Mullany 2010: 229; see section 3). Mullany (2010) adds that findings from the early work on gender can provide researchers with a useful starting point for conducting research into pragmatics and warns against the dangers of clear-cut accounts of gender research which may obscure this point. A review of the four waves of gender research, presented above, shows, for the most part, that there has always been a close connection between gender and pragmatics. Christie (2010: 172) claims that re-theorisations of gender has resulted due to questioning notions of difference in language and gender studies and has followed a path that parallels one within pragmatics. She states that just as pragmatic accounts have increasingly taken on board the extent to which communication is achieved through the dynamic construction of meaning via context-specific uses of language, so current research sees gender as a product of specific uses of linguistic and pragmatic resources, rather than a determinant of those uses. Cameron (2005: 332) states that this is perhaps the most important insight pragmatics has to offer “in that going beyond the data” or bringing “global assumptions to bear on local instances”, analysts need not be thought of as imposing their own procedures. Christie (2010: 172) makes reference to West and Zimmerman (1987) who argued that gender should be perceived as a verb rather than a noun: by engaging in particular linguistic and non-linguistic behaviours, individuals are actively “doing gender” (cf. Georgakopoulou and Charalambidou this volume on “doing age”). Eckert and McConnell-Ginet (1992) take the idea, Christie (2010: 173) states, that gendering is at least in part a linguistic practice and develop it further by linking it with a
Gender identities and discourse
57
model of context and communication that has proved to be extremely influential in the study of how men and women use language. Christie adds that the more recent developments in the theorisation of gender are perceived as capable of providing a more nuanced and explanatory account of variation in language use since they provide a model for addressing differences related to the individual such as class, race and age, some of which I will focus on later, as they emerge as relevant aspects of context in a given interaction. 1.2.
Gender, pragmatics and methodologies
Since the 1970s, studies on the exploration of gender and pragmatics have tended to draw on a range of different methodologies and frameworks. As highlighted by Bucholtz (2003: 43), the study of language and gender has increasingly become the study of gender and discourse and as a result, various perspectives on what discourse means has influenced the types of frameworks used to explore gender. Two of the most featured frameworks for exploring gender and discourse, in recent times, which feature commonly in the literature include Conversation Analysis (CA), and Critical Discourse Analysis (CDA); although not yet as common and not included in Bucholtz’s (2003) chapter, Corpus Linguistics is also becoming increasingly more common in studies of gender and pragmatics (Beeching 2002; Murphy 2010) and will be reviewed alongside the others below. For instance, CA (Sacks, Schegloff and Jefferson 1974) involves a detailed and systematic micro-analysis of spoken extracts which focus on the details of the interaction and on the analysis of only those social categories, such as gender, that are manifestly oriented to by the speakers in their discourse (Schegloff 1999; Chouliaraki and Fairclough 1999: 7; Stokoe and Smithson 2001; Speer 2005; Litosseliti 2006). Seedhouse (2004) states that “although CA’s main interest has been in how social acts are performed through language, it has always been interested in … the domain of pragmatics” (see Schegloff 1997). The use of pragmatics in association with CA, for instance, illustrates how exploring features of CA, in context, is essential in order to understand how they are being used by the speakers. Such an analysis takes into account the social structures which are paramount for analyses of gender and language (Stokoe and Smithson 2001; Weatherall 2002; Sunderland 2004; Cameron 2005) in order to produce rich understandings of the links between gender and discourse. Looking beyond the text to explore the related discursive practices, social issues and social relations of power (Litosseliti 2006: 54), research on language and gender has also used a CDA approach (cf. Wodak this volume). CDA is often drawn upon in order to show how language contributes to the representation and possible social construction of men and women (Mills 2003; Wodak 1997; Baxter et al. 2010). It is suitable for the studies of pragmatics and gender as it sees language as social practice (Fairclough and Wodak 1997) and considers the context of
58
Bróna Murphy
language use to be crucial (Weiss and Wodak 2003). Although it has been heavily criticised (Widdowson 1995; Schegloff 1997), Litosseliti (2006: 55) highlights its usefulness for the analysis of less obvious and implicit meanings behind spoken and written texts (including visuals and gestures). As a result, CDA has been particularly enlightening in relation to studies of language and gender as “many basic assumptions of feminist linguistics relate to and overlap with principles of critical linguistics and CDA” (Wodak 1997: 7). Used within frameworks of CA and CDA, more recently, studies have begun to provide gender-related insights into how language is used when explored from a Corpus Linguistics’ perspective (Beeching 2002; Holmes 2006; Murphy 2010). The usefulness of corpus linguistic tools and methodologies when exploring pragmatics and discourse in general has been widely documented (Andersen 2001; Stenström, Andersen and Hasund 2002; Aijmer 2002; Aijmer and Stenström 2004; Stenström 2006; Romero Trillo 2008; Adolphs 2008; Farr and Murphy 2009). Rühlemann (2010) also provides an interesting account of how corpora can be used in studies of pragmatics and highlights the merits and challenges of such an approach (see section 2.1 for full discussion). 1.3.
Gender, pragmatics and social contexts
Since the 1970s, literature on gender and pragmatics also reflects investigations which take place in a variety of different social contexts such as the classroom (Sunderland 2004), advertising (Velasco-Sacristán and Fuertes Olivera 2006), the workplace (Holmes 2006), as well as the media (Benwell 2003), among others. The context of casual conversation between friends and family, however, has been one of the most researched contexts as conversation is “the prototypical kind of language usage” and has “naturally been at the heart of pragmatics research” (Rühlemann 2010: 293). Conversations have been looked at both in terms of female talk where an abundance of research exists (see Coates 1989, 1998), male talk (Cameron 1998; Coates 2003; Benwell 2003, Pichler and Eppler 2009) although to a much lesser extent (see section 1.3.1) as well as mixed gender groups (Coates 1989; Eggins and Slade 1997; Pilkington 1998, Lampert and Ervin-Tripp 2006). As I will be exploring male discourse and swearing below, the following section will briefly review men’s talk, in particular, in relation to research on the use of insults. 1.3.1.
Men’s talk and the pragmatics of insults
Strides are being made to compensate for the lack of coverage that male discourse has received in comparison to females over the past decade (Coates 2003; Baker 2005; Pichler and Eppler 2009). Central to the investigation of male discourse in the recent literature is an association with swearing (Coates 1993; de Klerk 1990,
Gender identities and discourse
59
1991, 1997, 2005). De Klerk (1991) as well as Benwell (2001) and Coates (2003) all claim that swearing is a commonly accepted, and expected (see Stapleton 2003; 2010), way of presenting a “masculine” identity and also functions to build male (in-group) solidarity (Johnson and Meinhof 2003). Coates (2003: 104) claims that men tend to express solidarity with each other through the use of linguistic strategies such as swearing, ritual insults, sexist and homophobic remarks as well as competitive banter, while avoiding discussing personal topics and engaging in mutual self-disclosure (see Coates 1998). In section 3, this chapter will look in particular at insults, as a category of swearing across three life-stages of males in order to understand further why they engage in such a ritual and what it means in terms of male gender identity (see Andersson and Trudgill 1990; Allan and Burridge 2006; for definitions of swearing). Insults1 will be defined here as “emphatic manifestations of our intentions, evidence of our feelings and very often a clear exhibition of our aggressive nature” (Mateo and Yus 2000: 98). Insults: 1) are a sort of catharsis, a tension relief when humans feel stressed or in need to release a high emotional strain 2) require close interaction between the participants 3) can be uttered but without deliberate intention to insult anybody and function in a phatic way, for instance, as a sign of camaraderie Before focusing on the pragmatics of insults in terms of gender, the following section will provide details about the data and methodologies being used in this study.
2.
Describing the corpus
To investigate commonly-occurring insults, I use a spoken corpus (142,467 words) which includes male and female conversations across a range of age groups and which will be referred to as CAG-IE (Corpus of Age and Gender – Irish English) hereafter. The corpus has been compiled across gender into a Male Adult Corpus (MAC) and a Female Adult Corpus (FAC) and within those groups organised according to a number of age groups into three c. 15,000 word sub-corpora which represent three life-stages: the 20s/30s (young adulthood), the 40s/50s (mid-adulthood) and the 70s/80s (old adulthood) cohorts. The participants are largely made up of friends and family members. The recordings take place in various locations in the Republic of Ireland (Cork, Limerick, in particular, see Murphy 2010). All of the interactions are casual conversations and range in topics from the weather, politics and university life to local events and holidays. The methodology employed in this research is a combined quantitative and qualitative corpus-based discourse analysis. Frequently-occurring items are isolated using appropriate computer software WordSmith Tools (Scott 2008) and are examined using frequency
60
Bróna Murphy
lists, concordances and contextualised extracts. The main focus of the analysis is to investigate the high-frequency insults being used and to understand the why and how behind the men’s usage in every-day interactions. Interviews were carried out after each of the data collections and will also be used below as a form of data (see sections 2.1, 3.3 and 3.4). The study highlights the pivotal role of follow-up interviews in interpreting and fully understanding the corpus-based findings. 2.1.
Pragmatics and corpus linguistics: problematic issues
Carrying out research into pragmatics using a corpus-based approach is very useful (Adolphs 2008) but it can give rise to a number of problematic issues as highlighted by Rühlemann (2010). Rühlemann (2010: 289) discusses five potential issues: firstly, he discusses the fact that corpora simply record text and only provide a crude record of the context within which the text is produced. Secondly, he highlights how spoken corpora are based on transcriptions which may include stretches of inaudible or unintelligible data which as a result, tear the co-textual web apart preventing the analyst from understanding what is going on. Thirdly, he adds, that as analysts, we are in a less privileged situation than the participants who produced the text which impacts on the interpretation of the data. Finally, he states that as analysts, we are not the intended recipients but merely a kind of eavesdropper cut off from the wealth of background knowledge the participants share. As a result, we do not have access to the participants which would allow us to “negotiate” meaning and gain further elaboration and clarification (Rühlemann 2010: 290). While these issues are certainly problematic when it comes to large corpora, I have found them to be less significant, on this occasion, when dealing with a smaller corpus such as the CAG-IE which I compiled. Working with a small corpus, as Koester (2010) states, means that the analyst has familiarity with the corpus and as a result, s/he has access to background information to aid in the interpretation of the data. This is particularly true in the case of CAG-IE, where I knew the people taking part in the recordings as well as understood their socio-cultural background. Such insights into the social and cultural contexts, as a result, renders us, as Flowerdew (2008) highlights, less cut-off from the background knowledge and relevant contexts which is often a problem when dealing with larger corpora. With smaller corpora, there is also greater likelihood of having greater access to the participants. Certainly in the case of CAG-IE, the participants have been accessible which has allowed me, as the compiler, to seek clarification and further information through conducting informal semi-structured interviews. The informal interviews, ranging in duration from 15 minutes to one hour, were carried out after each of the 20 recordings and will form an additional dataset which will be drawn upon later on in this chapter. In terms of capturing context, as discussed also by Rühlemann (2010), there is no doubt that this is problematic as background noise or other issues can render ut-
Gender identities and discourse
61
terances inaudible or unintelligible. However, in the case of CAG-IE, while such problems were encountered sporadically in the transcription, (see Murphy 2010), it was rare that large stretches of utterances were missing. In addition, with regards to capturing context, as both the compiler and the analyst collecting data in a context to which I belonged, I was very familiar with the background knowledge of the context (and at times the conversation topics), the culture, and the people. This knowledge also came to bear when carrying out the analysis in that being a member of the community under scrutiny, as highlighted by Jaworski and Coupland (1999: 36) allowed me to often enter the private worlds of the speakers and gain an insider’s understanding, which is not necessarily possible when dealing with a large corpus. In this section, I put forward that the problems encountered when carrying out pragmatics research using a small corpus, that has been compiled by the analyst, are, often, more manageable. Despite certain problems when carrying out pragmatics research using a corpus (as addressed above), there are clear benefits to using corpus linguistics. For instance in light of the research analysing multifunctional elements (Rühlemann 2010: 290), a corpus is useful in providing the analyst with many examples which are authentic and which are also embedded in their co-texts therefore providing some evidence of the context in which they were used. Concordance lines shed light on lexico-grammatical patterning, but insights into what is taking place can also be uncovered by looking at extended concordance lines which provide access to stretches of discourse as well as to the greater text (see section 3.2).
3.
Investigating insults in CAG-IE
The analysis that follows in the next section will explore a microcosm of the pragmatics of society: all-male naturally-occurring interactions in close friendships. It will investigate the pragmatics of insults, which are a high-frequency item in the men’s discourse. The analysis consists of two main parts. The first section will provide a quantitative account of the insults in terms of gender differences, while the second part will look in depth at how the forms are used by two particular age cohorts. 3.1.
Insults and gender differences – A quantitative analysis
In order to establish insights into the use of insults in male and female discourse, frequency lists were run, using WordSmith Tools (Scott 2008) so as to isolate commonly-occurring forms. Each of the forms was then viewed in context in concordance lines so as to determine if the forms were being used as insults. From these analyses, 30 abusive forms, which make up six groups of insults across both the MAC and FAC (see Figure 1), were uncovered. (1) Mental insults (nuts, mad, lu-
62
Bróna Murphy
natic, crazy) are concerned with mental health problems. (2) Social status slurs (scumbag2, pathetic, scab3, hash-head, loser, square) are forms used to refer to one’s lowly status in society. (3) Sexual insults (fuck/fuck off, fucker, bitch, bastard/s, cunt, prick/s, bollocks, fecker4, assholes) are concerned with sex, sexual practices and body parts. (4) Homosexual insults include the use of the forms gay and homosexual to refer to heterosexual men. (5) Intellectual insults (stupid, idiot, eegit5, dope6, plonker7, simpleton, thick, dumb, gom8) refer to one’s low level of intelligence. (6) Animal insults (donkey) refer to humans in terms of their likeness to animals and their traits. Occurrences of the forms functioning in this way have been converted to words per million (wpm). From the figure, it is clear that abusive forms in the Irish English context tend to be predominantly associated with male discourse. There are 810 occurrences of the forms by the men in comparison to 497 by the women. The categories most frequently used by the males are sexual insults (467 occurrences), which are by far the most common, followed by intellectual insults (213 occurrences), homosexual insults (64 occurrences), social status slurs (22 occurrences) and animal insults (11 occurrences), which are the least common. Although used noticeably less often by the females, there is evidence of the women’s use of sexual and intellectual abusives; however the categories they use most commonly and more frequently than the males are mental insults and social status slurs. Interestingly, there are zero occurrences of homosexual slurs and animal abusives in FAC. These figures suggest that the men value sex and intellectual insults more than the women in their interactions while the women instead show preference for insults which could be considered to be less harsh, for instance, social status slurs, mental insults and references to animals. It is interesting with regard to the sex and intellectual insults, that not only do the women use the forms infrequently but they also demonstrate a very limited range (stupid and gom) in comparison to their male counterparts who use stupid, idiot, eegit/s, dope, plonker, simpleton, thick and gom. On the whole, these findings concur with the literature which has established that traditionally women use swearwords less often than males (de Klerk 1991, 1997; Coates 2003; Jay 1992, Jay and Janschewitz 2008; Murphy 2010), and women, as highlighted by Mateo and Jus (2000: 102), insult less than men even in highly egalitarian societies. The initial quantitative findings show the use of insults to indeed be an (in-) group activity which is associated with masculinity and a powerful resource for male identity construction (Stapleton 2010: 297). However, this is not entirely new and what I would like to focus in on in more depth is how the insults are used across different age groups of men and women. Looking beyond gender and including a variable such as age sheds more light on the differences across the age groups in terms of how frequently they use insults (see Figure 2). Figure 2 shows that across all of the age groups, the MAC 20s/30s are, by far, the most frequent users of the abusive terms. The graph illustrates that although the
Gender identities and discourse
63
Figure 1. Most frequently occurring groups of insults in CAG-IE
Figure 2. Distribution of insults across age and gender variables
females use the forms considerably less, it is, nonetheless, the 20s females who are also appearing on the graph as the most frequent users of the forms in their gender. What is enlightening about these findings is not necessarily that the insults are more frequent in the MAC 20s/30s, as swearwords and taboo language have been characterised as a feature of young male discourse across Englishes (de Klerk
64
Bróna Murphy
1990; Coates 2003; Murphy 2010), but that an exploration across different age groups or indeed life-stages of adulthood seems to provide us with more in-depth fine-grained insights into different gender identities. While the 20s/30s’ high frequency usage dominates the graph, it is clear that the 70s/80s barely feature on it at all. It would also seem that the use of insults is not a high frequency aspect in the discourse of men in the 40s/50s age group. Tentative suggestions may be that at this 40s/50s life-stage, they are fathers and/or professionals or indeed see themselves as more mature men, as was highlighted in the interviews and the use of insults is no longer befitting of their stage in life. The older men say that they never engaged in “that kind of talk” because they were not “brought up like that” and it is linked to religious faith and Catholic upbringing. From this picture, it is clear that the 20s/30s men, in particular, engage in a way of doing interaction that is different from the other two groups of males. The finding also highlights that while swearing is indeed a characteristic of masculinity, it cannot be attributed to all males, as there are fine-grained differences as the age analysis reveals. Looking in even more depth and focusing on the 20s/30s males and how they use insults in the three most frequent categories of insults (see Figure 3), there is evidence of even further variation. From Figure 3, it appears that the 30s use sexual insults slightly more often while the 20s men use intellectual and homosexual references almost predominantly. There are zero occurrences of homosexual references in the 30s’ men’s discourse and very few instances of intellectual insults. This observation indicates that even within the young men’s discourse, there is evidence of variation with regard to the groups of insults they engage in and the frequency with which they are used. These initial quantitative insights suggest that the consideration of other variables alongside gender clearly allows for a more holistic picture of gender and one which seems to indicate a nuanced definition of male identity which would support
Figure 3. Sexual, intellectual and homosexual insults in MAC 20s and MAC 30s
Gender identities and discourse
65
the view of gender as a dynamic, fluid notion (Mullany 2010: 225) and not a fixed static construct as highlighted by the early approaches. Such a view of gender as pluralised, as the findings suggest here in relation to the use of insults, moves away from binary thinking that there is simply one singular monolithic version of masculinity or indeed femininity (Mullany 2010: 232). Moving towards an exploration of the pragmatics of sexual, homosexual and intellectual insults reveals that functional variation also occurs in the men’s use of the forms. 3.2.
Insults in MAC 20s/30s
An exploration of the full set of concordance lines for the insults reveals that the 20s men’s insults are used primarily in direct reference to each other (see lines 1–4 for examples of intellectual and homosexual insults). They tend to use them to make fun of each other, which illustrates a very social function which is linked to camaraderie and social bonding. On the other hand, the 30s men do not use the forms to insult each other directly but prefer to use them about absent third parties with the main focus of creating humour. Their usage of the insults is commonly accompanied by instances of laughter (see Figure 4 lines 5–8 for examples). 1 2 3 4 5 6 7 8
three years old Come on ya You’re so listening to the tape you’re not Ho ha ha here’s our weight since I got him he is a fat for food (phone) he’s a hungry hold means in America it means be happier. what about the other
idiot dumb gay. homosexual cunt bastard fuck off fucker
You’re so man Lads in fairness like shut up buddy (laughter) he’s we got him and he w I have a conference call that I need t and leave me alone (laughter) then? Ha (laughter) Is she fuckin
Figure 4. Intellectual, homosexual and sexual insults in MAC 20s/30s
3.3.
Insults in MAC 20s – A qualitative analysis
Figure 4 shows examples of the 20s men’s use of insults (lines 1–4) as interactionoriented devices which are not meant to insult but are intended to signal camaraderie (see Coates 2003) as well as to establish and reinforce group identity among the men (see de Klerk 1991). The men’s interaction is characterised by informality, solidarity and lack of status differences (Montagu 2001). The following three extracts (taken from extended concordance lines) will illustrate how intellectual, homosexual and sexual insults are being used in terms of their pragmatic functions in the 20s men’s talk.
66 3.3.1.
Bróna Murphy
Intellectual insults in MAC 20s
In the first extract, the 20s men, who have lived together in a student house for five years or more, are discussing how Tom’s girlfriend (she), who is not present, has to repeat a university exam. In line 4, Diarmuid asks whether she is thick (meaning unintelligent) or something because she has failed a university exam. Extract 1. The use of intellectual insult thick in MAC 20s Kyle: But shure9 she has to repeat exams anyway doesn’t she? Tom: Yeah. Kyle: Because she was sick. Diarmuid: Is she thick of something? Kyle: Has she got a problem? Tom (and others) laughing < $E> Tom: I’m not laughing I’m laughing at his ignorance. Kyle: Shure10 don’t mind us we’re only joking anyway.
1 2 3 4 5 6 7 8
This extract is interesting because on the surface, it seems that the insult is referring to Tom’s girlfriend and is questioning her intelligence, however, if we look at the extended co-text, it is clear that the insult is firstly not intended to offend and secondly not directly related to Tom’s girlfriend. In fact, it seems that the insult is used in order to tease Tom. Kyle and Diarmuid take turns (for example, lines 3 and 4) to wind Tom up. It is clear that Tom recognises the intentions of his interlocutors in the interaction and understands that they are having fun at his expense. He finds the situation comical and in line 6, there is evidence of him laughing as well as in line 7. In line 7, he wants to make it clear, although he does not state it explicitly, that he is not laughing because they are being “offensive” or because he believes what they are saying about his girlfriend to be true, but because, as he says, his friend is being ignorant. The use of ignorant, in this instance, in an Irish English context, means rude in an explicit way for humorous effect but which borders on the unacceptable. He does not want to be seen as being disloyal, perhaps, to his girlfriend. It is clear, from the interaction, that the men believe that it is not because Tom’s girlfriend is unintelligent that she has not been able to sit her exam but because she has been ill. Kyle in line 8 adds that they are joking, possibly for the benefit of the recorder. This extract illustrates a way of doing humour which involves insulting but without the intention to offend (Mateo and Yus 2000). Although the insult involves Tom’s girlfriend, and appears as if the insult is directed at an absent third party, it becomes clear, in fact, that it is aimed at Tom and is typical of the jocular teasing men engage in and that the literature refers to (Coates 2003). The men are tending to the social dynamics among the group by teasing Tom. Their use of the intellectual insult is interaction-centred and signals that the men are bonding and engaging
Gender identities and discourse
67
in camaraderie which is regarded positively in male friendships. In interviews, Tom highlights that this kind of interaction is very common and it is how they interact. He says that winding each other up is an important part of their relationship. 3.3.2.
Homosexual insults
In the second extract, the 20s men, Colin and Joe, who also live in the student house, use the homosexual abusive, gay, to tease their friend Tom (Murphy 2010: 203). Extract 2. The use of homosexual insult gay in MAC 20s Colin: Tom: Colin: Joe: Tom: Colin:
I bet ya nobody in the town knows that Tom is gay I know yeah yeah yeah Why what do you make of it? You’d think he’d have come out by now like Mmh He’s got a boyfriend up here and all like …
1 2 3 4 5 6
Interestingly in the interaction, Colin begins the banter by talking about Tom in the third person as if he were not present. In line 1, he states I bet ya nobody in the town knows that Tom is gay. In line 2, Tom acknowledges the banter and receives it in the spirit that it is being given which is not offence-centred but very much interactionorientated (Mateo and Jus 2000). He agrees and repeats the form yeah three times which indicates that he is acknowledging listenership, has not taken offence but is perhaps tiring of the banter from Colin, which began a number of turns before. His friends Colin and Joe, however, continue to tease him by interacting among themselves (see line 3) about Tom who is still present. The function here would seem to continue the teasing by excluding him from the conversation. Tom reveals that he is very aware of what his friends are doing in line 5 when he uses the response token mmh which is accompanied by laughter. The banter (see line 6) continues from Tom’s friends and a few turns later, Joe refers to Tom as our homosexual buddy which again is framed by more laughter. The teasing and winding up of their friend is possible because they know he is not dealing with any sensitive issues about his sexuality. They are tending to their relationship by providing amusement and comic value and focusing on one of the friends to allow this to happen. This seems to be part of their in-group identity. Also underneath this abusive is the fact that being gay is seen as an insult to a heterosexual young man and thus it shines a light on the socio-cultural values within this group of young men. It is also interesting the way that extracts 1 and 2, as well as 3 below, focus on Tom as the source of amusement. In interviews, the young men highlight that they enjoy making fun of Tom as it usually provokes a reaction of some kind. This perhaps suggests that the men recognise certain roles within their friendship dynamics and also play to those roles.
68
Bróna Murphy
3.3.3.
Sexual insults
The third extract is the direct use of the sexual insult fucker by Colin to his interlocutor Tom. In this extract, the men are discussing an acquaintance who used to play football with them. Extract 3. The use of sexual insults fucker MAC 20s Tom: Colin: Tom: Colin: Tom: Colin: Tom: Colin: Tom: Colin:
Ah God they were mingin their second team like they had this Jamaican playing for them Yeah? Yeah How was that? I dunno there must be What did he look like? I think it was his first ever game Besides being black ya racist fucker I know you wanted to say it but I’ll say it for you Anyway I dunno shure11 Were they calling him Valentine?
1 2 3 4 5 6 7 8 9 10 11 12
Both men know that the acquaintance is black. Nevertheless, Colin deliberately asks Tom what he looked like (line 7). As Tom, it seems, is trying to be politically correct because the recorder is running, he overtly ignores the question (line 8) at which point Colin jokingly begins to tease him about being racist (line 9). Again Tom attempts to change the subject (line 11). This extract shows how the shared knowledge between these men allows for this interaction to take place. Colin has known Tom for more than a decade and he understands him and his personality. Colin knows, in this case, that Tom will feel awkward about saying that the acquaintance they are talking about is black and is teasing him and testing him by asking what did he look like? Both men know what the answer to the question is but Tom avoids it, by adding new information on the situation and changing the topic (see lines 8 and 11). Tom’s avoidance to say that the acquaintance was black could be regarded as a subject that he regards to be taboo and one he is not comfortable discussing because he is not sure of what vocabulary to use or what to say exactly. This reaction could be seen to reflect his own socio-cultural background12 and one which Colin is aware of and makes an effort to exploit for his and Tom’s own amusement. These extracts show the men interacting with each other and engaging in banter which is a characteristic of male interaction. The direct use of insults function in a humorous interactionally-centred way and no offence is intended. It is also clear that no offence is taken. Given the context, the participants involved, the relationship as well as the expected roles in this relationship, the situations have been clearly interpreted and the pragmatic meanings behind the insults have been under-
Gender identities and discourse
69
stood in light of the background information against which the men are processing the interaction. 3.4.
Insults in MAC 30s – A qualitative analysis
In the 30s men’s discourse, there is a different trend to what we see from the 20s men. Here the men predominantly use sexual insults, however, all of the referents of these insults are absent third parties and the main function seems to be to create humour through the use of what may be considered to be strong taboo forms. None of the occurrences are used to refer directly to the interlocutor and there seems to be no attempt to insult each other or directly tease each other. 3.4.1.
Sexual insults
Extract 4 shows the use of the abusive cunt in context and involves Darrell (30s) discussing the medication his sick cat was recently prescribed. Extract 4. The use of sexual insult cunt in MAC 30s Darrell:
Dean:
Yeah, it’s crazy yeah. So this one, the warnings that came with it … because we get it from the normal chemist, the normal prescription chemist and it says em you know, em may … like side effects include em anorexia and suicidal tendencies![both laughing heartily] It’s like how the fuck is that going to turn out in a cat? Is he going to want to throw himself off the deck? So it was just kind of interesting thinking about that. But it’s good that he might be a little anorexic because he’s … he’s literally doubled in weight since we got him … he is a fat cunt! [laughing – Dean and Darrell] … Yeah, but that’s your fucking fault
1 2 3 4 5 6 7 8 9
Darrell’s use of the abusive in the phrase he is a fat cunt (line 9) provokes positive judgement in the form of laughter from his interlocutor, Dean, as well as from the speaker himself. The abusive is purely an example of social swearing (Montagu 2001) used, it seems to create humour by the very fact that it is a strong form as well as the fact that the cat is being discussed in terms of its weight which is usually a topic associated to humans. It is clear from the interaction that the use of the form has not offended Dean, Darrell’s interlocutor. In interviews with Darrell and Dean, they highlight that they have known each other for a long time and that humour, levity and harsh taboo language are characteristic of how they usually behave. Dean states that in addition to the anthropomorphic references, the story is embellished by the use of the taboo form cunt and without it, the story would not have been as humorous. Dean adds that the use of such forms is almost expected when they interact with each other, which suggests
70
Bróna Murphy
that the laughter and positive judgement play a role in reinforcing their views and expectations of each other and the identity of their friendship and what they understand it to be. In light of this information, it is plausible to suggest that Darrell knows what Dean finds funny and plays that particular role which both men have acted in for many years, and which characterises their relationship. The use of the abusive form seems to be acting almost in the role of positive politeness. Knowing and recognising this characteristic of their friendship, Darrell nurtures it and they bond over the humorous story, as highlighted in the interview. Interestingly, during the interview, Dean states that he recognises the form cunt as a very strong taboo form and one that he would not use with some other friends (even though they have known each other as long as he and Darrell). He states that it is not taboo for either him or Darrell in their interactions because, he says, the way they interact is “different”. What Dean seems to be suggesting is that even within friendships, men do friendships differently depending on how they see their interlocutor. Not all male friendships are characterised by the use of insults or strong taboo forms (as established earlier). He indicates that choices are made about personality and what the other person will find funny or not. Levels of offence, on whatever personal subjective scale, are taken into account. With Darrell, however, he states that they are on the same level in terms of humour and the use of taboo forms to create humour, and this is an expected characteristic of their friendship. Here the findings reveal the role of insults as signs of male friendship identity which are acted out differently from the predominant trends in the MAC 20s. The findings reveal the dynamic nature of the male friendships in that through the interviews, in particular, we find that they associate certain ways of interacting with certain friends and not with others. It supports a finding that not all 30s men interact and display male identity by engaging in insults and taboo language. While earlier quantitative findings showed inter-group differences in male identities across the age cohorts, our interviews would seem to suggest that intra-group differences also exist within the 30s cohort and that the men, certainly in the case of Dean, seem to be aware of the differing nature of such identities within friendships. It is important to highlight that the fact the 30s men do not use insults directly to each other may be due, in this instance, to the fact that the men are at different stages of their lives than the 20s men. The 30s men are professionals while the 20s are students and this may be a factor in determining their interaction patterns. Also their non-usage may be reflective of them as a group and the way Dean and Darrell, in particular, interact. As Dean highlighted, men interact in different ways with different friends and reasons for such interactions are more complex than we think. The study is not suggesting that the 30s men do not engage in insults with each other on other occasions, contexts or in different situations, however, in order to explore this further, a larger corpus would need to be investigated.
Gender identities and discourse
4.
71
Conclusions
This chapter has explored links between gender and pragmatics using a corpusbased approach as well as insights from interview data. The analysis puts forward a view of gender, which straddles both the difference and dynamic approaches. The study provides insights into how not all male discourse and not all descriptions of male identity are characterised by swearing and taboo forms. The chapter shows that the use of insults in male talk is more nuanced and more complex than it seems and highlights that both inter- and intra-variation exist across different male age cohorts, both in terms of the frequency of the forms and their pragmatic functions. The chapter portrays the existence of male gender identities, in relation to the use of insults, which come to the surface when variables, other than gender, for instance, are considered. It also illustrates that even within age cohorts such as the MAC 20s/30s, insights into varying identities also reveal themselves. I conclude that these identities are infused and influenced not only by gender and age but also the men’s expectations and perceptions of their friendships, shared knowledge of each other and their roles in the friendship group-identity which are all connected and play a part in what the men say, how they say it, to whom they say it as well as how far they go. The chapter puts forward that care is needed when making generalisations about male identity and highlights the importance of considering issues of gender alongside other variables, such as age as well as more individual factors which are more characteristic of the particular participants, interaction and/or context.
Notes 1. See Culpeper (2011) and his contribution to this volume for definitions of impoliteness. 2. refers to a person who lacks integrity, honesty and other values deemed important in basic human interaction in society. 3. refers to a person who lacks integrity, does not work and is a drain on society. 4. is a weak form that is typical of Irish English and is used to refer to someone who has caused mild annoyance; in some contexts, it is sometimes regarded as a weak form of the stronger word “fucker”. 5. is typically used in an Irish English context and refers usually to someone whose actions are “silly” and/or considered to reflect a lack of intelligence. 6. refers to someone who lacks intelligence usually in terms of social etiquette. 7. refers to someone whose behaviour is considered to be reflecting a lack of intelligence. 8. is typically used in an Irish English context and is a shortened version of the form “gombeen”. It is sometimes used to refer to one’s lack of intelligence in terms of limited life experience or lack of social etiquette. 9. Shure is used, in Irish English, as a discourse marker to introduce a topic and also functions as a downtoner.
72
Bróna Murphy
10. Shure is used in line 8 as a hedge along with “don’t mind us” “we’re only joking anyway” 11. Shure is used as a hedge in this context alongside the use of cluster “I dunno” (see Murphy 2010) 12. Tom grew up in a rural Irish setting which would not have included mixed nationalities in the 1980s.
References Adolphs, Svenja 2008 Corpus and Context: Investigating Pragmatic Functions in Spoken Discourse. Amsterdam: John Benjamins. Aijmer, Karin 2002 English Discourse Particles: Evidence from a Corpus. Amsterdam: John Benjamins. Aijmer, Karin and Anna-Brita Stenström (eds.) 2004 Discourse Patterns in Spoken and Written Corpora. Amsterdam: John Benjamins. Allan, Keith and Kate Burridge 2006 Forbidden Words. Cambridge: Cambridge University Press. Andersen, Gisle 2001 Pragmatic Markers and Sociolinguistic Variation-A Relevance-Theoretic Approach to the Language of Adolescents. Amsterdam: John Benjamins. Andersson, Lars-Gunnar and Peter Trudgill 1990 Bad Language. Oxford: Basil Blackwell. Baker, Paul 2005 Public Discourses of Gay Men. London: Routledge. Baxter, Judith, Ann Weatherall, Maria Stubbe and Jane Sunderland 2010 Conversation analysis and critical discourse analysis in language and gender research: Approaches in dialogue. In: Janet Holmes and Meredith Marra (eds.), Femininity, Feminism and Gendered Discourse, 213–245. Newcastle: Cambridge Scholars Publishing. Beattie, Geoffrey 1982 Turn-taking and interruption in political interviews. Semiotica 39(1–2): 93–114. Beeching, Kate 2002 Gender, Politeness and Pragmatic Particles in French. Amsterdam: John Benjamins. Benwell, Bethan 2001 Male gossip and language play in the letters’ pages of men’s lifestyle magazines. Journal of Popular Culture 34: 19–34. Benwell, Bethan 2003 Masculinity and Men’s Lifestyle Magazines. Oxford: Blackwell. Bucholtz, Mary 2003 Theories of discourse as theories of gender: Discourse analysis in language and gender studies. In: Janet Holmes and Miriam Meyerhoff (eds.), The Handbook of Language and Gender, 43–68. Oxford: Blackwell.
Gender identities and discourse
73
Butler, Judith 1990 Gender Trouble: Feminism and the Subversion of Identity. London: Routledge. Butler, Judith 2004 Undoing Gender. London: Routledge. Cameron, Deborah 1998 ‘Is there any ketchup, Vera?’: Gender, power and pragmatics. Discourse and Society 9: 437–455. Cameron, Deborah 2005 Relativity and its discontents: Language, gender and pragmatics. Intercultural Pragmatics 2–3: 321–334. Chouliaraki, Lilie and Norman Fairclough 1999 Language and power in Bourdieu: on Hasan’s ‘the disempowerment game’. Linguistics and Education 10(4): 399–409. Christie, Chris 2000 Gender and Language: Towards a Feminist Pragmatics. Edinburgh: Edinburgh University Press. Christie, Chris 2010 Gender and pragmatics. In: Louise Cummings (ed.), The Pragmatics Encyclopedia, 171–174. London: Routledge. Coates, Jennifer and Deborah Cameron (eds.) 1988 Women in their Speech Communities. London: Longman. Coates, Jennifer 1986 Women, Men and Language. London: Longman. Coates, Jennifer 1989 Gossip revisited: languages in all-female groups. In: Jennifer Coates and Deborah Cameron (eds.), Women in their Speech Communities, 94–122. London: Longman. Coates, Jennifer 1993 Women, Men and Language. 2nd edition. London: Longman. Coates, Jennifer 1998 Language and Gender: A reader. Oxford: Blackwell. Coates, Jennifer 2003 Men Talk. Stories in the Making of Masculinities. Oxford: Blackwell. Crawford, Mary 2000 Only joking: Humor and sexuality. In: Cheryl Travis and Jaquelyn White (eds.), Sexuality, Society and Feminism, 213–236. Washington, DC: American Psychological Association. Culpeper, Jonathan 2011 Impoliteness: Using Language to cause Offence. Cambridge: Cambridge University Press. De Klerk, Vivian 1990 Slang: a male domain? Sex Roles 22(9/10): 589–602. De Klerk, Vivian 1991 Expletives: men only? Communications Monographs 58: 156–169. De Klerk, Vivian 1997 The role of expletives in the construction of masculinity. In: Sally Johnson and Ulrike Meinhof (eds.), Language and Masculinity, 144–158. Oxford: Basil Blackwell.
74
Bróna Murphy
De Klerk, Vivian 2005 Slang and swearing as markers of inclusion and exclusion. In: Angie Williams and Crispin Thurlow (eds.), Talking Adolescence: Perspectives on Communication in the Teenage Years, 109–206. New York: Peter Lang. Dubois, Betty-Lou and Isabel Crouch 1975 The question of tag questions in women’s speech: they don’t really use more of them, do they? Language in Society 4: 289–294. Eckert, Penelope and Sally McConnell Ginet 1992 Think practically and look locally: Language and gender as community-based practice. Annual Review of Anthropology 21: 461–490. Eggins, Suzanne and Diane Slade 1997 Analysing Casual Conversation. London: Cassell. Fairclough, Norman and Ruth Wodak 1997 Critical Discourse Analysis. In: Teun A. van Dijk (ed.), Discourse as Social Interaction. Discourse Studies: A Multidisciplinary Introduction, 258–284. London: Sage. Farr, Fiona and Bróna Murphy 2009 Religious references in contemporary Irish English: ‘for the love of God Almighty … I’m a holy terror for turf’. Journal of Intercultural Pragmatics 6(4): 535–559. Fishman, Pamela 1983 Interaction: the work women do. In: Barrie Thorne, Cheris Kramarae and Nancy Henley (eds.), Language, Gender and Society, 89–101. Rowley, Mass.: Newbury House. Flowerdew, Lynne 2008 Corpora and context in professional writing. In: Vijay K. Bhatia, John Flowerdew and Rodney H. Jones (eds.), Advances in Discourse Studies, 115–131. London: Routledge. Grice, Paul 1968 Utterer’s meaning, sentence meaning, and word meaning. Foundations of Language 4: 225–242. Gumperz, John J. 1982 Language and Social Identity. Cambridge: Cambridge University Press. Haberland, Hartmut and Jacob Mey 2002 Linguistics and pragmatics, 25 years after. Journal of Pragmatics 34: 1671–82. Holmes, Janet 2006 Sharing a laugh: Pragmatic aspects of humor and gender in the workplace. Journal of Pragmatics 38(1): 26–50. Jaworski, Adam and Justine Coupland 1999 The Discourse Reader. London: Routledge. Jay, Timothy and Kristin Janschewitz 2008 The pragmatics of swearing. Journal of Politeness Research 4: 267–288. Jay, Timothy 1992 Cursing in America. Amsterdam: John Benjamins. Johnson, Sally and Ulrike Meinhof (eds.) 1997 Language and Masculinity. Oxford: Blackwell.
Gender identities and discourse
75
Koester, Almut 2010 Building small specialised corpora. In: Anne O’Keeffe and Michael McCarthy (eds.), The Routledge Handbook of Corpus Linguistics. London: Routledge. Lakoff, Robin 1973 The logic of politeness or minding your p’s and q’s. Papers from the Ninth Regional Meeting of the Chicago Linguistic Society. Chicago: Linguistic Society: 292–305. Lakoff, Robin 1975 Language and Woman’s Place. New York: Harper and Row. Lampert, Martin and Susan Ervin-Tripp 2006 Risky laughter: Teasing and self-directed joking among male and female friends. Journal of Pragmatics 38: 51–72. Litosseliti, Lia 2006 Gender and Language: Theory and Practice. London: Hodder Arnold. Maltz, Daniel N. and Ruth A. Borker 1982 A cultural approach to male-female miscommunication. In: John J. Gumperz (ed.), Language and Social Identity, 195–216. Cambridge: Cambridge University Press. Mateo, José and Francisco Yus 2000 Insults: a relevance-theoretic taxonomical approach to their translation. International Journal of Translation 12(1): 97–130. Mills, Sara 1997 Discourse. London: Routledge. Mills, Sara 2002 Rethinking politeness, impoliteness and gender identity. In: Lia Litosseliti and Jane Sunderland (eds.), Gender Identity and Discourse, 69–89. Amsterdam: John Benjamins. Mills, Sara 2003 Gender and Politeness. Cambridge: Cambridge University Press. Montagu, Ashley 2001 The Anatomy of Swearing. Philadelphia: Philadelphia University Press. Mullany, Louise 2007 Gendered Discourse in the Professional Workplace. Basingstoke: Macmillan. Mullany, Louise 2010 Gender and interpersonal pragmatics. In: Miriam Locher and Sage L. Graham (eds.), The Handbook of Interpersonal Pragmatics, 225–250. Mouton de Gruyter. Murphy, Bróna 2010 Corpora and Sociolinguistics: Investigating age and gender in female talk. Amsterdam: John Benjamins. Pichler, Pia and Eva M. Eppler (eds.) 2009 Gender and Spoken Interaction. Basingstoke: Palgrave. Pilkington, Jane 1998 Don’t try to make out that I’m nice. The different strategies women and men use when gossiping. In: Jennifer Coates (ed.), Language and Gender, 254–69. Oxford: Blackwell.
76
Bróna Murphy
Romero Trillo, Jesús (ed.) 2008 Pragmatics and Corpus Linguistics: a mutualistic entente. Berlin: Mouton de Gruyter. Rühlemann, Christoph 2010 What can a corpus tell us about pragmatics? In: Anne O’Keeffe and Michael McCarthy (eds.), The Routledge Handbook of Corpus Linguistics, 288–301. London: Routledge. Sacks, Harvey, Emanuel A. Schegloff and Gail Jefferson 1974 A simplest systematic for the organisation of turn-taking for conversation. Language 50(4): 696–735. Schegloff, Emanuel A. 1997 Whose text? Whose context? Discourse and Society 8: 165–187. Schegloff, Emanuel A. 1999 ‘Schegloff’s texts’ as ‘Billig’s data’: a critical reply. Discourse and Society 8: 558–575. Scott, Michael 2008 WordSmith Tools. Oxford: Oxford University Press. Seedhouse, Paul 2004 The Interactional Architecture of the Language Classroom: A Conversation Analysis Perspective. Oxford: Blackwell. Speer, Susan 2005 Gender Talk: Feminism, Discourse and Conversation Analysis. London: Routledge. Spender, Dale 1980 Man Made Language. London: Routledge and Kegan Paul. Stapleton, Karyn 2003 Gender and swearing: a community practice. Women and Language 26: 22– 33. Stapleton, Karyn 2010 Swearing. In: Miriam Locher and Sage L. Graham (eds.), Interpersonal Pragmatics, 289–306. Mouton de Gruyter. Stenström, Anna-Brita 2006 Taboo words in teenage talk: London and Madrid girls’ conversation compared. Spanish in Context 3: 115–138. Stenström, Anna-Brita, Gisle Andersen and Ingrid K. Hasund 2002 Trends in Teenage Talk. Corpus, Compilation, Analysis and Findings. Amsterdam: John Benjamins. Stokoe, Elizabeth and Janet Smithson 2001 Making gender relevant: Conversation analysis and gender categories in interaction. Discourse and Society 12: 217–44. Sunderland, Jane 2004 Gendered Discourses. Basingstoke: Palgrave. Tannen, Deborah 1992 You Just Don’t Understand: Men and Women in Conversation. London: Virago Press. Tannen, Deborah (ed.) 1993 Gender and Conversational Interaction. Oxford: Oxford University Press.
Gender identities and discourse
77
Velasco-Sacristán and Fuertes Olivera 2006 Towards a critical cognitive pragmatic approach to gender metaphors in advertising English. Journal of Pragmatics 38(11): 1982–2002. Weatherall, Ann 2002 Towards understanding gender and talk-in-interaction. Discourse and Society 13: 767–780. Weiss, Gilbert and Ruth Wodak (eds.) 2003 CDA, Theory and Interdisciplinarity. London: Palgrave/MacMillan. West, Candace and Don H. Zimmerman 1987 Doing gender. Gender and Society 1: 125–151. Widdowson, Henry 1995 Discourse analysis: a critical view. Language and Literature 4(3): 152–172. Wodak, Ruth (ed.) 1997 Gender and Discourse. London: Sage. Zimmerman, Don and Candace West 1975 Sex roles, interruptions and silences in conversation. In: Barrie Thorne and Nancy Henley (eds.), Language and Sex: Differences and Dominance, 105–129. Rowley, MA: Newbury House.
3.
Regional pragmatic variation María Elena Placencia
1.
Introduction
It is a well-known fact that language use is influenced by a range of socio-cultural factors. These include situation-specific contextual factors such as the relationship between participants in an interaction, as well as factors relating to participants’ identities and their membership of different social and cultural groups. One of the largest of these groupings is the nation-state with which national varieties of languages and cultures are associated, although with immigration, globalization and technological developments in communication, there are more and more groupings that transcend national/geographical boundaries. Other groupings include sub-cultures linked to age, ethnicity, region, gender and social class, as well as to particular professions and organizations. Certain patterns of language use are associated with these and other groupings, or may even be constitutive (in part) of some of them. Language use and variation in language use can thus be examined in relation to a range of factors. This chapter will focus on the impact of region on language use, or regional pragmatic variation. It starts with a definition of regional pragmatic variation (2), followed by a consideration of the genesis and development of this area (3) as well as the levels and domains in which regional pragmatic variation can be studied (4). In section 5, some findings from studies published primarily in English are provided for the purposes of illustration, and some trends in the study of regional pragmatic variation are sketched out. Finally, a summary of the chapter is provided in section 6, together with a consideration of some methodological issues in the study of regional pragmatic variation as well as some prospects for the future.
2.
Defining regional pragmatic variation
Regional pragmatic variation is a sub-field of variational pragmatics, a discipline of recent constitution (Barron 2005a; Schneider and Barron 2005, 2008a) that deals with “the investigation of possible correlations between macro-social factors and the use of language in action” (Barron 2005a: 525). The macro-social factors in question include region, gender, ethnicity as well as social class. Regional pragmatic variation is concerned with the study of regionally-based intra-lingual pragmatic variation. Regional variation in the use of language can, in turn, refer to a
80
María Elena Placencia
wide range of phenomena such as variation in speech act realization, turn-taking, use of listener responses, etc. (see Domains of Analysis in section 4 below). The term regional is employed with reference to at least two levels of analysis: the national and the sub-national1 (Schneider and Barron 2008a). The first is concerned with variation at the level of national varieties of pluricentric languages, that is, languages that have several interacting centres (Kloss 1978; Clyne 1992),2 such as English (e.g. American English and British English), German (e.g. German German and Austrian German) and Spanish (e.g. Mexican Spanish and Peninsular Spanish). The second level deals with internal varieties of a language “geographically distributed within a given country or nation state” (Schneider and Barron 2008a: 17), such as Swabian and Bavarian German; Northern, Midland, and Southern American English, etc. Schneider and Barron (2008a) further distinguish between the local and sub-local levels: the first corresponds to the study of pragmatic variation across specific locations such as particular cities or towns, and the second one, across districts within a single city or town. The majority of studies currently available focus on the national level; however, in practice, most of these studies offer a contrast of language use of a sub-variety within the national variety in question. For instance, García’s (2004) study of reprimands in Peruvian and Venezuelan Spanish is a study that contrasts speech act realization in Lima, Peru, and Caracas, Venezuela; these are locations that represent particular sub-varieties of Peruvian and Venezuelan Spanish, respectively. The assumption behind contrastive studies of national varieties is that national varieties have at least some norms of their own (Clyne 1992: 1) and that national varieties are identified with a particular nation by both the in- and the outgroup (Clyne 1992: 2). Norms or patterns that characterize different national varieties can be found at the phonological, morphosyntactic, lexical and (socio) pragmatic levels as well as in relation to spelling conventions (cf. Baxter 1992). When it comes to sociopragmatic aspects, it is assumed that there are patterns or regularities of language use associated with socio-cultural practices and historical processes (Mesthrie and Bhatt 2008). As Hofstede (1991: 12), amongst others, has observed, “[m]any nations […] form historically developed wholes even if they consist of clearly different groups”, and national cultures are said to function as a source of cultural meanings (Hall 1992). We take culture to mean “ways of living, speaking, thinking and feeling widely shared in a particular society” (Wierzbicka 2005: 580), or as co-occurring regularities “found in a wide range of elements, including basic assumptions, fundamental values, deep-seated orientations to life, attitudes, beliefs, policies, procedures and behavioural conventions” (Spencer-Oatey 2008a: 4); these regularities are (partly) constructed by and manifested through talk. This is not to say that all members of a particular nation adhere to its norms to the same extent, that is, it is accepted that there is individual variation. Also, it is generally accepted that national culture is only one level of culture that may be addressed in interaction, in addition to other levels
Regional pragmatic variation
81
associated with macro-social factors such as gender and ethnicity, as well as with particular sub-groups linked, for example, to particular professions or organizations.
3.
Genesis and development of regional pragmatic variation as a field of study
The study of regional pragmatic variation is at the interface of modern dialectology, which looks at social as well as regional variation, and pragmatics (Schneider and Barron 2008a). Both disciplines deal with variation, but regional pragmatic variation has been largely overlooked in both disciplines, as well as in traditional dialectology, hence Schneider and Barron’s (2008a) proposal to combine pragmatics and (modern) dialectology under variational pragmatics. Within traditional and modern dialectology, the focus, as highlighted by Schneider and Barron (2008a), has been on phonetic, phonological, grammatical or lexical features; that is, pragmatic aspects have not been granted much attention. While these authors make this observation to refer mainly to German and English, their claim seems to hold true for other languages such as French3 and Spanish.4 On the other hand, within pragmatics, the study of pragmatic variation, which dates back to the late 1970s and early 1980s, has been largely confined to situational and cross-linguistic variation.5 That is, it has focussed mainly on examining the influence of factors of the micro-context, such as the degree of social distance between the interactants, on speakers’ choice of pragmatic strategies, and on the study of how pragmatic choices vary across different languages rather than crossregionally within varieties of the same language. Blum-Kulka and Olshtain (1984) and Blum-Kulka, House and Kasper’s (1989a) influential work on requests and apologies is illustrative of this kind of focus that is constitutive of, and that has prevailed in, cross-cultural pragmatics. Cross-linguistic studies of the 1980s and 1990s were closely linked to second/ foreign language teaching and learning, seeking to establish performance baselines through contrastive analyses of pragmatic features across different languages.6 This partly explains the focus on cross-linguistic rather than intralingual variation. A large number of these cross-linguistic studies were also prompted by Brown and Levinson’s ([1978]1987) seminal theory of politeness with its universality claims which many scholars sought to explore. Much work in this area, however, seems to have been built on the assumption that speech communities of the languages examined constitute homogeneous wholes (Schneider and Barron 2008a: 5), thus neglecting the examination of possible intralingual regional variation. While Variational Pragmatics is a discipline of recent vintage, interest in regional pragmatic variation is not new and can be traced back to the 1970s and early 1980s, with systematic studies beginning to emerge in the late 1980s and the 1990s.
82
María Elena Placencia
German, English and Spanish are the main languages associated with these early developments. With respect to German, Schneider and Barron (2008a) point the reader to the work of Schlieben-Lange and Weydt (1978), published in German, that argued for the incorporation of pragmatic phenomena in German dialectology. Their call, however, seems to have gone largely unheeded in studies of German dialectology. The first systematic studies of regional pragmatic variation can be found in the work of Muhr (1987; 1993; 1994) on, among other features, requests and apologies in Austrian and German German, and Birkner and Kern (2000) on the management of self-presentation in job interviews among Germans in the former German Democratic Republic and in the Federal Republic of Germany. More recently, Clyne, Fernandez and Muhr (2003), for example, looked at Austrians and Germans’ metapragmatic perceptions of apologies, among other aspects. In relation to English, the situation is similar. Different scholars started highlighting region as a factor of variation in language use since the early 1980s. Wolfson (1983: 66), for example, observed that “sociolinguistic rules are subject to considerable variation with respect to region and social class” and illustrated her point with examples of address form usage in the south and northeast of the U.S. Along the same lines, Tannen (1984a: 1) noted that conversational style can vary according to individual factors, but also with respect to differences such as gender, ethnicity, class and regional background. However, the first systematic studies, to our knowledge, can be traced back to the work of Herbert (1989) on compliments and compliment responses in American and South African English (see also Herbert and Straight 1989), Creese’s (1991) on the same topic in American and British English, and Tottie’s (1991) contrastive study of backchannels also in these last two varieties of English. It is only more recently that there has been renewed interest in the area, largely promoted by Barron and Schneider’s (Barron and Schneider 2005; Schneider and Barron 2008b) publications. Irish (cf. Schneider 2005; Barron 2005b; O’Keefe and Adolphs 2008) and other varieties of English such as New Zealand English (Jautz 2008) are now being examined. Finally, with reference to Spanish, the first studies also date back to the 1990s. During this decade, a number of scholars working on different varieties of Spanish sought to examine features of pragmatic variation between a Latin American variety and Peninsular Spanish. These studies were partly prompted by an interest in investigating a perception prevalent among some Latin Americans that Spaniards employ a more direct or abrupt communicative style than speakers of certain varieties of Latin American Spanish, which leads Spaniards to be perceived as less polite. Interest in this area also derived from the observation that, as Placencia (1994: 66) puts it, “although Spain and the different Spanish-speaking countries in Latin America share the ‘same’ linguistic code … they are quite diverse in terms of their geographical and ethnic configurations, as well as their historical, social, economic and political development”.
Regional pragmatic variation
83
Authors working in this area therefore judged that, in addition to phonological, lexical and grammatical differences that had been well documented, differences in the use of the language in context were also to be found. The varieties initially contrasted with Peninsular Spanish included Chilean (Puga Larraín 1997; 1999); Ecuadorian (Placencia 1994; 1998); and Mexican Spanish (Fant 1996; Bravo 1998; Curcó 1998; Wagner 1999). Other varieties such as Uruguayan Spanish were later examined also in contrast with Peninsular Spanish (cf. Márquez Reiter 2002). The first decade of the new millennium saw the emergence of contrastive work involving Latin American varieties such as Peruvian and Venezuelan Spanish (cf. García 2004) or Mexican and Dominican Spanish (cf. Félix-Brasdefer 2008a). All of these studies have demonstrated the productiveness of the area. With respect to other languages, developments have been slower. In relation to Greek and Portuguese, for example, some initial contrastive work on Greek from the 1990s (cf. Terkourafi 1999), and on Portuguese from the 1970s and 1980s (cf. Wilhelm 1979; Jensen 1981; Silva-Brummel 1984), shows the potential interest of the area; however, not much has been done since with respect to these two languages. This also applies to other pluricentric languages such as Swedish (cf. Reuter 1992; Clyne et al. 2006) and Chinese (cf. Spencer-Oatey, Ng and Li Dong 2000; Chen 2003). Schneider and Barron’s (2008b) publication, nonetheless, has played a pivotal role in opening up the area to other languages such as Dutch (Plevoets, Speelman and Geeraerts 2008) and French (Schölmberger 2008). In terms of the approach employed and the focus and framework of analysis, studies on regional pragmatic variation mirror, to a large extent, developments within cross-cultural pragmatics. Many studies, particularly those focussing on varieties of English, show, for example, the influence of Blum-Kulka et al.’s (1989a) seminal work on requests and apologies referred to above; this is in relation to the aims pursued (e.g. the testing of the universality of conventional indirectness), the descriptive framework for the analysis of requests and apologies employed, as well as the methodological approach and tools used for data collection such as the Discourse Completion Task (DCT), a type of production questionnaire aimed at eliciting single turns.7 An increasing number of studies, however, display the influence of conversation analysis, for example, with reference to the type of data employed (i.e., naturally occurring vs. elicited data) and/or the focus of analysis (e.g. examining aspects of turn-taking, the sequential, or the overall organization of conversation rather than speech act realization) (see section 4 below). Many studies, particularly in the 1990s, also display the influence of Brown and Levinson’s ([1978]1987) theory of politeness and modified versions of this theory (e.g. Scollon and Scollon [1995] 2001), or of Leech’s (1983) conceptualization of politeness, in the explanations offered to account for variation in language use. The impact of more current theories of politeness such as those of SpencerOatey (2000) and Locher and Watts (2005) can be seen in more recent studies.
84
María Elena Placencia
Other research in pragmatic variation, however, has been guided by work on national mentalities and cultural values (cf. Hofstede 1980). Some studies, on the other hand, follow the ethnography of speaking tradition as reflected in Manes and Wolfson’s (1981) research on compliments in American English. There is also a growing number of contrastive corpus linguistics works, particularly in the study of English. Researchers are making use of existing corpora such as the Cambridge and Nottingham Corpus of Discourse in English (CANCODE), the Cambridge North American Spoken Corpus and the Limerick Corpus of Irish English (LCIE) to explore variation in the use of a range of pragmatic features. The different approaches and frameworks employed in the study of regional pragmatic variation are necessarily reflected in the greater or lesser attention that is given to examining socio-cultural factors and interpreting the findings in terms of cultural values and politeness orientations. For instance, while studies in the crosscultural pragmatics tradition seek to explore the interrelation between language and culture, contrastive corpus linguistic studies may not necessarily have the same aims, since studies in this field have their origin in different areas of interest such as grammar and the lexicon.
4.
Levels, dimensions, and domains of analysis in the study of regional pragmatic variation
Pragmatic variation can be studied at different levels and in relation to different dimensions as well as different domains. The first broad level corresponds to the national and sub-national distinction mentioned above. Most studies have focussed on the national level as pointed out earlier. A second broad level relates to the differentiation between pragmalinguistic and sociopragmatic studies according to whether the focus is on linguistic rather than social aspects of language use (Leech 1983). Pragmalinguistics can be said to be concerned with “the particular resources which a given language provides for conveying particular illocutions” (Leech 1983: 11), for example, with linguistic devices available in a particular language to convey mitigation. Corpus linguistics studies tend to focus on this level. Sociopragmatics, on the other hand, is concerned with the study of uses of language “relative to specific social conditions” (Leech 1983: 10). Following Brown and Levinson ([1978]1987) and Blum-Kulka et al. (1989a), the specific social conditions in Leech’s quotation often have to do with examining language use in relation to situated features of the micro-context, such as social distance and power. Another key difference between pragmalinguistic and sociopragmatic studies is that the latter, unlike the former, often seek to explain the motivations behind particular uses in terms of socio-cultural factors.
Regional pragmatic variation
85
These explanations are usually encapsulated under the notions of cultural values and beliefs and/or politeness orientations/ideologies. Most studies available can be described as sociopragmatic although not all fit neatly into the pragmalinguistic/ sociopragmatic categorization. Several dimensions of cross-linguistic/cross-cultural differences have been proposed for the study of language use. These dimensions can be, and in some cases have already been employed in the examination of regional pragmatic variation. Tannen (1984b: 189–194), for example, discussed eight ‘levels of communication differences’: when to talk, what to say, pacing and pausing, listenership, intonation, formulaicity, indirectness, and cohesion and coherence. Scollon and Scollon ([1995] 2001), on the basis of Brown and Levinson’s ([1978]1987) work, proposed the notions of involvement vs. independence to refer to different face preferences (cf. Felix-Brasdefer 2008a). In her contrastive studies of German and English, House (1996; 2000) found the following dimensions useful: Orientation towards content Orientation towards self Directness Explicitness Ad hoc formulation
D D D D D
Orientation towards addressee Orientation towards other Indirectness Implicitness Verbal routines
In the same vein, but approaching the contrastive analysis of the communicative style of Swedes and Spaniards, Fant (1989: 249) suggested that culture-based differences in communicative patterns could be accounted for in terms of “priorities, or socially based preferences, among conditions set on communicative interaction” (e.g. how important it is to display consensus or to be economical with one’s contributions, etc.). In a later contrastive work, Fant (1995) found the person- vs. taskorientedness dimension useful in the characterization of the communicative style of Swedish and Spanish negotiators;8 this is also a dimension that has been employed in some studies dealing with regional pragmatic variation (cf. Placencia 2005). Spencer-Oatey and Jiang (2003) in turn proposed what they refer to as sociopragmatic interactional principles or SIPs (e.g. routinization – novelty; clarity – vagueness; cordiality – restraint). Finally, an allied proposal comes from Muhr (2008) who has put forward a set of pragmatic discourse features linked to cultural standards. Pragmatic studies can also be classified according to domain of analysis, that is, on the basis of whether they deal with the utterance level or connected discourse, with verbal and/or non-verbal aspects of communication, etc. SpencerOatey’s (2000; 2008b) classification for the study of rapport management, which appears to be suitable for the examination of pragmatic features of language use in general, has also been used in the study of regional pragmatic variation. It includes the following domains:
86 x
x
x
x
x
María Elena Placencia
Illocutionary domain: This domain covers the analysis of the linguistic realization of requests, compliments and other speech acts. With requests and other directives, for example, it typically involves the study of levels of (in)directness and the use of devices that upgrade or downgrade the illocutionary force of the directive (cf. Breuer and Geluykens 2007). Some recent studies also incorporate prosodic features in the contrastive examination of speech acts (cf. Félix-Brasdefer 2008b). With other speech acts such as compliments and compliment responses, the analysis may involve the identification of particular semantic and syntactic formulae and preferred types of responses that convey, for instance, (partial) acceptance or rejection (cf. Schneider 1999). This domain is the one that has received the most attention in the study of regional pragmatic variation. Discourse domain: This domain covers discourse content and discourse structure. It would include, among other features, the examination of topic choice and topic management as well as the organization and sequencing of information in which discourse markers, for example, play a part. The analysis of features of the overall organization of conversation, such as how particular interactions are opened and closed, would also come under this domain (cf. Schneider 2008; Márquez Reiter and Placencia 2004; Placencia 2008). Participation domain: In this domain, the analysis deals with “procedural aspects of an interchange” (Spencer-Oatey 2000: 20), that is, aspects of turn-taking, including the analysis of turn-taking mechanisms and listener responses which can be verbal or non-verbal (cf. Fant 1996; McCarthy 2002; O’Keefe and Adolphs 2008). Stylistic domain: Here the focus is on “the stylistic aspects of an interchange, such as choice of tone (for example, serious or joking), choice of genre-appropriate lexis and syntax and choice of genre-appropriate terms of address or use of honorifics” (Spencer-Oatey 2000: 20) (cf. Silva-Brummel 1984; Clyne et al. 2006). Non-verbal domain: This last domain covers non-verbal aspects of the interchange which include “gestures, and other body movements, eye contact and proxemics” (Spencer-Oatey 2000: 20). Following Poyatos (2002), amongst others, this domain would also include paralanguage which comprises, for example, the class of differentiators such as laughter, the object of analysis in a few contrastive studies (cf. Bravo 1998). This domain is the one that has received the least attention in the study of regional pragmatic variation.
It has to be noted, however, that this and other classifications available9 are normally taken as a guide only as they are not entirely clear-cut or problem-free in their application. As Spencer-Oatey (2000: 19) rightly observes, the different domains outlined above are “interrelated”. For example, gaze and gesture, categorized under the non-verbal domain, may play a role in the exchange of turns and may
Regional pragmatic variation
87
therefore need to be considered as features of both the participation and the nonverbal domains. Likewise there are studies that examine speech acts such as invitations (cf. García 2008) both in terms of the semantic formulas employed and as exchanges performed across sequences, therefore straddling two domains – the illocutionary and the discourse. 5.
Studies on regional pragmatic variation: An illustrative account
This section reports on a number of studies dealing with regional pragmatic variation. It does not aim to provide an exhaustive account, but offers an illustration of topics that have been dealt with, some of the results obtained, as well as certain trends that can be identified. The focus is on Spanish and English as the languages that have been most widely examined. 5.1.
Spanish
Spanish is the language that has received the most attention in the study of regional pragmatic variation. As can be seen in Table 1 below, there are several (sub-) national varieties that have been examined contrastively. Table 1.
Studies on regional pragmatic variation in Spanish in chronological order
Author(s)
Year
Focus
Varieties of Spanish Examined
Placencia
1994
Fant
1996
Telephone management requests Turn delivery and turn-taking features in business negotiation
Puga Larraín Bravo
1997
Mitigation
1998
Curcó
1998
Placencia
1998
Puga Larraín Wagner
1999
Laughter in business negotiation Requests & mitigation Requests in service encounters Mitigation
1999
Apologies
†
†
Grindsted
2000
Markers of social proximity/distance
†
†
AS ChS DoS
ES
MxS PnS
†
† †
†
†
† †
†
†
†
†
†
†
†
PvS
UrS USS VS
88
María Elena Placencia
Author(s)
Year
Focus
Varieties of Spanish Examined AS ChS DoS
ES
MxS PnS
PvS
UrS USS VS
De los Heros
2001
Compliments
2 subvars.
Hardin
2001
Pragmatic features of television commercials
Curcó & de Fina
2002
Requests & mitigation
Márquez Reiter
2002
Requests in various situations
†
†
Márquez Reiter
2003
Requests in various situations
†
†
García
2004
Reprimands & responses
Márquez Reiter & Placencia
2004
Openings & closings of service encounters; selling strategies
†
Placencia
2005
Requests in service encounters
†
†
†
†
†
†
†
† †
†
†
†
Alba Juez
2008
Impolite actions
Albelda Marco
2008
Mitigation in youth talk
FélixBrasdefer
2008a Refusals to various directives
†
†
FélixBrasdefer
2008b Requests
†
†
GarcésConejos Blitvich & Bou Franch
2008
Internet-based service encounters
García
2008
Invitations
Jørgensen
2008
Discourse markers in youth talk
Martínez Camino
2008
Television advertising
Placencia
2008
Requests; small talk in service encounters
García
2009
Reprimands
†
†
†
†
†
† †
† †
†
2 subvars. †
†
†
Regional pragmatic variation
89
The majority of studies consist of comparisons of features across national varieties, and, as can also be seen from Table 1, the majority involve a contrast between a Latin American variety of Spanish and Peninsular Spanish. US Spanish is also an object of analysis in Hardin’s (2001) study, for example, in contrast with Chilean and Peninsular Spanish. The illocutionary domain is the area that has been most widely explored, and requests are the speech acts that have received the most attention (cf. Curcó 1998; Curcó and de Fina 2002; Félix Brasdefer 2008b; Márquez Reiter 2002, 2003; Placencia 1998, 2005, 2008). Only scant attention has been given to other speech acts such as apologies (cf. Wagner 1999), invitations (cf. García 2008) and reprimands (cf. García 2004). The participation (cf. Fant 1996), discourse (cf. Márquez Reiter and Placencia 2004; Placencia 2008), stylistic (cf. Jørgensen 2008; Placencia 2005, 2008) and non-verbal domains (cf. Bravo 1998) have also been explored, albeit to a much smaller degree. Most published studies deal with face-to-interaction in a range of contexts (e.g. in interactions in service encounters and among friends or roommates, boss and employer, business negotiators); however, there is an increasing number of studies that focus on media discourse such as television advertising (Hardin 2001; Martínez Camino 2008), and a few that deal with computer mediated discourse (cf. Garcés-Conejos Blitvich and Bou Franch 2008). In terms of findings, direct comparisons between the studies available are not easy to make given that the corpora employed are often not comparable and given that different frameworks of analysis are often used in different studies. Some preliminary trends can nonetheless be identified. In relation to Latin Americans’ perceptions of Spaniards’ communicative style, as referred to in section 3, research has shown that there are indeed some differences in the way Spaniards and speakers of other Latin American varieties perform different actions in comparable situations. Nonetheless, and not surprisingly, differences have also been found between Latin American national varieties and, even more interestingly, at the sub-national level. A few studies involving Peninsular, Ecuadorian, Uruguayan, Chilean and Mexican Spanish are considered below to illustrate some of these differences. Placencia (1994), for example, focused on the realization of telephone management requests (e.g. requests to speak to the person called, requests for self-identification) and their corresponding replies in Ecuadorian Spanish (Quito) and Peninsular Spanish (various locations). She based her study on both naturally-occurring data and oral production questionnaires. Among the differences Placencia observed were that when requesting a self-identification, Spaniards’ repertoire includes direct forms such as ¿quién eres? ‘who are you?’ which Ecuadorians would regard as abrupt and impolite. Ecuadorians in turn were found to prefer covert realizations such as ¿con quién hablo? ‘with whom am I speaking?’; the latter are forms that have a speaker rather than a hearer perspective. Likewise, in replies to
90
María Elena Placencia
such requests, the author found that Spaniards commonly employ the direct form soy X ‘I’m X’, whereas Ecuadorians employ a range of less direct forms that go from the neutral (e.g., Habla con Carmen ‘You are speaking with Carmen’) to the deferential, marked by lexical choices and word order (e.g., Carmen le saluda ‘Carmen greets you’). Placencia also found that Ecuadorians use more mitigation than Spaniards. Indeed, the author suggests that the lesser use of mitigation in Peninsular Spanish is one of the features that may lead Ecuadorians to perceive Spaniards as direct and brusque. The key features of variation identified in Placencia (1994) also appear in Placencia’s (1998; 2005) studies involving the same two varieties, albeit in service encounter interactions. For instance, greater use of mitigation in Quito, compared to Madrid, is a characteristic of requests at hospital reception desks (Placencia 1998) as well as in corner shop interactions (Placencia 2005). All in all, Placencia found more interpersonal padding in the Quiteño service encounters compared to the Madrileño ones. Quiteños in this sense, using Fant’s (1995) person vs. task-orientedness dimension, can be described as more person-oriented than Madrileños. As in Placencia (1994), Placencia’s other studies (1998, 2005) also show the use of deferential forms in Quito which are absent in Madrid in the contexts examined. Additionally, in relation to address usage, formal nominal and pronominal address was found to be common in the service encounters examined in Quito, but not in Madrid. Placencia suggests that the use of formal forms and deference markers points to a more hierarchical society in Quito and, conversely, to a more egalitarian one in Madrid. On the deference/hierarchy vs. equality dimension, Quiteños in (certain) service encounters would therefore be located towards the deference/hierarchy pole, and Madrileños towards the opposite end. Nonetheless, Placencia’s (2008) study at the sub-national level sheds a different light on such a differentiation between Ecuadorian and Peninsular Spanish. Placencia (2008) compared corner shop transactions in Ecuadorian Andean (Quito) and Coastal Spanish (Manta) based on audio-recordings and observation of naturally occurring interactions. At the illocutionary level, as in Placencia (1998, 2005), this study shows an overall preference for direct requests in both Quito and Manta for service transactions (e.g. un litro de leche ‘one litre of milk’). However, Quiteños were found to employ a great deal more internal modification in their request formulation in the form of diminutives and other mechanisms (e.g. por favor regáleme pancito, literally, ‘please give away to me some bread+ diminutive’). This was also the case with politeness formulas: a higher use and a wider range of politeness formulas – from neutral to deferential – were found in the Quiteño corpus; by contrast, only the neutral or standard por favor ‘please’ was employed in the Manteño service encounters. At the discourse level, Quiteños were found to produce more relational talk overall. These results for Manteños echo Placencia’s findings relating to Madrileños’ communicative style (Placencia 2005). Where Manteños differ from Madrileños and coincide with Quiteños is in terms of the sty-
Regional pragmatic variation
91
listic domain: both Manteños and Quiteños prefer the formal V address form unlike Madrileños who favour the T form. This study thus illustrates the need to examine pragmatic variation at the sub-national level for a safer ground for generalizations at the national level. Moving on from Ecuadorian Spanish, Puga Larraín (1997; 1999) also found a greater use of mitigation devices (atenuación) in Chilean Spanish (Santiago) compared to Peninsular Spanish (Valencia). This finding is presented on the basis of her analysis of a range of spoken and written texts as well as her own observations. The author describes mitigation as a means of creating distance (from the speaker, his/ her interlocutor, the message or time) and suggests that mitigation is motivated by the need to convey deference; the higher incidence of mitigation in Chile compared to Spain would be explained by the fact that Chilean society is more stratified than Spanish society. Again, the deference/hierarchy vs. equality dimension seems to be relevant for Chileans in relation to Spaniards. This dimension also surfaces in Hardin’s (2001) study on television advertising in Chile, Spain and the US. With reference to Brown and Levinson’s ([1978] 1987) work, Hardin found, for example, that negative politeness strategies displaying distance and power were most frequently used in Chilean adverts, with the Spanish data containing the smallest number of such strategies. Márquez Reiter (2002), who examined requests in Uruguayan Spanish and Peninsular Spanish based on data from role play interactions, found that Uruguayans employ internal modification devices more frequently than their Spanish counterparts. On the other hand, with respect to overall request strategies, she found that Uruguayans and Spaniards both prefer conventional indirectness in interactions where participants are not familiar with each other. There were, however, some differences with respect to the linguistic formulae employed. Also, there were differences in relation to the use of alerters or precursors (e.g. oye ‘listen’ type; perdona ‘excuse me’). The author found a higher use and a wider range of precursors among Uruguayans, which resulted in slightly longer and more deferential preambles (see also Márquez Reiter 2003). In brief, while Uruguayans and Spaniards did not differ on the directness-indirectness scale, they did differ on what could be described as the tentativeness-terseness scale, as Spaniards were found to be less tentative. The relevance of the deference/hierarchy-equality dimension also surfaces in Márquez Reiter’s study. In another study conducted by Márquez Reiter and Placencia (2004), looking at the discourse domain, longer and more deferential preambles were found this time among Quiteños compared to Montevideans in the context of service encounter interactions in shops selling clothes and accessories. On the scale of deference/hierarchy-equality, in certain contexts at least, Quiteños would need to be placed closer to the deference pole, Spaniards at the other extreme, with Montevideans possibly towards the middle. Quiteños’ openings were found to be not only longer but also more ritualistic than those of Montevideans; in this respect, on the ad hoc formu-
92
María Elena Placencia
lation vs. verbal routines scale (House 1996; 2000), Quiteños would be closer to the verbal routines end, and Montevideans, to the ad hoc formulation pole. Differences in the use (and perception) of mitigation strategies also feature in Curco’s (1998) study on Mexican and Peninsular Spanish. Employing a rating scales questionnaire which Curcó administered to university students in Mexico City and Barcelona, the author found, for example, that Mexicans, unlike Spaniards, rated utterances with imperatives without mitigation or only minimal mitigation as less polite. In relation to the use of diminutive suffixes, Curcó observes, that while for Mexicans politeness increases with the inclusion of a diminutive (e.g. Oye, mueve tu coche, por favorcito’ ‘Hey, move your car please +diminutive’), it decreases for Spaniards. Curcó explains her results in relation to what she considers to be different face wants among Mexicans and Spaniards, with the former paying more attention to their interlocutor’s positive face wants than the latter (see also Curcó and de Fina 2002). Two other studies dealing with Mexican Spanish examined in contrast with Peninsular Spanish are Fant (1996) and Grindsted (2000); these studies also illustrate some differences between the two varieties in relation to various domains, particularly the participation and stylistic domains. Drawing on conversation analysis and work on national mentalities and cultural values (Hofstede 1980; 1991), Fant (1996) contrasted the communicative style of Mexican and Spanish business negotiators, focussing on features of turn delivery and the exchange of turns. His study is based on a corpus of simulated negotiations collected for training purposes in Mexico and Spain. The author found, for instance, that overlaps in turn-exchange occurred in both groups but were more frequent among Spaniards. Also, there was a higher proportion of interruptive (as opposed to non-interruptive) overlaps in the Spanish corpus. The author interprets these findings as reflecting a higher tolerance among Spaniards of direct confrontation in negotiation. Overall, Fant suggests that his findings lend support to different proposals that have been made about Spaniards’ orientation towards individualism and self-affirmation (cf. Thurén 1988), and Mexicans, towards collectivistic behaviour. Finally, Grindsted (2000), drawing on Hall’s (1966) work on proxemics, more specifically, on the notions of interactive space and interactive centre, examined how the spatial dimension of interaction can be reflected metaphorically in the use of language. She focused on the use of social deictic expressions in simulated business negotiations in Mexico and Spain (see also Fant 1996); this was with the aim of identifying markers of social proximity that metaphorically regulate interpersonal social distance in interaction. According to Grindsted, utterances such as imperatives employed as attention getters (e.g. oye ‘listen’) help establish a higher degree of proximity among interlocutors (2000: 212). She found that utterances of this type are much much more frequent among Spaniards than Mexicans. She also found that Mexican, unlike Spaniards, employ expressions like permítame ‘if you will let me’, caballero ‘gentleman’ and señor ‘Mr’ orienting themselves towards a
Regional pragmatic variation
93
‘hierarchy principle’; this principle is related to the expression of respect for the social status of one’s interlocutor. The expressions Spaniards employ (e.g. first names, oye ‘listen’) on the other hand, suggest that they have a preference for an “egalitarian principle” (Grindsted 2000: 212). Like Fant (1990) Grindsted concludes that her findings support previous proposals that suggest an orientation of Spaniards towards individualism and self-affirmation, and of Mexicans towards collectivism, where it is important to show respect. In brief, from the account above, various points stand out concerning the communicative style of Latin Americans and Spaniards; it is clear, for example, that mitigation, associated with tentativeness and perhaps also deference, is employed more frequently among speakers of several Latin American varieties of Spanish than among Spaniards. The notions of deference and showing respect emerge as linked to certain Latin American varieties of Spanish only. For Peninsular Spanish, other notions such as self-affirmation seem to be more relevant. Finally, while Spaniards seem to display more tolerance of disagreement/confrontation, Mexicans, for example, seemingly orientate themselves towards agreement/conflict avoidance. However, it has to be taken into account that there is also variation at the sub-national, as remarked earlier. In relation to similarities and differences between Latin American varieties of Spanish, García’s (2004) is one of the first studies in this area. García compared the realization of reprimands and responses to reprimands in Peruvian (Lima) and Venezuelan (Caracas) Spanish based on role-play interactions. She found that in the issuing of reprimands, direct, face-threatening strategies are preferred by both sets of informants. Nonetheless, Venezuelans were found to use more and a wider range of strategies than Peruvians, thus leaving room for some negotiation. Also, Venezuelans were found to employ more and a wider range of supportive moves than Peruvians, with more aggravators rather than mitigators being produced by Peruvians, and more mitigators rather than aggravators by Venezuelans. In responding to reprimands, both sets of informants were found to use different positive and negative politeness strategies (Brown and Levinson [1978]1987) to realize the head act as well as a range of supportive moves. On the whole, however, Peruvians exhibited a preference for negative politeness strategies, and Venezuelans, for direct, positive politeness strategies. These preferences are interpreted as showing that Peruvians tend to display deference and respect, which, in turn, would signal an acknowledgement of a power differential. On the other hand, Venezuelans’ use of direct strategies is, according to the author, a way of counteracting the power and authority of the person issuing the reprimand (García 2004: 140). As we can see, García’s study also shows the relevance of the deference/hierarchy vs. equality dimension in the examination of certain Latin American varieties in contrast. Most of the studies referred to so far deal with face-to-face interaction. When it comes to certain computer mediated or media discourse genres, the situation may be different. For instance, Garcés-Conejos Blitvich and Bou Franch (2008) exam-
94
María Elena Placencia
ined the use of politeness strategies in e-commerce, more specifically in a number of web pages in Spanish in a range of companies in the U.S. and Spain. They did not find important differences. They attribute this to the internet-based nature of e-commerce where the effects of globalization are more obvious. Their conclusion is that when it comes to e-business the conventions of the genre appear to play a more important role than the national culture conventions. 5.2.
English
Studies dealing with regional pragmatic variation in English have also covered a range of topics in different contexts (e.g. face-to-face interaction, radio phone-in programmes, etc.) focussing on various domains. In relation to the illocutionary domain, for example, several speech acts have been examined, including requests (cf. Breuer and Geluykens 2007; Barron 2008a, 2008b), offers (Barron 2005b), thanks or expressions of gratitude (Jautz 2008) and compliments (cf. Herbert 1989); the discourse/participation domain has also been explored with studies on small talk (Schneider 2008), listener responses (Tottie 1991; McCarthy 2002; O’Keefe and Adolphs 2008) and discourse markers (Kallen 2005), among other aspects. Below, we look at results from some of these studies, restricting ourselves to works of inner circle Englishes (Kachru 1985), that is, those varieties where English is the first language for the majority of the population; this choice reflects the focus of current and past research on regional pragmatic variation relating to English. The inner circle varieties that have been examined contrastively include the following: American English, British English/English English, Irish English, New Zealand English and South African English (see Table 2 below). Rather surprisingly, there appear to be no works involving a contrast with Australian English. Nonetheless, some works available on individual varieties of outer circle Englishes (Kachru 1985) that is, those varieties spoken by people for whom English is a second language, show the interest there is in expanding the study of regional pragmatic variation beyond inner circle Englishes. A case in point is Sridhar’s (1996) work on request realization in South India (Bangalore), based on data from DCTs administered to university students from different institutions. This study shows some situational variation as well as variation related to the background of the students – whether they come from a more ‘westernized’ or a more ‘traditional’ educational background; at the same time, some common uses associated with the local culture emerge. For instance, certain kinship terms are employed to address elders even if one is not related to them. Another is the more frequent use of imperative constructions in some of the situations by students from the more traditional background in particular, possibly reflecting local practices.10 Going back to contrastive work on inner circle Englishes, studies in the illocutionary domain tend to show, like some studies on varieties of Spanish, that the
95
Regional pragmatic variation Table 2. Author(s)
Studies on regional pragmatic variation in English in chronological order Year
Focus
Varieties of English Examined American English
British English
English English
Irish New South English Zealand African English English
†
†
†
†
Herbert
1989
Herbert & Straight
1989
Creese
1991
Compliments and responses and other speech acts
†
†
Tottie
1991
Backchannels
†
†
Schneider
1999
Compliment responses
†
McCarthy
2002
Listener responses
†
Barron
2005b Offers
†
†
Kallen
2005
Silence and mitigation
†
†
Schneider
2005
Thanks minimizers
†
†
†
Requests
†
† †
† †
Breuer & 2007 Geluykens
Compliments and responses
British English
Barron
2008a Requests 2008b Requests
Jautz
2008
Gratitude
O’Keefe & 2008 Adolphs
Listener responses
Schneider
Small talk
2008
† †
†
† † †
† †
†
overall strategies employed in the realization of particular speech acts are often similar, and that what is subject to variation is the choice of sub-strategy, as well as the use of internal and external modification of the head act. For example, Barron (2005b) investigated offers in English English (South of England) and Irish English (South-East of Ireland), based on data from free discourse completion tasks (FDCTs),11 administered to female school students. With reference to Schneider’s (2003) characterization of offer strategies, Barron found that both Irish and English informants prefer conventional indirectness in the form of ‘preference’ and ‘execu-
96
María Elena Placencia
tion’ strategies; nonetheless, Barron found that there was some variation with respect to the sub-strategies employed. For instance, ‘predication of a future act’ was used more frequently by the Irish participants. Also, the Irish were found to use more ‘directive’ strategies than the English in some situations. In terms of external modification, Barron reports that both sets of informants employed two basic types: conditionals and grounders (cf. Aijmer 1996). However, the Irish informants were found to use more (explicit) mitigating conditionals and more grounders than the English informants. Likewise, Breuer and Geluykens (2007) examined how British and American university students realized requests, employing data elicited by means of a DCT. They started out with the hypothesis that the British informants would produce more indirect requests; however, conventional indirectness was found to be the preferred overall strategy of both sets of informants. Nonetheless, they did find some variation in terms of sub-strategies. For example, the British informants employed negation more often than the Americans in their study. Breuer and Gelukyens also predicted that the British informants would make more use of internal and external modification than their American counterparts. This hypothesis was indeed confirmed: the British informants employed a much higher proportion of syntactic and lexical downgrading of the head act and displayed a higher use of external modification. Another speech act that has received some attention is complimenting and responding to compliments (cf. Herbert 1989; Creese 1991; Schneider 1999). One aspect that has caught researchers’ interest in particular is whether there is cultural variability in how people respond to compliments. This interest stems in part from Pomerantz’s (1978) work on compliment responses in American English in which she highlighted the fact that recipients of compliments are faced with a dilemma between expressing agreement with their interlocutor and avoiding self-praise; it also comes from Leech’s (1983) model of politeness, with a Politeness Principle and maxims, including the modesty maxim whereby self-praise is to be minimized. Researchers therefore were interested in investigating how users of different varieties of English deal with Pomerantz’s dilemma and/or whether they adhere to Leech’s modesty maxim. Herbert (1989), for example, compared the form and function of compliment responses among university students in South Africa and the U.S. based on data collected through participant observation. He found that agreement strategies predominated overall in both varieties over disagreement strategies, but were more frequent in the South African English data; he also found some important differences in the frequency of certain sub-strategies. For instance, the strategy Comment Acceptance (e.g. Yeah, it’s cool in response to I like your jacket) was found to be a great deal more frequent in the South African English data than in the American English data. The low frequency of explicit acceptance of this type among speakers of American English would be an indication of their preference for avoiding self-praise; Herbert interprets this and other find-
Regional pragmatic variation
97
ings for Americans in his study as an expression of solidarity, tied to “notions of equality and democratic idealism” (Hebert 1989: 29) that can be found in American society. Schneider (1999), in turn, compared the realization of compliment responses in American English and Irish English with reference to Chen’s (1993) work on compliment responses in American English and Chinese, and Leech’s (1983) theory of politeness. More specifically, Schneider replicated Chen’s study, based on DCTs, in Ireland (Dublin) and compared the results of the corpus that he collected with Chen’s. In terms of super-strategies, Schneider found that the Irish employed one more strategy than Americans and three more than the Chinese. There were also differences in relation to the frequency of use of particular super-strategies that appear in the three data sets. For example, rejecting was the most frequently employed super-strategy in Irish English, although used not nearly as much as in Chinese, while accepting was the most common super-strategy in American English. Grouping the super-strategies according to whether they essentially convey acceptance or rejection, Scheneider found that while Americans appear predominantly to follow Leech’s agreement maxim, and the Chinese, Leech’s modesty maxim, the Irish give relatively equal weight to agreement and modesty. Finally, Jautz (2008) looked at expressions of gratitude in phone-in radio programmes and broadcast interviews in British and New Zealand English. The data she examined comes from the British National Corpus (BNC) (spoken section) and the Wellington Corpus of Spoken New Zealand English (WSC). To start with, one of her findings is that expressions of gratitude are more than twice as frequent in the BNC than in the WSC. In terms of form, thank (you) occurred more frequently in the BNC, and thanks in the WSC. The author suggests that this finding supports the stereotype that “the British are very formal, while New Zealanders are said to be relaxed and easy going” (Jautz 2008: 150). Jautz also examined optional elements that occur with expressions of gratitude. One of them, “naming the benefactor” (Jautz 2008: 150), was found to occur with similar frequencies in both language varieties, with the British corpus showing a slightly higher occurrence of names, and the New Zealand corpus, of terms of endearment. Another optional element examined was that of “naming a reason for one’s gratitude” (Jautz 2008: 153). The author found that New Zealanders produce this element twice as many times as the British. In terms of the functions of expressions of gratitude, the superordinate function that prevailed in both data sets was that of organizing discourse, particularly the closing phase, followed by the “phatic communion” (Malinowski [1923]1972) function. There were some differences, however, in the frequency of these functions in both data sets. For example, there were more instances of the phatic function in the WSC. The author suggests that some of these findings could be interpreted as New Zealanders paying more attention to maintaining interpersonal relations whereas for the British it is “perhaps more of a question of a distancing, non-intruding formality” (Jautz 2008: 172).
98
María Elena Placencia
To turn now to the participation and discourse domains, Tottie (1991) examined backchannel items (Yngve 1970) or listener responses as features of conversational style (Tannen 1984a) in British and American English, based on data from the London-Lund Corpus and the Santa Barbara corpora. One of her key findings was that backchannels were considerably more common in the American English data compared to the British English data. There were also some differences about the frequency of certain forms employed in the two corpora. However, Tottie advises caution in any generalization of her findings given the limited amount of data that she examined (one conversation in AE and two in BE), as well as issues of comparability of the data, raising instead a number of questions such as the extent to which the subject matter determines the use of backchannelling, whether backchannels are elicited by certain types of utterances, etc. The study of regional pragmatic variation in the use of backchannels or listener responses was taken up later by McCarthy (2002) and O’Keefe and Adolphs (2008). McCarthy (2002), like Tottie (1991), offers a corpus linguistics contrastive study of listener responses in British and American English, based in his case on data from the Cambridge and Nottingham Corpus of Discourse in English (CANCODE) and the Cambridge North American Spoken Corpus. The author focuses on non-minimal response tokens (e.g. lovely, absolutely) in everyday conversation whose function goes beyond that of confirming listenership into “show[ing] engagement and interactional bonding with interlocutors” (McCarthy 2002: 49). Indeed, McCarthy (2002: 49) proposes that studying these forms can help understand what constitutes good listenership. Comparing the two data sets, he finds more commonalities than differences. A number of items (e.g. true, great) have similar distributions in both data sets; likewise, double (e.g. right, fine) and triple (e.g. right, right, right) tokens can be found in both varieties, and so forth. The author also observes that many of these tokens perform similar functions in both varieties. In terms of differences, McCarthy finds, for example, that while right is the most common single-word response item in the British English corpus, wow is the most common in the American English corpus. The author suggests, however, that the predominance of right in the British corpus may have to do with the fact that there are more service encounters in the British English data where such markers serve as “transactional boundary markers” (McCarthy 2002: 58). O’Keefe and Adolphs (2008), on the other hand, examine the form and function of listener response tokens (cf. Tottie 1991) in British English and Irish English based on data from the Cambridge and Nottingham Corpus of Discourse in English (CANCODE) and the Limerick Corpus of Irish English (LCIE). The authors find that a broader range of forms are used by speakers of British English, also with a higher frequency. Additionally, there are certain forms like yes and quite that were found only in British English; likewise, are you serious as a response token was found only in Irish English. Interestingly, but perhaps not surprisingly, the authors found a wider range of religious references (e.g. Jesus, Jesus Christ, my
Regional pragmatic variation
99
God) (2008: 78–79) employed as response tokens in Irish English. They suggest that this phenomenon is related to the role of religion in Irish society as a result of which religious references would have a greater “taboo value” in Irish English (O’Keefe and Adolphs 2008: 92). In relation to the function of response tokens, like McCarthy, they did not find any important differences across the data sets that they examined. Finally, the authors highlight the limitations of corpora studies such as theirs, where the researcher has no access to visual clues such as head nods, given that head nods, for example, “could operate as surrogate response tokens” (O’Keefe and Adolphs 2008: 93). To conclude, some of the dimensions that seem relevant to differentiate varieties of English include the directness-indirectness dimension, with an overall preference for indirectness across varieties in the realization of offers and requests surfacing in the above studies. When it comes to tentativeness-terseness among Americans and the British, the orientation of the latter to tentativeness becomes apparent in Breuer and Geluykens’s (2007) study on requests. Looking at British and New Zealand’s English, in Jautz’s (2008) study, involvement and informality appear to be more relevant for New Zealanders vis-à-vis the British who seem to favour independence and formality in the context examined. In terms of variation in the use of backchannel / response tokens, their use, which may be interpreted as signaling involvement, can be tentatively described as a continuum with Americans towards to the pole of higher use, the Irish towards the other end, with the British possibly in the middle. McCarthy (2002) nonetheless did not find any important differences in the use of non-minimal responses between American and British English. O’Keefe and Adolphs (2008) in turn warn us about the possibility of non-verbal mechanisms, not captured in the corpora employed, being used instead of verbal ones. Finally, in relation to responses to compliments, South Africans appear to be more oriented towards agreement than Americans or the Irish, in that order.
6.
Summary and conclusions
This chapter has presented a sketch of regional pragmatic variation as a sub-field of Variational Pragmatics, a recently constituted discipline. It has outlined its basis and development, the levels at which regional pragmatic variation can be analyzed, verbal and non-verbal domains of study as well as dimensions of cross-cultural variation that can be employed in the classification and interpretation of findings. The account of empirical works provided shows the impact of region on language use with respect to a wide range of phenomena including, for example, the realization of speech acts, the use of mitigating devices, address forms, turn-taking and listener responses; however, its impact on certain activities, namely internet based service encounters (see 5.1), seems to be less apparent.
100
María Elena Placencia
The account also shows that while the study of regional pragmatic variation underwent a slow and uneven development in the 1980s and early 1990s, it has experienced a steadier growth since the mid 1990s, especially with respect to Spanish and English, and it has gained pace and momentum within the past few years, with Barron and Schneider (2005) and Schneider and Barron’s (2008b) publications in particular. Yet, for numerous languages, it appears to be an area still awaiting development. The diversity of perspectives and methodologies that can be employed in the study of regional pragmatic variation becomes visible in the account provided. Yet, looking at the range of studies across languages a certain prevalence of speech act studies examined through production questionnaires can be discerned. This can be explained as a result of methodological constraints, namely the difficulty there is in obtaining comparable data (with relative ease). This is a difficulty not exclusive to the study of regional pragmatic variation, but inherent in contrastive/cross-cultural studies (cf. Kasper 2008). Some studies, however, show innovative methodological developments aimed at counteracting some of the problems intrinsic in prototypical production questionnaires. For example, Barron’s (2005b) use of free DCTs allowed her to go beyond the utterance level into the discourse level, enabling the investigation of aspects of the sequential and overall organization of interactions (see also Schneider 2005). Nonetheless, production questionnaires, as researchers who use them acknowledge, give access to (metapragmatic) perceptions of appropriate behaviour; therefore, the wider use of (a combination of) other methodologies is always a desideratum. The corpus linguistics perspective offers a valuable alternative for the qualitative and quantitative contrastive study of a range of pragmatic phenomena. Yet it is not problem free either. There can be comparability issues, as highlighted by McCarthy (2002), among others. Additionally, corpus linguistics does not necessarily give access to enough data on particular speech acts such as requests and offers. Ultimately, the choice of research instrument and hence type of data employed has to be made in relation to the goals pursued by the researcher; also practical considerations need to be taken into account. The interplay of region with other macro-social factors such as age and gender is less apparent in the brief account provided above; it is nevertheless something that certain scholars address and others acknowledge, for example, by restricting their studies to specific age and gender populations, since it is difficult to examine several variables within a single study. For instance, in his study of small talk, Schneider (2008) focuses on female teenagers; Félix-Brasdefer (2008a), on the other hand, looks at the behaviour of male university students in his examination of refusals of certain directives. In terms of the medium of the interactions examined, a prevalence of studies representing face-to-face interactions can also be detected in the field, with comparatively little attention being given to interactions mediated by different technologies. The development of studies of interactions mediated by different technol-
Regional pragmatic variation
101
ogies is therefore another desideratum for future studies. So is the broadening of the study of regional pragmatic variation to encompass a wider range of populations. Again, for reasons of accessibility, there seems to be a preponderance of studies on language use among (normally middle-class) university students between the ages of 18 and 25, although contrastive studies of teenage talk, for instance, are becoming more numerous (see, for example, Barron 2008a, 2008b; Jørgensen 2008; Warga 2008). As far as actual findings are concerned, more or less subtle features of variation and trends across varieties of the same language, and even across languages, can be discerned from the studies that we have surveyed. One feature that Barron (2008a: 360) has observed, and that also surfaces in many of the studies referred to in section 5, is that regional pragmatic variation is visible often less at the level of overall strategies than at the level of sub-strategies and/or linguistic realizations. Also, while speakers of different language varieties usually have the same range of devices at their disposal, what is a matter of variation is the frequency with which certain devices are employed. In relation to dimensions of cross-cultural variation, looking at the results of studies in different languages, one can see the usefulness of the range of dimensions that have been proposed. Among the dimensions that recur (implicitly or explicitly) in addition to the prototypical directness-indirectness dichotomy are the following: involvement vs. independence; tentativeness vs. terseness; verbosity vs. restraint; deference/hierarchy vs. equality; person- vs. task-orientedness, etc. Nonetheless, caution needs to be exerted in making generalizations and extrapolations to the larger population on the basis of the results of isolated studies without taking into account, for example, possible situational as well as regional variation at the sub-national level. As Barron (2008a: 362) rightly observes, research in regional pragmatic variation is still at “a very early stage”. Also, in comparing national varieties, there is the danger of overgeneralizing and “glossing over internal differences”, to use Terkourafi’s (2007: 62) phrase. Placencia’s (2008) study (section 5.1), for example, shows differences between two dialectal areas in Ecuadorian Spanish that suggest perhaps more similarities in communicative style between one of the Ecuadorian sub-varieties examined and Madrileño Spanish than between the two Ecuadorian sub-varieties. Also, it can be problematic to make generalizations for given groups across studies because different frameworks of analysis are employed in different studies, the data sets are often not comparable, and there may not be agreement among researchers about key notions such as (in)directness. In this sense, a comprehensive study across varieties of different languages such as English or Spanish of the scale of Blum-Kulka et al.’s (1989a), where the same methodology and analytical framework across studies are employed, is a desideratum for the future of regional pragmatic variation studies. On the other hand, the focus on variation may obscure similarities across varieties. This is something that needs to be borne in mind in studies of (regional)
102
María Elena Placencia
pragmatic variation. Contrastive studies are, precisely, necessary “for an overall understanding of the common ground that typically exists alongside differences between one variety and another.” (McCarthy 2002: 70). McCarthy refers to crosscorpora intervarietal studies in particular, but his observations also apply to intervarietal studies from other perspectives. In fact, as McCarthy also maintains, intervarietal studies are essential in that “they may confirm or may refute functional interpretations made from one variety or language as not being idiosyncratic to one culture or speech community”, hence “it is only in intervarietal and interlingual studies that one can find safer ground for generalisations” (McCarthy 2002: 69). The greater fluidity between contemporary societies, with increased borderless communication (Jiang 2006), brought about by globalization forces, including wider access to modern technologies and increased population movements, has led some critics to predict the demise of national cultures and consequently, one might say, of pragmatic variation across national language varieties. Indeed some suggest that future research should possibly focus on commonalities of behaviour rather than cultural variation (cf. Martinell Gifre 2007).12 However, the majority of the studies surveyed in this chapter attest to the interest there still is in studying regional pragmatic variation and the likelihood that this will continue for some considerable time yet. It is, of course, probable that the study of certain spheres of social life such as computer-mediated and media discourse, areas significantly under-researched within regional pragmatic variation, may show less variation than others. These are but two of a number of avenues of scholarly investigation which are currently opening up in this relatively new field of research.
Notes 1. Other authors refer to this level as the intra-national level (cf. Holmes 2008). 2. Stewart (1968) employed the term polycentric instead. 3. This can be seen in introductions to French linguistics from the 1990s and even more recently, such as Battye, Hintze and Rowlett ([1992]2000), where consideration of regional variation is limited to phonological, lexical and morphosyntactic features, or Antes’s (2006) which purports to provide “a complete examination of the French language” (book cover), but focuses on phonetics, phonology, morphology, syntax, and semantics, only. Fagyal, Kibbee and Jenkins (2006), on the other hand, seems to be an exception in this respect as it includes a chapter on pragmatics. 4. The reader is referred to the detailed overview that Cortés Rodríguez (2002) provides of works dealing with Spanish language varieties in the second half of the 20th century, including studies in traditional dialectology, social/urban dialectology as well as variational sociolinguistics. 5. See Blum-Kulka, House and Kasper (1989b) for an overview of contrastive pragmatic studies in the late 1970s and early 1980s. 6. And yet, it is in manuals linked to EFL, for example, where impressionistic observations
Regional pragmatic variation
7.
8.
9. 10.
11.
12.
103
of regional pragmatic variation seem to abound. See, for example, Ronowicz and Yallop’s (2007) collection of papers. Cross-cultural pragmatics, as we know, is also closely associated with the works of Anna Wierzbicka (see for example Wierzbicka 1985; 1991); her influence in studies of regional pragmatic variation, however, can only be seen indirectly in the fact that most of her published work sets out to examine the influence of culture on language use. On the other hand her framework, built as it is around the notion of cultural scripts (cf. Wierzbicka 1997; Goddard and Wierzbicka 2004), has not as yet had much direct influence in this area. This dimension resembles House’s (1996, 2000) orientation towards content vs. orientation towards addressee. Related notions can be found, for example, in Triandis (1967: 51) where Greeks’ and Americans’ approaches to decision-taking in organizations is described as “friendship-oriented” for Greeks and “efficiency-oriented” for Americans. Bateson’s (1972) notion of “frames of interpretation” has also been employed in cross-cultural studies to explain differences in the way participants approach a particular communicative activity, as “this is business” or “this is a friendly encounter” (cf. García 1989). More recently, Gudykunst and Kim (1997) contrast “personalness” with “impersonalness,” and Spencer-Oatey and Jiang (2003), “concern for face/rapport” with “concern for task.” See, for example, Schneider and Barron’s (2008a) categorization of levels of pragmatic analysis into formal, actional, interactional, topic, and organisational. The reader is also referred to Kuiper and Tan Gek Lin (1989) as another example of work that shows some of the pragmatic particularities of a variety of English – Singapore English in this case. The free discourse completion task (FDCT) is a kind of production questionnaire aimed at eliciting sequential aspects of the speech acts under examination by requiring respondents to imagine themselves in particular scenarios and to recreate in writing the full interaction with an interlocutor in the imagined scenario (Barron 2005b: 147–148). Martinell Gifre (2007) refers to the study of non-verbal communication specifically.
References Aijmer, Karin 1996 Conversational Routines in English: Convention and Creativity. London: Longman. Alba-Juez, Laura 2008 Sobre algunas estrategias y marcadores de descortesía en español peninsular y argentino: ¿Son españoles y argentinos igualmente descorteses? In: Antonio Briz, Antonio Hidalgo, Marta Albelda, Josefa Contreras and Nieves Hernández Flores (eds.), Cortesía y conversación: de lo escrito a lo oral. III Coloquio Internacional del Programa EDICE (CD-ROM), 80–97. Valencia: Universitat de València. Albelda Marco, Marta 2008 Atenuantes en Chile y en España: distancia o acercamiento. In: Antonio Briz, Antonio Hidalgo, Marta Albelda, Josefa Contreras and Nieves Hernández
104
María Elena Placencia
Flores (eds.), Cortesía y conversación: de lo escrito a lo oral. III Coloquio Internacional del Programa EDICE (CD-ROM) 98–113. Valencia: Universitat de València. Antes, Theresa A. 2006 Analyse linguistique de la langue française. New Haven, CT ; London: Yale University Press. Barron, Anne 2005a Variational pragmatics in the foreign language classroom. System 33: 519–536. Barron, Anne 2005b Offering in Ireland and England. In: Anne Barron and Klaus P. Schneider (eds.), The Pragmatics of Irish English, 141–177. Berlin: Mouton de Gruyter. Barron, Anne 2008a Contrasting requests in Inner Circle Englishes: A study in variational pragmatics. In: Martin Pütz and JoAnne Neff-van Aertselaer (eds.), Developing Contrastive Pragmatics: Interlanguage and Cross-cultural Perspectives, 355–402. Berlin: Mouton de Gruyter. Barron, Anne 2008b The structure of requests in Irish English and English English. In: Klaus P. Schneider and Anne Barron (eds.), Variational Pragmatics. A Focus on Regional Varieties in Pluricentric Languages, 35–67. Amsterdam: John Benjamins. Barron, Anne and Klaus P. Schneider 2005 The Pragmatics of Irish English. Berlin: Mouton de Gruyter. Bateson, Gregory 1972 Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution, and Epistemology. Aylesbury: Intertext. Battye, Adrian, Marie-Anne Hintze and Paul Rowlett [1992]2000 The French Language Today: A Linguistic Introduction. London: Routledge. Baxter, Alan N. 1992 Portuguese as a pluricentric language In: Michael Clyne (ed.), Pluricentric Languages: Differing Norms in Different Nations, 11–43. Berlin: Mouton de Gruyter. Birkner, Karin and Friederike Kern 2000 Impression management in East and West German job interviews. In: Helen Spencer-Oatey (ed.), Culturally Speaking: Managing Rapport through Talk across Cultures, 255–271. London: Continuum. Blum-Kulka, Shoshana, Juliane House and Gabriele Kasper (eds.) 1989a Cross-cultural Pragmatics: Requests and Apologies. Norwood, NJ: Ablex. Blum-Kulka, Shoshana, Juliane House and Gabriele Kasper 1989b Investigating cross-cultural pragmatics: An introductory overview. In: Shoshana Blum-Kulka, Juliane House and Gabriele Kasper (eds.), Cross-cultural Pragmatics: Requests and Apologies, 1–34. Norwood, NJ: Ablex Blum-Kulka, Shoshana and Elite Olshtain 1984 Request and apologies: A cross-cultural study of speech act realization patterns (CCSARP). Applied Linguistics 5: 196–213. Bravo, Diana 1998 ¿Reírse juntos? Un estudio de las imágenes sociales de hablantes españoles mexicanos y suecos. In: Henk Haverkate, Gijs Mulder and Carolina Fraile
Regional pragmatic variation
105
Maldonado (eds.), La pragmática lingüística del español. Recientes desarrollos, Diálogos Hispánicos 22, 315–364. Amsterdam: Rodopi. Breuer, Anja and Ronald Geluykens 2007 Variation in British and American English requests: A contrastive analysis. In: Bettina Kraft and Ronald Geluykens (eds.), Cross-cultural Pragmatics and Interlanguage English,107–126. Munich: Lincom. Brown, Penelope and Stephen C. Levinson [1978]1987 Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Chen, Rong 1993 Responding to compliments. A contrastive study of politeness strategies between American English and Chinese speakers. Journal of Pragmatics 20: 49–75. Chen, Shu-hui Eileen 2003 Compliment response strategies in Mandarin Chinese: Politeness phenomenon revisited. Concentric: Studies in English Literature and Linguistics 29: 157– 184. Clyne, Michael 1992 Pluricentric languages. Introduction. In: Michael Clyne (ed.), Pluricentric Languages: Differing Norms in Different Nations, 1–9. Berlin: Mouton de Gruyter. Clyne, Michael, Sue Fernandez and Rudolf Muhr 2003 Communicative styles in a contact situation: Two German national varieties in a third country. Journal of Germanic Linguistics 15: 95–154. Clyne, Michael, Heinz-Leo Kretzenbacher, Catrin Norrby and Doris Schupbach 2006 Perceptions of variation and change in German and Swedish address. Journal of Sociolinguistics 10: 287–319. Cortés Rodríguez, Luis 2002 Los estudios del español hablado entre 1950 y 1999. Anejo de Oralia. Madrid: Arco/Libros. Creese, Angela 1991 Speech act variation in British and American English. Penn Working Papers in Educational Linguistics 7: 37–58. Curcó, Carmen 1998 No me harías un favorcito?: Reflexiones en torno a la expresión de la cortesía verbal en el español de México y el español peninsular. In: Henk Haverkate, Gijs Mulder and Carolina Fraile Maldonado (eds.), La pragmática lingüística del español. Recientes desarrollos, Diálogos Hispánicos 22, 129–171. Amsterdam: Rodopi. Curcó, Carmen and Ana de Fina 2002 Modo imperativo, negación y diminutivos en la expresión de la cortesía en español: el contraste entre México y España. In: M.E. Placencia and D. Bravo (eds.), Actos de habla y cortesía en español, 107–140. Munich: Lincom. De los Heros, Susana 2001 Discurso, identidad y género en el castellano peruano. Lima: Pontificia Universidad Católica del Perú/Fondo Editorial.
106
María Elena Placencia
Fagyal, Zsuzsanna, Douglas A. Kibbee and Fred Jenkins 2006 French: A Linguistic Introduction. Cambridge: Cambridge University Press. Fant, Lars 1989 Cultural mismatch in conversation: Spanish and Scandinavian communicative behaviour in negotiation settings. Hermes Journal of Linguistics 3: 247–265. Fant, Lars 1995 Negotiation discourse and interaction in a cross-cultural perspective: The case of Sweden and Spain. In: Konrad Ehlich and Johannes Wagner (eds.), The Discourse of Business Negotiation, 177–201. Berlin/New York: Mouton de Gruyter. Fant, Lars 1996 Regulación conversacional en la negociación: una comparación entre pautas mexicanas y peninsulares. In: Thomas Kotschi, Wulf Oesterreicher and Klaus Zimmerman (eds.), El español hablado y la cultura oral en España e Hispanoamérica, 147–183. Frankfurt/Madrid: Vervuert-Iberoamericana. Félix-Brasdefer, César 2008a Sociopragmatic variation: Dispreferred responses in Mexican and Dominican Spanish. Journal of Politeness Research 4: 81–110. Félix-Brasdefer, César 2008b Peticiones y prosodia: variación pragmática regional en México y la República Dominicana. Paper presented at the XV Congreso Internacional de la Asociación de Linguística y Filología de América Latina (Montevideo, 18–21 August 2008). Garcés-Conejos Blitvich, Pilar and Patricia Bou Franch 2008 Cortesía en las páginas web interactivas. El comercio electrónico. In: Antonio Briz, Antonio Hidalgo, Marta Albelda, Josefa Contreras and Nieves Hernández Flores (eds.), Cortesía y conversación: de lo escrito a lo oral. III Coloquio Internacional del Programa EDICE (CD-ROM), 468–488. Valencia: Universitat de València. García, Carmen 1989 Disagreeing and requesting by Americans and Venezuelans. Linguistics and Education 1: 299–322. García, Carmen 2004 Reprendiendo y respondiendo a una reprimenda. Similitudes y diferencias entre peruanos y venezolanos. Spanish in Context 1: 113–147. García, Carmen 2008 Different realizations of solidarity politeness: Comparing Venezuelan and Argentinean invitations. In: Klaus P. Schneider and Anne Barron (eds.), Variational Pragmatics: A Focus on Regional Varieties in Pluricentric Languages, 269–305. Amsterdam: John Benjamins. García, Carmen 2009 Intra-lingual pragmatic variation in the performance of reprimanding. Intercultural Pragmatics 6: 443–472. Goddard, Cliff and Anna Wierzbicka 2004 Cultural scripts: What are they and what are they good for? Intercultural Pragmatics 1: 167–187.
Regional pragmatic variation
107
Grindsted, Annette 2000 Categorías gramaticales y pragmática en el español mexicano y el peninsular. Journal of the Modern Language Teachers’ Association of Sweden 94: 205– 213. Gudykunst, William B. and Young Yun Kim 1997 Communicating with Strangers: An Approach to Intercultural Communication. New York: McGraw-Hill. Hall, Edward T. 1966 The Hidden Dimension. Garden City, N.Y.: Doubleday. Hall, Stuart 1992 The question of cultural identity. In: Stuart Hall, David Held and Tony McGrew (eds.), Modernity and its Futures, 273–325. Cambridge: Polity Press / The Open University Press. Hardin, Karol J. 2001 Pragmatics in Persuasive Discourse of Spanish Television Advertising. Dallas, Tex.: SIL International. Herbert, Robert K. 1989 The ethnography of English compliments and compliment responses: A contrastive sketch. In: Wieslaw Oleksy (ed.), Contrastive Pragmatics, 3–35. Amsterdam: John Benjamins. Herbert, Robert K. and Straight, H. Stephen 1989 Compliment-rejection versus compliment-avoidance: Listener-based versus speaker-based pragmatic strategies. Language & Communication 9: 35–47. Hofstede, Geert 1980 Culture’s Consequences: International Differences in Work-related Values. Beverly Hills ; London: Sage. Hofstede, Geert 1991 Cultures and Organizations: Software of the Mind. London: McGraw-Hill. Holmes, Janet 2008 An Introduction to Sociolinguistics. Harlow: Pearson/Longman. House, Juliane 1996 Contrastive discourse analysis and misunderstanding: The case of German and English. In: Marlis Hellinger and Ulrich Ammon (eds.), Contrastive Sociolinguistics, 345–361. Berlin; New York: Mouton de Gruyter. House, Juliane 2000 Understanding misunderstanding: A pragmatic-discourse approach to analysing mismanaged rapport in talk across cultures. In: Helen Spencer-Oatey (ed.), Culturally Speaking: Managing Rapport through Talk across Cultures, 145– 164. London: Continuum. Jautz, Sabine 2008 Gratitude in British English and New Zealand radio programmes. Nothing but gushing? In: Klaus P. Schneider and Anne Barron (eds.), Variational Pragmatics: A Focus on Regional Varieties in Pluricentric Languages, 141–178. Amsterdam: John Benjamins. Jensen, John B. 1981 Forms of address in Brazilian Portuguese: Standard European or Oriental honorifics? In: Francis Millet Rogers and Bernard H. Bichakjian (eds.), From
108
María Elena Placencia
Linguistics to Literature: Romance Studies Offered to Francis M. Rogers, 45–66. Amsterdam: John Benjamins. Jiang, Xiaoping 2006 Towards intercultural communication: From micro to macro perspectives. Intercultural Education 17: 407–419. Jørgensen, Annette M. 2008 Diferencias y similitudes en el uso de los vocativos en el lenguaje juvenil de Madrid y Santiago de Chile. Paper presented at the XV Congreso Internacional de la Asociación de Linguística y Filología de América Latina (Montevideo, 18–21 August, 2008). Kachru, Braj B. 1985 Standards, codification and sociolinguistic realism: The English language in the Outer Circle. In: Randolph Quirk and Henry G. Widdowson (eds.), English in the World: Teaching and Learning the Languages and Literatures, 11–36. Cambridge: Cambridge University Press (for the British Council). Kallen, Jeffrey L. 2005 Silence and mitigation in Irish English discourse. In: Anne Barron and Klaus P. Schneider (eds.), The Pragmatics of Irish English, 47–71. Berlin / New York: Mouton de Gruyter. Kasper, Gabriele 2008 Data collection in pragmatics research. In: Helen Spencer-Oatey (ed.), Culturally Speaking. Culture, Communication and Politeness Theory, 279–303. London: Continuum. Kloss, Heinz 1978 Die Entwicklung neuer germanischer Kultursprachen seit 1800 [i.e. achtzehnhundert]. Düsseldorf: Pädagogischer Verlag Schwann. Kuiper, Koenraad and Daphne Tan Gek Lin 1989 Cultural congruence and conflict in the acquisition of formulae in a second language. In: Ofelia García and Ricardo Otheguy (eds.), English across Cultures. Cultures across English. A Reader in Cross-cultural Communication, 281– 304. Berlin: Mouton de Gruyter. Leech, Geoffrey N. 1983 Principles of Pragmatics. London: Longman. Locher, Miriam A. and Richard J. Watts 2005 Politeness theory and relational work. Journal of Politeness Research 1: 9–33. Malinowski, Bronislaw [1923]1972 Phatic communion. In: John Laver and Sandy Hutcheson (eds.), Communication in Face-to-Face Interaction, 146–152. Harmondsworth, Middlesex: Penguin. Manes, Joan and Nessa Wolfson 1981 The compliment formula. In: Florian Coulmas (ed.), Conversational Routine: Explorations in Standardised Communication Situations and Prepatterned Speech, 115–132. The Hague: Mouton. Márquez Reiter, Rosina 2002 A contrastive study of conventional indirectness in Spanish: Evidence from Peninsular and Uruguayan Spanish. Pragmatics 12: 135–151.
Regional pragmatic variation
109
Márquez Reiter, Rosina 2003 Pragmatic variation in Spanish: External request modifications in Uruguayan and Peninsular Spanish. In: Rafael Núñez Cedeño, Luis López and Richard Cameron (eds.), A Romance Perspective on Language Knowledge and Use, 166–180. Amsterdam: John Benjamins. Márquez Reiter, Rosina and María Elena Placencia 2004 Displaying closeness and respectful distance in Montevidean and Quiteño service encounters. In: Rosina Márquez Reiter and María Elena Placencia (eds.), Current Trends in the Pragmatics of Spanish, 121–155. Amsterdam: John Benjamins. Martinell Gifre, Emma 2007 La gestualidad hoy, en el marco de la competencia intercultural y de la tendencia a la globalización. In: Enrique Balmaseda Maestu (ed.), Las destrezas orales en la enseñanza del español L2-LE: Logroño, 27–30 de septiembre de 2006, 65–82. La Rioja: Universidad de La Rioja. Martínez Camino, Gonzalo 2008 Análisis comparativo del dominio interactivo en los diálogos ficticios de la publicidad mexicana y de la española. Paper presented at Congreso Internacional sobre (des)cortesía y violencia verbal en español (Seville, 5–7 Nevember, 2008). McCarthy, Michael 2002 Good listenership made plain: British and American non-minimal response tokens in everyday conversation: In: Randi Reppen, Susan M. Fitzmaurice and Douglas Biber (eds.), Using Corpora to Explore Linguistic Variation, 49–72. Amsterdam: John Benjamins. Mesthrie, Rajend and Rakesh Mohan Bhatt 2008 World Englishes: The Study of New Linguistic Varieties. Cambridge ; New York: Cambridge University Press. Muhr, Rudolf 1987 Regionale Unterschiede im Gebrauch von Beziehungsindikatoren zwischen der Bundesrepublik Deutschland und Österreich und ihre Auswirkungen auf den Unterricht in Deutsch als Fremdsprache – dargestellt am Beispiel der Modalpartikel. In: Lutz Götze (ed.), Deutsch als Fremdsprache: Situation eines Faches, 144–156. Bonn-Bad Godesberg: Dürr. Muhr, Rudolf 1993 Pragmatische Unterschiede in der deutschsprachigen Kommunikation: Österreich-Deutschland. In: Rudolf Muhr (ed.), Internationale Arbeiten zum österreichischen Deutsch und seinem nachbarsprachlichen Bezügen, 26–38. Wien: Hölder-Pichler-Tempsky. Muhr, Rudolf 1994 ‘Entschuldigen Sie Frau Kollegin …’: Sprechaktrealisierungsunterschiede an Universitäten in Österreich und Deutschland. In: Gudrun Held (ed.), Verbale Interaktion, 126–143. Hamburg: Kovacˇ. Muhr, Rudolf 2008 The pragmatics of a pluricentric language. A comparison between Austrian German and German German. In: Klaus P. Schneider and Anne Barron (eds.), Variational Pragmatics: A Focus on Regional Varieties in Pluricentric Languages, 211–244. Amsterdam: John Benjamins.
110
María Elena Placencia
O’Keefe, Anne and Svenja Adolphs 2008 Response tokens in British and Irish discourse. Corpus, context and variational pragmatics. In: Klaus P. Schneider and Anne Barron (eds.), Variational Pragmatics. A Focus on Regional Varities in Pluricentric Languages, 69–98. Amsterdam: John Benjamins. Placencia, María Elena 1994 Pragmatics across varieties of Spanish. Donaire 2: 65–77. Placencia, María Elena 1998 Pragmatic variation: Ecuadorian Spanish vs Peninsular Spanish. Spanish Applied Linguistics 2: 71–106. Placencia, María Elena 2005 Pragmatic variation in corner store interactions in Quito and Madrid. Hispania 88: 583–598. Placencia, María Elena 2008 Pragmatic variation in corner shop transactions in Ecuadorian Andean and Coastal Spanish. In: Klaus P. Schneider and Anne Barron (eds.), Variational Pragmatics. A Focus on Regional Varieties in Pluricentric Languages, 307–332. Amsterdam: John Benjamins. Plevoets, Koen, Dirk Speelman and Dirk Geeraerts 2008 The distribution of T/V pronouns in Netherlandic and Belgian Dutch. In: Klaus P. Schneider and Anne Barron (eds.), Variational Pragmatics. A Focus on Regional Varieties in Pluricentric Languages, 181–209. Amsterdam: John Benjamins. Pomerantz, Anita 1978 Compliment responses: Notes on the co-operation of multiple constraints. In: Jim Schenkein (ed.), Studies in the Organization of Conversational Interaction, 79–112. New York: Academic Press. Poyatos, Fernando 2002 Nonverbal Communication across Disciplines. Amsterdam: John Benjamins. Puga Larraín, Juana 1997 La atenuación en el castellano de Chile: un enfoque pragmalingüístico. Valencia: Tirant Lo Blanch Libros, Universitat de València. Puga Larraín, Juana 1999 Elusión e insinuación: la atenuación en el castellano de Chile. RLA Revista de Lingüística Teórica y Aplicada 37: 123–138. Reuter, Mikael 1992 Swedish as a pluricentric language. In: Michael Clyne (ed.), Pluricentric Languages: Differing Norms in Different Nations, 101–116. Berlin: Mouton de Gruyter. Ronowicz, Eddie and Colin Yallop 2007 English: One Language, Different Cultures. London: Continuum. Schlieben-Lange, Brigitte and Harald Weydt 1978 Für eine Pragmatisierung der Dialektologie. Zeitschrift für Germanistische Linguistik 6: 257–282. Schneider, Klaus P. 1999 Compliment responses across cultures. In: Maria Wysocka (ed.), On Language in Theory and Practice: In Honour of Janusz Arabski on the Occasion of His
Regional pragmatic variation
111
60th Birthday. Vol 1: Language Theory and Language Use, 162–172. Katowice: Wydawnictwo Uniwersytetu S´la˛kiego. Schneider, Klaus P. 2003 Diminutives in English. Tübingen: Niemeyer. Schneider, Klaus P. 2005 No problem, you’re welcome, anytime: Responding to thanks in Ireland, England, and the USA. In: Anne Barron and Klaus P. Schneider (eds.), The Pragmatics of Irish English, 101–139. Berlin: Mouton de Gruyter. Schneider, Klaus P. 2008 Small talk in England, Ireland, and the U.S.A. In: Klaus P. Schneider and Anne Barron (eds.), Variational Pragmatics. A Focus on Regional Varieties in Pluricentric Languages, 99–139. Amsterdam: John Benjamins. Schneider, Klaus P. and Anne Barron 2005 Variational pragmatics: Contours of a new discipline. 9th International Pragmatics Conference Riva del Garda, July 10–15, 2005. Schneider, Klaus P. and Anne Barron 2008a Where pragmatics and dialectology meet: Introducing variational pragmatics. In: Klaus P. Schneider and Anne Barron (eds.), Variational Pragmatics: A Focus on Regional Varieties in Pluricentric Languages, 1–32. Amsterdam: John Benjamins. Schneider, Klaus P. and Anne Barron 2008b Variational Pragmatics: A focus on Regional Varieties in Pluricentric Languages. Amsterdam: John Benjamins. Schölmberger, Ursula 2008 Apologizing in French French and Canadian French. In: Klaus P. Schneider and Anne Barron (eds.), Variational Pragmatics. A Focus on Regional Varieties in Pluricentric Languages, 333–354. Amsterdam: John Benjamins. Scollon, Ronald and Suzanne W. Scollon [1995] 2001 Intercultural Communication: A Discourse Approach. Second edition. Oxford, UK ; Cambridge, Mass.: Blackwell. Silva-Brummel, Maria Fernanda 1984 As formas de tratamento no português angolano. In: Günter Holtus, Heinz Kröll and Edgar Radtke (eds.), Umgangssprache in der Iberoromania: Festschrift für Heinz Kröll, 271–286. Tübingen: Narr. Spencer-Oatey, Helen 2000 Rapport management: A framework for analysis. In: Helen Spencer-Oatey (ed.), Culturally Speaking: Managing Rapport through Talk across Cultures, 11–46. London: Continuum. Spencer-Oatey, Helen 2008a Introduction. In: Helen Spencer-Oatey (ed.), Culturally Speaking. Culture, Communication and Politeness Theory, 1–8. London: Continuum. Spencer-Oatey, Helen 2008b Face, (im)politeness and rapport. In: Helen Spencer-Oatey (ed.), Culturally Speaking. Culture, Communication and Politeness Theory, 11–47. London: Continuum.
112
María Elena Placencia
Spencer-Oatey, Helen, Patrick Ng and Li Dong 2000 Responding to compliments: British and Chinese evaluative judgements. In: Helen Spencer-Oatey (ed.), Culturally Speaking: Managing Rapport in Talk across Cultures, 98–120. London: Continuum. Spencer-Oatey, Helen and Wenying Jiang 2003 Explaining cross-cultural pragmatic findings: Moving from politeness maxims to sociopragmatic interactional principles (SIPs). Journal of Pragmatics 35: 1633–1650. Sridhar, Kamal K. 1996 The pragmatics of South Asian English. In: Robert J. Baumgardner (ed.), South Asian English: Structure, Use, and Users, 141–157. Urbana: University of Illinois Press. Stewart, William 1968 A sociolinguistic typology for describing national multilingualism. In: Joshua Fishman (ed.), Readings in the Sociology of Languages, 531–545. The Hague: Mouton. Tannen, Deborah 1984a Conversational Style: Analyzing Talk among Friends. Norwood, N.J.: Ablex. Tannen, Deborah 1984b The pragmatics of cross-cultural communication. Applied Linguistics 5: 189–195. Terkourafi, Marina 1999 Frames for politeness: A case study. Pragmatics 9: 97–117. Terkourafi, Marina 2007 Perceptions of difference in the Greek sphere: The case of Cyprus. Journal of Greek Linguistics 8: 60–96. Thurén, Britt-Marie 1988 Left Hand Left Behind: The Changing Gender System of a Barrio in Valencia, Spain. Stockholm: University of Stockholm. Tottie, Gunnel 1991 Conversational style in British and American English: The case of backchannels. In: Karin Aijmer and Bengt Altenberg (eds.), English Corpus Linguistics, 254–271. London: Longman. Triandis, Harry C. 1967 Interpersonal relations in international organizations. Organizational Behaviour and Human Performance 2: 26–55. Wagner, Lisa C. 1999 Towards a sociopragmatic characterization of apologies in Mexican Spanish. Unpublished doctoral disseration. Ohio State University. Warga, Muriel 2008 Requesting in German as a pluricentric language. In: Klaus P. Schneider and Anne Barron (eds.), Variational Pragmatics. A Focus on Regional Varieties in Pluricentric Languages, 245–266. Amsterdam: John Benjamins. Wierzbicka, Anna 1985 Different cultures, different languages, different speech acts: Polish vs. English. Journal of Pragmatics 9: 145–178. Wierzbicka, Anna 1991 Cross-cultural Pragmatics: The Semantics of Human Interaction. Berlin; New York: Mouton de Gruyter.
Regional pragmatic variation
113
Wierzbicka, Anna 1997 Understanding Cultures through Their Key Words: English, Russian, Polish, German, and Japanese. New York: Oxford University Press. Wierzbicka, Anna 2005 In defense of ‘culture’. Theory & Psychology 15: 575–597. Wilhelm, Eberhard Axel 1979 Pronomes de distância do português actual em Portugal e no Brasil. Lisboa: Instituto Nacional de Investigação Científica. Wolfson, Nessa 1983 Rules of speaking. In: Jack C. Richards and Richard W. Schmidt (eds.), Language and Communication, 61–87. London: Longman. Yngve, Victor H. 1970 On getting a word in edgewise. Papers from the Sixth Regional Meeting, Chicago Linguistic Society, 567–578. Chicago, IL: University of Chicago.
4.
Pragmatics in multilingual language situations Susanne Mühleisen
1.
Introduction
Pragmatic research has often assumed universality in patterns of verbal behaviour (e.g. Brown and Levinson 1987) or investigated the production and interpretation of meaning in seemingly homogeneous speech communities. However, the implicit equation “one language = one culture = one community”, often found in the description of speech act realisations in one language like German or Polish, or a comparison of two languages, a highly prolific field in cross-cultural pragmatics ever since Blum-Kulka and Olshtain’s (1984) Cross-Cultural Speech Act Research Project (CCSARP, also in Blum-Kulka et al. 1989), does not do justice to many language situations where more than one language is used by all members of the community, or by part of the group. Here, interactional verbal behaviour may vary considerably according to the individual speaker’s affiliation with a linguistic, cultural or ethnic group and the values that are associated with it. In politeness studies in particular, a neglect of such complexities as well as an overgeneralisation of politeness rules with respect to a particular group and/or country might then contribute to a production or reproduction of national stereotypes. In a recent volume of politeness studies in Europe (Hickey and Stewart 2005), the editors recognise this potential trap in their introduction: It emerges […] that language is not necessarily co-terminous with culture and a number of the chapters address issues such as the differences between rural and urban norms […] Furthermore, within the same country there may co-exist different, often conflicting, language communities. (Hickey and Stewart 2005: 2)
Hence, some contributions in Hickey and Stewart’s collection that deal with multilingual countries in Europe, for instance, Switzerland (Manno 2005), Luxemburg (Kramer 2005) and Belgium (Danblon, de Clerck and van Noppen 2005), do consider varied politeness behaviour with regard to the language groups within the country. Regional and national variation of pluricentric languages such as English, Spanish, French or German has also been largely neglected in pragmatic studies, perhaps with the exception of obvious differences between British and American English (e.g., discourse markers, cf. Lenk 1998). There are also some studies on particular speech acts in which their realisation in different varieties of English is compared, for example, on compliments in American English and South African English (Herbert 1989) or on requests in Indian versus Singaporean English (Kachru 1998). However, such studies are relatively few in comparison to interlin-
116
Susanne Mühleisen
gual speech act comparisons. While this gap is slightly different in focus from pragmatics in multilingual and postcolonial societies, the basic problem remains the same: an idealisation of a homogeneous group of speakers of a language and their assumed uniform verbal behaviour regardless of context, intralingual variation and the possibility of co-existing cultural variation. As Wierzbicka (1991) notes with regard to the area of speech acts, “from the outset, studies in speech acts have suffered from an astonishing ethnocentrism” (Wierzbicka 1991: 25). Her recognition of intralinguistic pragmatic variation must be seen as rather an exception at the time of publication: The cultural norms reflected in speech acts differ not only from one language to another, but also from one regional and social variety to another. There are considerable differences between Australian English and American English, between mainstream American English and American Black English, between middle-class English and workingclass English, and so on (Wierzbicka 1991: 26).
A recent volume (Schneider and Barron 2008) on Variational Pragmatics addresses this lacuna and explores intralingual pragmatic differences between, for instance, Irish English and British English, Netherlandic and Belgian Dutch, Austrian and German German, or Venezuelan and Argentinian Spanish. But what about pragmatic meaning and interpretation in Nigerian, Cameroonian or South African English, in Caribbean Englishes and Creoles, in New Zealand English spoken by Pakehas (New Zealanders of European descent) versus Maoris? In such postcolonial language situations, value and norm systems (e.g. European versus non-European) can often be found to be parallel or competing and their evaluation has to include dimensions of power, agency and identity. Apart from substantial work on New Zealand English (e.g. Holmes 1995, 2003, 2005, Stubbe 1998, Holmes and Stubbe 2004), few studies (e.g. Kasanga and LwangaLumu 2008) have been conducted in postcolonial language situations and only recently, whole volumes have been devoted to issues in postcolonial pragmatics (Mühleisen and Migge 2005, Anchimbe and Janney 2011). This contribution explores verbal behaviour (speech act realisations, organisation of conversation) and associated values (e.g. directness/indirectness) in multilingual and postcolonial societies where, it is argued, norms of behaviour and interpretation show a discrepancy according to factors such as linguistic, social or ethnic group. In the following, three different types of multilingual or postcolonial language situations will be discussed: Firstly, in section 2.1, some European multilingual countries (Switzerland, Belgium and Luxemburg) will serve as a case of pragmatic variation according to language group. The discussion will focus on language situations with established groups of speaker in these countries rather than on multilingual situations due to recent migration movements. While the latter situation also produces an interesting site for the exploration of intercultural communication issues (see, for instance,
Pragmatics in multilingual language situations
117
Clyne’s 1999 work on pragmatics in work situations in Australia’s multilingual situation), its complexities including issues of second language acquisition and native/ non-native speaker communication go beyond the scope of the present paper. Secondly, in section 2.2, Holmes’s (e.g. 1995, 2003) research on differences in pragmatic behaviour between Pakeha and Maori speakers in New Zealand will serve as a first case in point for variation according to ethnic identity in postcolonial language situations. The third situation will be a more detailed investigation of postcolonial ambiguity and pragmatic variation in the Anglophone Caribbean. It is argued that, in the complex diglossic Creole/English language situation of Caribbean countries, pragmatic variation does not correspond straightforwardly with certain linguistic or ethnic groups but is part of a complex negotiation of one’s alignment with Creole versus English/ European values and communication norms. In this paper, I am limiting my discussion to those areas which are characterised by a Creole-English diglossic situation (e.g. Jamaica or Trinidad and Tobago). There are, however, other territories where the multilingual situation is more complex: in the Central American country Belize, for instance, a large proportion of speakers use Spanish, or, to a lesser extent, Garifuna or Maya in addition or as an alternative to Creole and English. In section 2.3, two brief case studies as examples of pragmatics in a postcolonial environment, i.e., in the Eastern Caribbean country Trinidad and Tobago, will be included.
2.
Pragmatics in multilingual language situations
One of the reasons why multilingual and postcolonial language situations are highly interesting for pragmatics is the very connection between language practices and cultural values and norms of behaviour. Areas such as politeness strategies, speech act realisations or conversational routines have been found to be particularly culture-sensitive, despite an overwhelming initial concentration in early studies on (often idealised) English. In multilingual situations, differing cultural values and expected forms of pragmatic behaviour along with choice of language might then become an area of negotiation or even conflict – either between interlocutors with different linguistic or ethnic background, or even within the same multilingual speaker who has to adapt his or her own affiliations and alignments according to the specific situation. The identification with a particular language/group in a multilingual situation can be asserted overtly by choice of language and more covertly by pragmatic behaviour associated with that language. Accommodation or non-accommodation might then be a strategic face-relevant choice, as Danblon et al. (2005: 45–46) illustrate with regard to the Belgian situation:
118
Susanne Mühleisen In bilingual situations, especially in Brussels, where language is both an instrument and symbol in a shifting power situation […] the question of ‘who speaks what language to whom and when’ […] is a strategic one. Anecdotal accounts report on forms of non-accommodating behaviour whereby speakers demand to be served or addressed in their own language, and/or respond to someone’s failure to do so with irritation, anger or even insults, showing that language choice is a negotiable demand in a potentially conflictive situation.
Strategic code-switching (see also Holmes and Stubbe 2004) is therefore a common phenomenon in multilingual situations. But beyond issues of language choice, there are a number of areas prone to miscommunication or conflict: terms of address, for example, are easily translatable from one language to another, but the effect of a choice of pronominal address, for instance, in French Tu/Vous, German Du/Sie or Italian Tu/Lei, or of first name versus title followed by last name, is not. Rather, the appropriateness of a form of address in a given situation is highly dependent on its indexical force within each language and the conventions given in each cultural environment. Pragmatic transfer, i.e., the influence of pragmatic knowledge of a speaker’s first language on the comprehension and production of linguistic action in a second language (cf. Kasper 1992), might also be a difficulty in highly formulaic patterns, as for example in greetings, introductions, openings and closings of face-to-face or telephone conversations, etc. Unlike in intercultural situations (cf. Rehbein and Fienemann 2004 on introductions in multilingual settings; see also House, this volume) or in native/non-native speaker communication, a familiarity of the patterns relevant to the situation can be assumed in multilingual situations, and transfer as well as interference1 (e.g. Danblon et al. 2005), are not uncommon. Diverging values (e.g. directness/indirectness, self-assertion) in pragmatic behaviour or differences in verbal routines beyond the purely formulaic are less overt sources of miscommunication and conflict. Miscommunication arising from these sources serves as an interesting example of subtle discrepancies in speaker intention and hearer interpretation in communication due to either inability to recognise, or unwillingness to adapt to the interlocutor’s norms. In postcolonial societies, an unequal power dimension between languages and cultures is an added factor to multilingual pragmatic research. The postcolonial nature of the setting is not to be equated purely with a state of “post-independence” of former European colonies. Rather, it implies specific circumstances in which important implications for pragmatic research are given: x
x
x
The multilingual situation is usually a diglossic one in which a number of languages or varieties co-exist in functional distribution. The multilingual situation is shaped by diverse cultural and linguistic influences which comprise European as well as indigenous traditions. Attitudes towards European linguistic and cultural models are often ambiguous and shift between acceptance and rejection of the European language as the
Pragmatics in multilingual language situations
119
H-prestige variety. In between such postcolonial processes of assimilation and abrogation, there are also kinds of creative reinterpretation of the European form. The interpretation of verbal acts therefore requires an awareness of the various linguistic and cultural traditions which may have been alternately maintained, merged or transformed in the course of the colonial and postcolonial period. Moreover, an interpretation of meaning has to take into account the presence of parallel or competing value systems. 2.1.
Pragmatic variation according to language group: Switzerland, Belgium and Luxemburg
In Europe, countries that have maintained a tradition of multilingualism are more the exception than the norm. Switzerland, Belgium and Luxemburg are examples of multilingual European countries which are all characterised by a certain dichotomy between Germanic (German, Dutch/Flemish, Luxemburgish) and Romance (French, Italian) language traditions. While official language policy encourages the idea of multilingualism,2 the reality is more often than not that individuals are not primarily multilingual. In Switzerland, consciousness of differences in verbal behaviour between German-speaking Swiss and French-speaking Swiss seems to be most pronounced. Manno (2005: 100–101) quotes and translates Büchi (2001) who summarises some of the perceived differences in convention with regard to forms of address and interactions with strangers: … the German-speaking Swiss are occasionally surprised at the courteous manners which prevail in French-speaking Switzerland. The French-speaking Swiss are happy to call each other ‘dear friend’; to profess their ‘friendship’ for a bare acquaintance; to drink copiously to each other’s good health even with people they detest and to kiss each other on the cheek three times rather than once. They address each other with ‘Madame’, ‘Monsieur’ and even with ‘Mademoiselle’, which despite protests from feminists, has not been eradicated. The German-speaking Swiss, in contrast, often look awkward when they are to greet someone as their customs require them to pronounce the name of the person to whom they are speaking. Therefore they sometimes prefer to call them ‘Herr Doktor’ or … cross the road.
Despite such stereotypes, Manno (2005: 112) attests the existence of a “specific Swiss ethos, whose common features belong to the average Northern-European norms (preference for negative politeness and for autonomy).” He sees a disparity between the linguistic areas and their corresponding nations, Germany and France. For instance, the readiness to move from formal pronominal forms of address (V) to more informal ones (T) in both Deutsch-Schweiz (DS) and Suisse Romande (SR) differs from that of speakers in Germany and France in that DS speakers seem
120
Susanne Mühleisen
to switch to T less easily than Germans do, but more readily than their SR compatriots who are, again, more “progressive” than French speakers are (Manno 2005: 102–103). Helvetic conversational style has been found to differ from, for instance, German conversational conventions in that it is less “conflictual” and interruptions are scarcely tolerated (cf. also Luginbühl 1999). Norms in verbal routines, such as in telephone conversation openings are distinctly Swiss in all three common languages (German, French and Italian) in that the self-identification of the person answering the phone is expected – quite in contrast to the French (Oui Allô) or Italian (Pronto) forms (Manno 2005: 111). For the Belgian situation, Danblon, et al. (2005: 46) focus strongly on transfer and interference in politeness strategies and claim that while “lexical, grammatical and phonological interference between Dutch and French works both ways”, i.e., from Dutch to French and French to Duch, “the transfer of politeness strategies, however, seems to take place especially from Dutch to French”. One prominent example given is the transfer of eens (Dutch ‘one’) and mal (German ‘times’) into French une fois ‘once’ as a politeness strategy in requests (e.g. Tu [ne] veux pas une fois ouvrir la porte pour le chat?) with several contextual interpretations (Danblon et al. 2005: 47). Their study focuses on politeness markers (formulas) in 300 real-life service encounters in Belgium, 100 from each region and the corresponding languages: Flanders – Dutch/Flemish, Wallonia – French, Brussels – theoretically a “mixed” region, but in reality overwhelmingly French. While borrowing of French politeness markers (voilà, merci) was noted to be frequent in Dutch speakers’ interactions, the reverse borrowing was not observed. Politeness markers in sale requests were generally used more in Flanders (60 %) than in Brussels (“a little less than half”) or Wallonia (42 %; cf. Danblon et al. 2005: 50). By and large, however, Danblon et al. found that differences in rural versus urban communication seem to be more pronounced than between the different regions. Despite the small size of the territory, Luxemburg boasts a rather heterogeneous linguistic state of affairs (Kramer 2005). Most native Luxemburgers use Luxemburgish as their language of daily affairs but are also fluent in French and German. However, one third of its population of 432,000 are recent migrants to Luxemburg. In addition, there are a sizable number of foreign commuters from France, Belgium and Germany who presumably use French and German as their main means of communication. Luxemburgish is therefore marked as an in-group language and a language of nationhood and identity. The use of the two other main languages, French and German, is also not neutral, however, since there are distinct prestige differences between the two languages, with French being regarded as the most respected language and German as a “necessary evil” (Kramer 2005: 61). Again, pragmatic transfer between verbal routine formulas can be observed: Kramer (2005: 63–64) cites the Luxemburgish French s’il vous (te) plaît and Luxemburgish wann iech glieft to be used in the same ways as Germans use bitte, e.g.
Pragmatics in multilingual language situations
121
when giving something to someone (where voilà would be used in French French), or if a hearer has not understood something another person has said (which would elicit pardon? in French French). Despite a vital need for more structured research that goes beyond preliminary explorations of forms of address and formulaic verbal routines in these multilingual territories, two points have become very evident in these studies: Firstly, that language attitudes play a decisive role in multilingual pragmatic usage and secondly, that pragmatics may serve as an area of tension where both national unity and internal and external cultural affiliations are negotiated. 2.2.
Variation according to ethnic group identity: New Zealand
In one of her early studies on politeness and gender in New Zealand (Holmes 1993), Janet Holmes self-consciously concludes the discussion of her findings with the following disclaimer: I have presented the arguments without qualifications as if New Zealand women were a monolithic group. Yet there are many qualifications to be borne in mind. Most of the contributors are well-educated middle class Pakehas. The linguistic behaviour of lower or working class women and of Maori women has been little studied to date. (Holmes 1993: 112)
In subsequent publications Holmes herself and other researchers working on New Zealand English (e.g. Meyerhoff 1994, Holmes 1995, Stubbe 1998, Holmes 2003, Holmes and Stubbe 2004) haved filled this lacuna, especially with regard to pragmatic differences between the English used in New Zealand by the European, mainly British derived population – Pakehas – and the English used by the indigenous population of Polynesian origin, the Maoris. The language situation that can be found here is typical for that of former settler colonies with a European demographic majority like New Zealand or Australia: while indigenous languages are still present (and, in the case of New Zealand, also promoted), a regional form of English plays the overwhelmingly dominant role in most social contexts. Holmes (2005: 91), citing a 1996 New Zealand Census, states that only 26 % of Maori report themselves as able to hold an average conversation in the Maori language (te reo Maori). Proficiency in the Maori language thus seems to be declining despite language political efforts to maintain and revitalise it, and most Maoris would cite New Zealand English as their first and foremost language. While this paper does not focus on pragmatic variation between English and Maori but, rather, on uses of English by the two major ethnic groups in New Zealand, these differences must be seen as influenced by linguistic and cultural factors in the latent multilingual and postcolonial situation. Among the features that have been studied in this regard are firstly the prosodic feature of the High-Rising Terminal Contour (HRT) (Allan 1990, Britain 1992) and secondly, the pragmatic
122
Susanne Mühleisen
tag eh (Meyerhoff 1994, Holmes 1995, 2005) in Pakeha and Maori English. As Holmes explains (1995: 102), “the functions of the HRT are similar to those of pragmatic particles like you know and tags such as eh. Despite its superficial similarity to interrogative intonation, the HRT does not function to mark an utterance as a straightforward question. Rather it serves interpersonal politeness functions,” as a signal of high involvement in the conversation, in that the “HRT invites the addressee to acknowledge the speaker’s talk and indicate that they are following and understanding the speaker” (Holmes 1995: 102–103). Likewise, the pragmatic tag eh can serve as a positive politeness device to stress a similarity in experience or attitude between speaker and hearer. It is therefore interesting to see that both features were found to be employed significantly more often by Maori speakers than their Pakeha counterparts, as well as more by Pakeha women than Pakeha men (see Holmes 1995, 2005). Apart from their function as an ethnic group identification marker for Maori speakers, their use has also been interpreted as a sign of involvement and cooperation in discourse which is characteristic for “members of Pacific cultures” as well as for “women in Western societies” (Britain 1992: 95). Further fields of investigation were the use of verbal feedback (minimal responses/cooperative overlaps vs. overtly supportive responses) in conversation (Stubbe 1998) and differences in narrative strategies (Holmes 2003). In Stubbe’s quantitative and qualitative analysis of data from the Wellington Corpus of Spoken New Zealand English (WCSNZE) she focussed on the amount and type of listener feedback as well as on the “information various formal cues provide about an interlocutor’s intention or attitudes” (Stubbe 1998: 257). Given the emphasis placed by Maori speakers and by women in both larger ethnic groups in New Zealand on involvement and cooperation in discourse, one might suspect that verbal feedback would also be found in large amounts in these groups. However, not only did Maori informants produce approximately a third less verbal feedback than did their Pakeha counterparts (1998: 268) but, secondly, “there is a tendency for the female dyads to produce feedback at a somewhat lower rate than the men, contrary to what might have been expected from previous research” (1998: 269). Types of verbal feedback did not differ significantly. The qualitative analysis shows, however, that the lower feedback rate among Maori speakers might be related to the function of pauses and silence in Maori interaction: Silences occur relatively often in the interaction […]: they seem to be accepted by the speakers, and do not usually disrupt the topical flow. This is consistent with another observation of Metge and Kinloch (1984) that Maori speakers are more tolerant of silence than Pakehas. Metge and Kinloch (1984: 22) suggest that Maoris are more willing than Pakehas to recognise silence as having several causes, meanings and values in interpersonal relations, in comparison with the negative interpretation often placed upon silence by Pakehas. […] For Maori speakers, then, attentive silence may function in certain contexts as the most appropriate strategy for providing collaborative feedback, probably
Pragmatics in multilingual language situations
123
in conjunction with non-verbal signals, and as such indicates a willingness to keep listening to what the speaker has to say in a similar way to verbal feedback in other contexts and for other groups (Stubbe 1998: 275)
In communication across cultures/ethnic groups in New Zealand, such differences in conversational patterns may have contributed to misunderstandings and misinterpretations. Drawing on Metge and Kinloch (1984), Stubbe (1998: 286) reports that “Maori people at times perceive much Pakeha talk as being unnecessary, and therefore ‘cut out’, later responding inappropriately because they did not get the message, with the result that Pakeha find Maori unresponsive and hard to talk to […].” In particular oral genres, misunderstandings can also occur due to differences in narrative structure. For example, as Holmes (2003) shows with regard to story-telling in New Zealand English, the classic narrative structure outlined by Labov and Waletzky (1967) frequently does not apply to Maori narratives, in that Maori storytellers are likely to omit certain narrative components (resolution and coda), thus, from a Pakeha perspective, leaving the story “unfinished” (Holmes 2003: 178). While differences in cultural norms in communication may first of all lead to misunderstandings in everyday interaction, miscommunication due to differences in pragmatic verbal behaviour and interpretation can have the most drastic consequences in institutional contexts. As Eades (2003) demonstrates in a number of cases in Australia, Aboriginal English ways of speaking have been misunderstood or mis-constructed in the legal system. Sources of miscommunication are, on the one hand, a widely reported Aboriginal accommodation strategy called “gratuitous concurrence”3 to answer ‘yes’ to a question and ‘no’ to a negative question, regardless of the Aboriginal speakers’ understanding of the question or the truth or falsity of the proposition made. On the other hand, differences in the interpretation of silence again play a significant role, with considerable implications: The difference in the way in which silence works in Aboriginal English and mainstream Australian English contexts has serious implications for police, lawyer and courtroom interviews of Aboriginal people. Aboriginal silence in these settings can easily be interpreted as evasion, ignorance, confusion, insolence, or even guilt. According to Australian law, silence should not be taken as admission of guilt, but it can be difficult for police officers, legal professionals or jurors to set aside strong cultural intuitions about the meaning of silence, especially when they are not aware of cultural differences in the use and interpretation of silence (Eades 2003: 203).
It is in institutional contexts such as courtroom, school and health sector that differences in power relations between different language groups or ethnic groups in a postcolonial language situation are played out and maintained on the level of communicative norms and interpretations. Institutions are therefore important sites of further research in multilingual and postcolonial language situations.
124 2.3.
Susanne Mühleisen
Variation and postcolonial ambiguity in the Anglophone Caribbean
While the evaluation and norm divide in countries like New Zealand and Australia seems to be very much along ethnic lines, the same does not necessarily follow for all postcolonial societies. In the Caribbean, for example, the presence of (mainly) European and African traditions and their concomitant value systems have led to the emergence of a cultural and linguistic ambiguity (Reisman 1974) where pragmatic variation is more context-dependent and, at times, part of the positioning of the individual speaker (Mühleisen 2011). Most Anglophone Caribbean countries like Jamaica, Trinidad and Tobago or Guyana are characterised by a diglossic language situation with a regional variety of English as the traditional high prestige form and an English-lexicon Creole as the low prestige form.4 In such a situation, speakers usually have access to a range of “lects” – from more basilectal forms of Creole to standard Caribbean English – which they employ according to situation, purpose of the interaction and interlocutor. Code-switching between English and Creole is a frequent phenomenon and can be observed not only in casual conversations but also in public media, for example, in radio phone-in programmes (cf. Shields-Brodber 1992). Therefore, the type of multilingualism found in most parts of the Anglophone Caribbean is not only inherently part of the postcolonial language situation but also part of the competence of most individual speakers. While there has been little research on the pragmatics of English and Creole in the Caribbean (for some exceptions, see Edwards 1979, Shields-Brodber 1992, Shields-Brodber 1998, Mühleisen and Migge 2005), studies in linguistic anthropology from the 1970s on have investigated dichotomies of the European and Africanderived value systems which can be found in linguistic and cultural forms in the Caribbean. Reisman, in his classic article (1974) on “Cultural and linguistic ambiguity in a West Indian village,” talks about a “duality of cultural patterning, both of Creole vs. English speech and of “African” vs. English culture” (1974: 129). Code-choice as well as the concomitant pragmatic verbal behaviour is largely a matter of identification with either culture and the values associated with it. In public contexts in the Caribbean, Abrahams (1983) distinguishes between two types of verbal performers as personas: the broad talker who brings Creole “into stylised use, in the form of wit, repartee, and directed slander” and the “sweet talker, who emphasises eloquence and manners through the use of formal Standard English.” (1983: 57) In the following, I will use two examples from recent research to illustrate how a) this has resulted in the development of particular norms in certain types of verbal interaction and b) how ambiguity and vagueness of meaning are strategically employed in Caribbean contexts. The first example uses material from current work on conversational style in a Trinidadian radio phone-in programme (Mühleisen 2008), the second example looks at particular forms of pronominal address and their pragmatic uses (Mühleisen 2011).
Pragmatics in multilingual language situations
2.3.1
125
Turn-taking and interruptions in radio phone-in programmes
Both the broad talker and the sweet talker find a platform in Trinidadian radio phone-in programmes. While in earlier days, Standard English was the principal code that was heard in a public medium (see also Shields-Brodber 1992: 494), Code-switching between English and Creole is now almost the norm in interactive listener programmes on radio. The choice of code has also important implications on norms in the organisation of conversation. Shields-Brodber (1998: 197) notes, for example, that the classic turn-taking model with minimal overlap in conversation does not hold for the Jamaican situation. In a society in which “assertiveness in conversation is often regarded as a positive indicator of confidence and knowledge about the subject of conversation” (Shields-Brodber 1992: 490), a considerable amount of simultaneous speech in conversation occurs regularly. Interruption and overlap in many exchanges are therefore unmarked and not interpreted as uncooperative: This means that, in many cases, the consequence of creating total silence in the one interrupted does not materialise; most often, both speakers continue making their contributions simultaneously, incorporating a response to any potentially damaging point made by the other party. The speaker who eventually yields the floor is not necessarily the one who was interrupted; but rather, the one who first completes his/her point to his/ her satisfaction.
Due to the inequality of channels in phone-in programmes – studio microphone versus telephone line in – simultaneous speech can, however, lead to a dominance of host or studio guest versus caller. In the following excerpt, taken from a study on the negotiation of power and authority in a particular Trinidadian radio phone-in programme (Mühleisen 2008), the caller is practically outperformed by the studio guest even though she continues in a simultaneous stream of talk which is, however, almost incomprehensible to the audience of listeners: (1) The Morning Show, Power 102 FM, 22. 3. 07 Participants: H = Host, C = Caller, G = Guest 01 02 03 04 05 06 07 08 09 10 11 12 13 14
H: C: H: C:
G: C: G: C: G: C:
Good mornin’ caller= =Hello, good morning Gladiator (.) Morning I agree to to the second to last caller who called a minute ago. All dey want to do, Gladiator, is to depend on even Patrick Manning, depend on di government to hand feed ’em, spoon feed ‘em everythin’, don’t want to get up on di[. . . [Ma’am, ma’am, what you mean depend on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I have to cut yuh, yuh know I have to cut you, nobody didn’t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . come here, I didn’t come here for Patrick Manning and dem to . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
126 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
Susanne Mühleisen G: C: G: C: G: C: G: C: G:
C: G: C: G: C: G:
spoon feed me, I didn’t come here for no handouts, I not on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . no programme, I don’t want no feedback, I don’t want no UIP, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . with dem is very good programme, dey girl, I was working dere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I woulda rally de people to make sure dey get better wages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (.) and I ain’t come here for dat, what I come here for, where people have to depend on di government for is: proper roads, proper drainage, water every day (.) proper in infrastructures, dat is why I come here for (.) not handout (2.0) Gladiator you see, people like dat go trow off, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . what I come here, I come here for di people mi eh coming for . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . handouts let somebody else who call it a positive world call . . . .[and STOP IT. in and [speak.
Transcription conventions: = latch [ overlap (.) micro pause (2.0) pause in seconds … incomprehensible stream of speech in background (phone line in) CAPITALS increased loudness
During almost the whole conversation above (line 09–33), none of the participants is willing to give up the floor and, although the radio listener can only hear the studio guest of the programme (G), simultaneous speech is maintained for a long period of time. In the Trinidadian programme The Morning Show, from which this excerpt was taken, broad talk, including slander, wit and repartee, are part of a complex face negotiation of participants (host, caller, studio guest) in a discourse community whose communicative norms differ from standardised conventions of the genre (cf. Mühleisen 2008). Strategies of inclusion or exclusion, of power and authority, are exercised on a number of levels: particular forms of greetings, address forms or their avoidance in the openings are used to mark membership in the community. For example, the host with the alias name Gladiator (line 02) makes a point of never addressing callers by their name but always simply by “caller” (line 01). It is then up to the callers, many of whom are frequent participants in the discussions, to mark their identity and stake their claim to membership in this phone-in community (cf. Mühleisen 2008). In interruptions, simultaneous speech (see above, lines 09 – 33) and closings, power and authority are negotiated according to politeness rules situated in the Caribbean context of broad talk in public verbal performances and contests. It is one of the characteristics in this Caribbean postcolonial language situation that politeness rules are not homogeneous but, rather, are complex and part of the positioning of the speaker. Thus, in radio pro-
Pragmatics in multilingual language situations
127
grammes where emphasis is placed on the use of Standard English and highly erudite forms of speech are heard, in other words, sweet talk, long stretches of uninterrupted speech by either host or caller can be noted. 2.3.2
Ambiguity and second person plural pronoun use in Trinidad
One of the characteristics of Caribbean Creole languages is that they make use of a second person plural pronoun where Standard English has lost the second person singular vs. plural distinction. In Trinidadian English Creole (TEC), allyuh is an optional form which can be used as an alternative to you – the more standard English form for both second person singular and plural – for various functions: It might potentially be used to emphasise the inclusion of a number of persons addressed, but it can also be employed in single person address. Since it is not an obligatory plural and also not a honorific in the sense of some European language V-forms (Vous, Sie, Lei etc.), its precise purpose is not clear and, until recently, its use had been seen as a more or less random one. A fresh look at pragmatic functions of pronominal address in Trinidad (Mühleisen 2005, Mühleisen 2011) investigates the uses of allyuh in the light of strategic uses of plurality and indirectness in face-threatening situations. Plural address gives the hearer the option of interpreting the address as not applying to him- or herself. It can therefore serve as a means to lower a negative face threat as, for instance, in requests, or also a positive face-threat like in accusations, criticism or insults (cf. Brown and Levinson 1987: 198). In Trinidad, allyuh is frequently employed in such face-threatening situations. Because of the customary employment of the pronoun in conflict situations, its use may also be interpreted as aggressive. The following two examples illustrate an aggressive tone in the use of allyuh to a collective group (‘farmers’), example (2) as well as to a single interlocutor (3). Example 2 is taken from a blog entry in response to a newspaper article on farmers’ claims in Trinidad, example 3 is an excerpt of an autobiographical report which was quoted in the Trinidad Express, one of the two major daily newspapers: (2) Farmers allyuh crazy or what? a little rain fall and allyuh want compensation who benefiting from that? when the price of the produce sky high, allyuh think nobody eh noticing allyuh or what, ripping of the Gov’t with allyuh stupidness, look nah allyuh start to plant properly and stop the sh. for me please and let us get reasonably priced produce […] (www.trinidadexpress.com/news, 09/09/10) (3) Inevitably, his mother would catch him and scold, Move from the jalousie before they chook out all yuh5 eye. Before all yuh say yuh prayers and go to bed, all yuh looking at stupidness. (Bertie Marshall 1972, as cited in www.trinidad express.com, 09/09/08)
128
Susanne Mühleisen
In a questionnaire survey on address forms with 160 participants in Trinidad (see Mühleisen 2011), respondents were asked to provide examples of allyuh uses in dialogue situations. Most of the responses therefore are in the form of a dialogue sequence, some also included metalinguistic comments like the following ones: (4) If I am pissed off with a group of people and I am responding to them in a harsh way, or if I am calling a group of people, e.g. my class, workers, etc. (5) I’m expressing a turn in anger or of all circumstances involving people you don’t want to call their names. In their responses, five different categories of the function of allyuh in TEC could be singled out: a) to emphasise the plurality of addressees and make sure that all persons present feel included; b) to address a singular person as a representative of a particular group: c) to avoid directness and therefore lower a potential face-threatening act; examples here include requests and orders but also advice; d) to signal aggression in positive face-threats like accusations and criticism; e) to signal humour. An example to illustrate category a) would be an utterance like (6) Talking to friends. ‘Allyuh going to church later?’ In this statement from the questionnaire, it is clear that there is more than one addressee and that the motivation for the choice of pronoun might be to make all hearers feel included as addressees of the question. However, the use of allyuh could also be interpreted as a means of indirectness since no one hearer in particular needs to feel addressed directly and therefore obliged to respond to the question or follow the request or invitation implied. The next example (7) is also taken from the questionnaire survey and falls into category b): (7) I had some of allyuh food for lunch (i.e. Indian food) Here, the individual hearer is a collective addressee, i.e., he or she stands for a group of people, in this case, an ethnic group in Trinidad. In requests and orders as negative face threatening acts – category c) – allyuh is employed frequently, as for example in: (8) ‘Allyuh get up please, I want to get the bench’ The use of the second person plural as a less direct form of address can have the effect of lowering the face-threatening act here. The same goes for the speech act of advice (also category c): (9) Talking to friends. ‘Allyuh shouldn’t go there’ Positive face threatening acts like criticism also very often evoke the use of allyuh. An example from the questionnaire survey of this use of the pronoun in category d) would be:
Pragmatics in multilingual language situations
129
(10) ‘Allyuh does throw garbage all over the place’ In category e) “to signal humour”, the statements in the survey were mostly accompanied meta-linguistic comments: (11) I would use ‘all-yuh’ when I am liming or making a joke The quantitative analysis of the statements made in the questionnaire revealed that category a) – emphasis of plurality and inclusion – made up the largest group of responses with 41.0 per cent. This was followed by category d) – to signal anger and aggression in positive face-threatening acts – with 34.3 per cent and then category c) – in requests and advice – with 15.7 per cent. Humour (category e)) and explicit address of an individual as part of a group comprised 6.0 per cent and 3.0 per cent respectively. This means that half of the situations in which allyuh would be typically placed included some kind of face-threatening acts and it demonstrates that its use is to be interpreted not merely as a random marker of plurality. On the other hand, when it was used for emphasising the plurality of addressees, i.e. in 41 % of all cases in the questionnaire, it might also be taken as a positive face strategy that a speaker uses to make everyone feel included. Therefore, allyuh constitutes a useful strategic tool in a postcolonial society like Trinidad and Tobago: its potential ambiguity makes its meaning subject to interpretation along the lines of situation and positioning of the speaker (sweet talking or broad talking). The capacity of allyuh to be taken variably as a sign of friendliness and inclusion, a marker of aggression and anger, or as a symbol of humour, makes its variable employment an ideal example of pragmatic verbal behaviour in a postcolonial context.
3.
Conclusion
This contribution has looked at variation in pragmatic verbal behaviour in a number of language situations that are characterised by multilingualism and the existence of parallel or competing norms of communication. Three different types of multilingual or postcolonial language situations were singled out: a) variation according to linguistic group, b) variation according to ethnic group in a postcolonial situation and c) variation according to the positioning of the speaker in a postcolonial situation. The review of studies in a multilingual European context (Switzerland, Belgium, Luxemburg) has confirmed the importance of language as a means of identifying with values and communicative norms, but has also shown the level of pragmatic transfer and mixture that may arise from the communication between speakers of different language groups. In addition, identification with the larger national community may also result in reducing differences between the linguistic groups, such as for instance, in Switzerland, where a pan-Helvetic conversational style common to both the Deutsch-Schweiz and the Suisse Romande and different
130
Susanne Mühleisen
from that in French (in France) and German (in Germany) was noted in Manno (2005). In a focus on ethnic variation in pragmatic behaviour in New Zealand (e.g. in Holmes 1995, 2005), the multilingual situation was more implicit: while the variety of New Zealand English spoken by the European group of New Zealanders (Pakeha) conforms more with Anglo-norms of pragmatic behaviour (e.g. in the management of silence in conversation), the pragmatics of New Zealand Maori English is possibly influenced by a transfer of communicative norms from Pacific languages. In institutional contexts, power differences also play a large role in such a postcolonial situation where miscommunication due to differences in pragmatic norms can have dramatic consequences (see Eades 2003). Finally, the Caribbean postcolonial language situation provides an example of a transfer and creative adaptation of the many linguistic and pragmatic influences (mainly European and African linguistic influences) in the Creole versus English used in the Anglophone Caribbean. It was shown here how pragmatic rules – for instance, politeness rules – depend largely on the positioning of the individual speaker in the particular situation. A form of multilingualism is the norm for a large number of speakers worldwide, whether it has arisen in a colonial context or not – while the majority of studies in pragmatics still operates with rather homogeneous language frameworks. The quantity of work in pragmatics – like in many other areas of linguistic analysis – seems to closely correlate with the political and economic power of the larger language group under investigation. So far, little work has been done on pragmatic transfer and pragmatic variation in multilingual and postcolonial African and Caribbean communities. Research in institutional contexts in multilingual situations, especially in education, is often restricted to the investigation of interference on the phonological and grammatical level rather than on the level of pragmatic transfer. In the health sector or the legal domain where miscommunication might have serious consequences much more pragmatic research is needed in order to limit the perpetuation of linguistic and cultural hierarchies, especially in communities with a colonial past. A consideration of pragmatic variation according to linguistic groups, ethnic groups and speaker positioning is a large field with much room for further exploration.
Notes 1. Interference is used here in the sense of automatic transfer of lexical, grammatical and phonological patterns from the speaker’s first language to a second language. 2. To various degrees: In Article 4 of the Swiss Bundesverfassung ‘federal constitution’ Switzerland commits itself to the preservation of all four national languages (German, French, Italian and Rumantsch). The federation and the cantons are furthermore re-
Pragmatics in multilingual language situations
131
quested to promote exchange between the different language communities (Art. 70.3). The constitution of Belgium (Art. 4) lists Dutch, French and German as its national languages each of which is matched with a particular region – with the exception of Brussels which is officially bilingual. French and German are official languages in Luxemburg since the 1848 constitution became effective; a language law of 1984 incorporated Letzeburgsch as additional national language. 3. There is an alleged tendency among Australian Aboriginal persons to gratuitously agree with a questioner in court as a means of conveying readiness for cooperative interaction: “Aboriginal English speakers often agree to a question even if they do not understand it. That is, when Aboriginal people say ”yes“ to a question it often does not mean ”I agree with what you are asking me“. Instead, it often means ”I think that if I say ‘yes’ you will see that I am obliging, and socially amenable and you will think well of me, and things will work out well between us“ (Eades 1992: 26). 4. For a detailed discussion of language prestige change in the Anglophone Caribbean, cf. Mühleisen 2002. 5. Since there is no standardized orthographic form for writing Trinidadian English Creole, all yuh is used as a spelling variant of allyuh.
References Abrahams, Roger 1983 The Man-of-Words in the West Indies. Performance and the Emergence of Creole Culture. Baltimore: Johns Hopkins University Press. Allan, Scott 1990 The rise of New Zealand intonation. In: Allan Bell and Janet Holmes (eds.), New Zealand Ways of Speaking, 115–128. Clevedon: Multilingual Matters Anchimbe, Eric and Richard Janney 2011 Postcolonial Pragmatics. Journal of Pragmatics (special issue) 43 (6). Blum-Kulka, Shoshana and Elite Olshtain 1984 Requests and apologies: A cross-cultural study of speech act realization patters (CCSARP). Applied Linguistics 5 (3): 196–213. Blum-Kulka, Shoshana, Juliane House and Gabriele Kasper (eds.) 1989 Cross-cultural Pragmatics: Requests and Apologies. Norwood, NJ: Ablex. Britain, David 1992 Linguistic change in intonation: The use of high rising terminals in New Zealand English. Language Variation and Change 4: 77–104. Brown, Penelope and Stephen Levinson 1987 Politeness. Some Universals in Language Usage. Cambridge: Cambridge University Press. Büchi, Christoph 2001 Mariage de raison. Romands et alémaniques: une histoire suisse. Genève: Zoé. Clyne, Michael 1999 Inter-cultural Communication at Work. Cultural Values in Discourse. Cambridge: Cambridge University Press.
132
Susanne Mühleisen
Danblon, Emanuelle, Bernard de Clerck and Jean-Pierre van Noppen 2005 Politeness in Belgium: Face, distance and sincerity in service-exchange rituals. In: Hickey and Stewart 2005a (eds.), 45–57. Eades, Diana 1992 Aboriginal English and the Law: Communicating with Aboriginal English speaking clients. A handbook for Legal Practitioners. Brisbane: Queensland Law Society. Eades, Diana 2003 The politics of misunderstanding in the legal system. Aboriginal English in Queensland. In: Juliane House, Gabriele Kasper and Steven Ross (eds.), Misunderstanding in Social Life: Discourse Approaches to Problematic Talk 196–223. London: Longman. Edwards, Walter F. 1979 Speech Acts in Guyana: Communicating ritual and personal insults. Journal of Black Studies 10: 20–39. Herbert, Robert K. 1989 The ethnography of compliments and compliment responses: A contrastive sketch. In: Wieslaw Oleksy (ed.), Contrastive Pragmatics, 3–35. Amsterdam/ Philadelphia: John Benjamins. Hickey, Leo and Miranda Stewart (eds.) 2005a Politeness in Europe. Clevedon: Multilingual Matters. Hickey, Leo and Miranda Stewart 2005b Introduction. In: Hickey, Stewart 2005 (eds.), 1–12. Holmes, Janet 1993 New Zealand women are good to talk to: An analysis of politeness strategies in interaction. Journal of Pragmatics 20: 91–116. Holmes, Janet 1995 Women, Men and Politeness. London: Longman. Holmes, Janet 2003 ‘I couldn’t follow her story …’: Ethnic differences in New Zealand narratives. In: House, Kasper, Ross 2003 (eds.), 173–198. Holmes, Janet 2005 Using Maori English in New Zealand. International Journal of the Sociology of Language 172: 91–115. Holmes, Janet and Maria Stubbe 2004 Strategic code-switching in New Zealand workplaces. Scaffolding, solidarity and identity construction. In: House, Rehbein 2004 (eds.), 133–154. House, Juliane, Gabriele Kasper and Steven Ross (eds.) 2003 Misunderstanding in Social Life. Discourse Approaches to Problematic Talk. London: Longman. House, Juliane and Jochen Rehbein (eds.) 2004 Multilingual Communication. (Hamburg Studies on Multilingualism 3). Amsterdam: Benjamins. Kachru, Yamuna 1998 Culture and speech acts: Evidence from Indian and Singaporean English. Studies in the Linguistic Sciences 28(1): 79–98.
Pragmatics in multilingual language situations
133
Kasanga, Luanga A. and Joy-Christine Lwanga-Lumu 2007 Cross-cultural linguistic realizations of politeness: A study of apologies in English and Setswana. Journal of Politeness Research 3: 65–92. Kasper, Gabriele 1992 Pragmatic transfer. Second Language Research 8: 203–231. Kramer, Johannes 2005 Politeness in Luxemburg: Greetings from foreign parts. In: Hickey, Stewart 2005 (eds.), 58–65. Labov, William and Joshua Waletzky 1967 Narrative analysis: oral versions of personal experience. In: June Helm (ed.), Essays on the Verbal and Visual Arts, 12–44. Seattle: University of Washington Press. Lenk, Uta 1998 Discourse markers and global coherence in conversation. Journal of Pragmatics 30(2): 245–257. Luginbühl, Martin 1999 Gewalt im Gespräch: Verbale Gewalt in politischen Fernsehdiskussionen am Beispiel der ‘Arena’. Bern: Peter Lang. Manno, Guiseppe 2005 Politeness in Switzerland: Between respect and acceptance. In: Hickey, Stewart 2005 (eds.), 100–115. Metge, Joan and Patricia Kinloch 1984 Talking Past Each Other: Problems of Cross-Cultural Communication. Wellington: Victoria University Press. Meyerhoff, Miriam 1994 Sounds pretty ethnic, eh?: A pragmatic particle in New Zealand English. Language in Society 23: 367–388. Mühleisen, Susanne 2002 Creole Discourse. Exploring Prestige Formation and Change across Caribbean English-lexicon Creoles. (Creole Language Library 24). Amsterdam: Benjamins. Mühleisen, Susanne 2005 Forms of address in English-lexicon Creoles: the presentation of selves and others in the Caribbean context. In: Mühleisen, Migge 2005 (eds.), 195–223. Mühleisen, Susanne 2008 “Mornin’ Caller” – negotiating power and authority in a Trinidadian phone-in discourse community. Paper presented at the Conference of the Society for Pidgin and Creole Languages / Society for Caribbean Linguistics held in Cayenne, French Guyana, August 2008. Mühleisen, Susanne 2011 Forms of address and ambiguity in Caribbean Creoles: strategic interactions in a postcolonial language situation. Postcolonial Pragmatics. Journal of Pragmatics (special issue). 43 (6): 1460–1471. Mühleisen, Susanne and Bettina Migge (eds.) 2005 Politeness and Face in Caribben Creoles. (Varieties of English around the World G34). Amsterdam: Benjamins.
134
Susanne Mühleisen
Reisman, Karl 1974 Cultural and linguistic ambiguity in a West Indian village, In: Norman E. Whitten and John F. Szwed (eds.), Afro-American Anthropology, 129–144. New York: The Free Press. Reisman, Karl 1989 Contrapuntal conversations in an Antiguan village. In: Richard Bauman and Joel Sherzer (eds.), Explorations in the Ethnography of Speaking, 110–124. (2nd edition). Cambridge: Cambridge University Press. Schneider, Klaus P. and Anne Barron (eds.) 2008a Variational Pragmatics. A Focus on Regional Varieties in Pluricentric Languages. Amsterdam: Benjamins. Schneider, Klaus P. and Anne Barron 2008b Where pragmatics and dialectology meet: Introducing variational pragmatics. In: Schneider, Barron 2008 (eds.), 1–32. Shields-Brodber, Kathryn 1998 Hens can crow too: The female voice of authority on air in Jamaica. In: Pauline Christie, Barbara Lalla, Velma Pollard and Lawrence Carrington (eds.), Studies in Caribbean Language II, 1987–203. St. Augustine: Society for Caribbean Linguistics. Shields-Brodber, Kathryn 1992 Dynamics and assertiveness in the public voice: turn-taking and code-switching in radio talk shows in Jamaica. Pragmatics 2(4): 487–504. Stubbe, Maria 1998 Are you listening? Cultural influences on the use of supportive verbal feedback in conversation. Journal of Pragmatics 29(3): 257–289. Wierzbicka, Anna 1991 Cross-cultural Pragmatics. The Semantics of Human Interaction. Berlin: de Gruyter.
II. The language system and pragmalinguistic features
5.
Speech and writing: linguistic styles enabled by the technology of literacy Douglas Biber
1.
Introduction
Over the past several decades, researchers in linguistics, anthropology, and communication studies have been interested in comparisons of the spoken and written modes. However, there has been considerable disagreement on the extent to which the two modes differ linguistically. Research studies in the 1970s and early 1980s usually argued that there are fundamental linguistic differences between speech and writing. For example, researchers such as O’Donnell (1974), Olson (1977), and Chafe (1982) argued that written language generally differs from speech in being more structurally complex, elaborated, and/or explicit. This view was moderated later, when researchers such as Tannen (1982), Beaman (1984), and Chafe and Danielewicz (1986) argued that communicative task is also an important predictor of linguistic variation; therefore equivalent communicative tasks (e.g. narration) should be compared in speech versus writing to isolate the possibility of mode differences. Multi-Dimensional (MD) studies of register variation in English (e.g. Biber 1986, 1988) went a step further by analyzing linguistic variation within each mode, in addition to analyzing linguistic similarities and differences across speech and writing. These studies found ‘dimensions’ of variation that distinguish between stereotypical ‘oral’ versus ‘literate’ registers, although there are few (if any) absolute linguistic differences between speech and writing with respect to any dimension. Rather, particular spoken and written registers are more or less similar/different with respect to each underlying dimension of variation. More recently, some scholars have taken the exact opposite position from earlier researchers, claiming that there are essentially no linguistic correlates of literacy as a technology. Many of these scholars have been active in ethnographic research, studying the specific functions of literacy practices in particular communities. Having noticed that those functions do not necessarily include the stereotypical purposes of informational exposition, these researchers have made general claims minimizing the importance of literacy as a technology; for example: Literacy can be used (or not used) in so many different ways that the technology it offers, taken on its own, probably has no implications at all (Bloch 1993: 87). It seems quite evident that speech may have all the characteristics Olson ascribes to text and written prose may have none of them … Thus, the characteristics of linguistic per-
138
Douglas Biber
formance at issue here have no intrinsic relation to whether performance is spoken or written (Halverson 1991: 625). In sum, orality and literacy share many common features and the features that have been identified with one or the other have more to do with the context in which language is used than with oral versus literate use. Hornberger (1994: 114)
Thus there is little consensus on the linguistic characteristics of written and spoken discourse, with views ranging from claims that the two are dramatically different to views that there are essentially no linguistic differences between the two. The present paper argues that neither of these extreme views is correct. Rather, it is argued here that the spoken and written modes differ in their potential for linguistic variation: speech is highly constrained in its typical linguistic characteristics, so there is comparatively little linguistic variation among the different kinds of spoken registers (excluding scripted texts that are read or recited). In contrast, the written mode permits a wide range of linguistic expression, including linguistic styles not attested in speech. Thus, written texts can be highly similar to spoken texts, or they can be dramatically different. This difference is attributed to the differing production circumstances of the two modes: real-time production in speech versus the opportunity for careful revision and editing in writing. As a result, the written mode provides the potential for styles of linguistic expression not found in the spoken mode. I present evidence for these linguistic differences between the modes from a survey of corpus-based research studies carried out over the past 20 years. These studies consistently document the same general patterns of linguistic variation: 1) few, if any, absolute differences between speech and writing; 2) large differences in the typical linguistic characteristics of the two modes; and 3) a much larger range of linguistic variation in the written mode than the spoken mode.
2.
Grammatical variation within and across the two modes
One analytical approach that has proven to be especially useful for the study of linguistic variation is corpus-based analysis (see e.g. Biber, Conrad and Reppen 1998; McEnery, Xiao and Tono 2006). Several corpus-based studies have undertaken detailed lexico-grammatical descriptions of spoken and written registers. For example, Biber et al. (1999) is a reference grammar of English that is based on empirical analysis of corpora from four registers: conversation, fiction, newspaper language, and academic prose (c. 20 million words of text overall, with c. 4–5 million words from each of these four registers). A second study that is useful for the purposes here is Biber (2006), which describes the typical linguistic characteristics of university spoken and written registers (both academic and non-academic). I focus in this paper on seven of the registers described in these earlier studies: conversation, classroom teaching, and office hours within speech; and university
Speech and writing: linguistic styles enabled by the technology of literacy
139
textbooks, institutional texts (e.g. university catalogs or handbooks), fiction, and personal e-mail messages within writing. These registers were chosen to illustrate the range of grammatical variation within the spoken and written modes, as well as the strong linguistic differences between the two modes. Those differences are not absolute. Rather, the differences apply in only one direction: writing can be similar to any kind of speech, but no spoken registers are similar to informational writing. Tables 1 and 2 summarize many of the most important grammatical characteristics of these seven registers, based on a survey of the patterns of use documented in the Longman Grammar of Spoken and Written English (LGSWE; Biber et al. 1999), the description of spoken and written university registers in Biber (2006), and the descriptions of e-mail messages, fiction, and academic prose in Biber and Conrad (2009). Table 1 lists linguistic features that are especially common in the spoken registers. These are mostly features relating to pronouns (rather than nouns), the verb phrase (verbs and adverbs), and finite clauses. Surprisingly, there are several kinds of dependent clause included here: finite adverbial clauses (e.g. if I’m lucky, cause he can’t smoke), that complement clauses controlled by verbs (e.g. I don’t think [that] he does), and WH-clauses (e.g. I don’t know what’s happening). Two patterns are especially noteworthy in Table 1: First, there is little variation within speech in the use of these features, despite important differences in situational contexts. Thus, conversation, university advising sessions (office hours), and university classroom teaching are dramatically different in their communicative purposes and in the relation between speaker and addressee. At one extreme, university classroom teaching has been pre-planned, with a primarily informational purpose and comparatively little interaction between speaker and addressee. At the other extreme, conversation usually has personal communicative purposes (e.g. revealing personal stance and narrating previous personal events) and is highly interactive. Office hours has more task-oriented communicative purposes but is highly interactive. The interesting findings for our purposes here, however, are that all three of these spoken registers are highly similar in their grammatical characteristics, characterized especially by a dense use of verbs, adverbs, and pronouns. In addition, as noted above, all three registers show a heavy reliance on finite dependent clauses, especially finite complement clauses and finite adverbial clauses. Thus, regardless of communicative purpose and interactivity, there is a very strong influence from the spoken mode, resulting in a discourse style that is fundamentally clausal. Text samples 1–3 illustrate how similar these registers are in the use of these characteristic grammatical features.
140
Douglas Biber
Table 1.
Distinctive linguistic characteristics of spoken registers compared to written registers [based on Biber et al. 1999, Biber 2006, and Biber and Conrad 2009] SPOKEN UNIVERSITY REGISTERS
OTHER SPEECH
WRITTEN REGISTERS
Linguistic Feature
Classroom Teaching
Office Hours
Conversation
Textbooks
verbs (e.g. get, go, see)
**
**
progressive aspect (e.g. bringing, making)
*
adverbs (e.g. here, now, again)
Institutional Writing
Fiction
Personal E-mails
**
**
**
*
*
*
*
*
*
*
*
*
certainty adverbs (e.g. really, actually)
**
**
**
*
likelihood adverbs (e.g. probably, maybe)
*
**
*
*
pronouns (e.g. I, you, it)
*
**
**
discourse markers (e.g. ok, well)
*
**
**
conditional clauses (e.g. if you read …)
*
**
*
causative clauses (e.g. because …)
*
**
*
temporal clauses (e.g. when/while …)
*
**
*
*
*
**
*
*
*
likelihood verb + that-cl. * (e.g. I think/guess [that] …)
**
**
*
*
communication V + that- cl. (e.g. he said [that] …)
*
*
*
*
**
**
*
*
certainty verb + that-cl. (e.g. I know [that] …)
**
*
WH-clauses * (e.g. you know how to …)
** = extremely common; much more frequent than in other registers * = very common; more frequent than in other registers
Speech and writing: linguistic styles enabled by the technology of literacy
141
Text Sample 1. Conversation Pronouns are bold underlined; verbs are underlined italics; finite adverbial clauses and finite complement clauses are marked by […]. Peter: Gayle: Peter: Gayle: Peter: Gayle:
Peter:
Oh brother. They might not even have left there yet … the hotel. Yeah they were just getting organized. Yeah. Were Bob and Dorothy up already? Oh yeah they were up. I think [we better wait.] You know [we go out to breakfast every Sunday after church]. And they’ll never, they’ll never stay there. I mean [they always, Bob’s always gotta go home for some reason]. He’s got to have his bacon and egg muffin. We took him to breakfast on Sunday, all he did was complain. Of course he gets mad [cause he can’t smoke [cause we always take non-smoking. ] ] Oh well.
Text Sample 2. Classroom teaching; English Pronouns are bold underlined; verbs are underlined italics; finite adverbial clauses and finite complement clauses are marked by […]. Instructor:
[What I want you to do in your free writes] is kind of reflect on [what do you think [he means here] ]. Maybe – and [what you could answer] is would you want to live in that kind of place. Would you want to live there? And [if you do], Why? and do not, Why? And how does Rymmer give you clues? I think [Rymmer, especially in a poem like this, he talks about this hollowness at his core, sort of the absence of the bona fide, legitimate purpose to the whole thing]. I think [clues like this are embedded throughout that suggest [that Rymmer’s pretty negative, or skeptical about this whole project] ], right? And [what I wanna know] is, [if you do want to live there], why is that, and [if you don’t], what is it about Rymmer’s writing, or Rymmer’s ideas that lead you to believe [that you wouldn’t want to live there].
Text Sample 3. Office Hour; Natural science Pronouns are bold underlined; verbs are underlined italics; finite adverbial clauses and finite complement clauses are marked by […]. Student:
I mean [ [if you were going to be running the test for four days] you’d look at it for fourdays] and I think [that ’s [what you said in class] ] is [that you should do it at least [as long as you ’re going to run the test] ]
142
Douglas Biber
Instructor:
Yeah … yeah I guess [ I just hadn’t seen [when I read it] [that um you defined those two things very well] ], but um but um I guess [ you could have meant that].
The second pattern to note from Table 1 is that these same grammatical characteristics can be employed in writing. Written registers like e-mail messages are especially similar to conversation. For example, consider the dense use of verbs and pronouns in the following e-mail message: Text Sample 4. E-mail message Pronouns are bold underlined; verbs are underlined italics. Hey there, How’s it going? It won’t be long now before you’re down here (or at least close to where I am now). I need your arrival time, flight number, etc. We are getting to CC earlier than I planned, so I will pick up the car on the 8th and may do something with these folks before you arrive. Let me know if you have any other questions too before you head off. Cheers, LC Fiction is a complex written register, composed of prose sections interspersed with dialogue. It is not surprising that fictional dialogue is similar to natural conversation in its grammatical characteristics. In many ways, though, fictional prose is also similar to spoken discourse, especially in its reliance on verbs, pronouns, and finite clauses. Text Sample 5. Fictional prose Pronouns are bold underlined; verbs are underlined italics. Penelope was glad that Claire had left Frizzle, for he made being alone not exactly alone. The dog loved her and offered the nearest thing to human companionship. At least he was waiting when she returned home from work in the evenings. Aside from Frizzle to help combat the chilling silence of the apartment, she brought home papers to grade and plunged into completing her book on John Milton. In summary, Table 1 shows that: 1) speech is surprisingly homogeneous in its grammatical characteristics, relying on verbs, pronouns, and finite clauses regardless of the communicative purpose for specific spoken registers; and 2) it is possible in writing to employ these same grammatical characteristics. Table 2 lists features that are especially common in informational written registers.
Speech and writing: linguistic styles enabled by the technology of literacy Table 2.
143
Distinctive linguistic characteristics of (informational) written registers compared to spoken registers [based on Biber et al. 1999, Biber 2006, and Biber and Conrad 2009]
Linguistic Feature
SPOKEN UNIVERSITY REGISTERS
OTHER WRITTEN REGISTERS SPEECH
Classroom Teaching
Conversation
Office Hours
Textbooks
Institutional Writing
Fiction
Personal E-mails
**
*
*
*
*
diversified vocabulary (e.g. sanctimonious)
*
nouns/nominalizations (e.g. term, assumption)
**
rare nouns (e.g. abscission, ambivalence)
*
common abstract / process nouns (e.g. system, factor, problem)
*
style adverbs (e.g. generally, typically)
*
adjectives (e.g. important, likely)
*
*
linking adverbials (e.g. however, for example)
**
*
passive voice (e.g. was determined)
*
*
relative clauses (e.g. the sequence which determines …)
*
*
*
prepositional phrases (e.g. patterning of behavior by households)
**
**
*
noun + that-clause (e.g. the fact/assumption that …)
*
mental verb + to-clause (e.g. remember to …)
*
adjective + to-clause (e.g. unlikely/difficult to …)
*
noun + to-clause (e.g. the opportunity to learn) ** = extremely common; much more frequent than in other registers * = very common; more frequent than in other registers
*
*
*
*
144
Douglas Biber
Many of these features are nouns or features that can be used for noun phrase modification, such as adjectives, prepositional phrases, or relative clauses. Similar to Table 1, Table 2 shows that there is extensive linguistic variation among written registers. Thus, these features are very common in informational written registers, but many of them are not common in e-mail messages or fiction. Text sample 6 illustrates the dense use of nouns, attributive adjectives, and post-nominal modifiers typical of informational writing, Text Sample 6. Medical textbook There was no significant difference between the two groups regarding blood pressure, family history of ischaemic heart disease, obesity or alcohol consumption. There was, however, a high incidence of heavy alcohol consumption amongst patients who subsequently required coronary artery surgery. A similar discourse style is typical of institutional writing. For example, consider the dense use of nouns, attributive adjectives, and prepositional phrases in Text Sample 7, taken from a university web site: Text Sample 7. Department web page (Anthropology) Cultural and Social Anthropology deal with the many aspects of the social lives of people around the world, including our own society: their economic systems, legal practices, kinship, religions, medical practices, folklore, arts and political systems, as well as the interrelationship of these systems in environmental adaptation and social change. Physical Anthropology describes and compares world human biology. Its focus is on humans and the primate order to which they belong as part of nature, and it seeks to document and understand the interplay of culture and biology in the course of human evolution and adaptation. Anthropological Linguistics deals with varied aspects of human language, and the characteristics of non-human communication systems, in order to achieve an understanding of past and present human language systems and their significance in social life. The most distinctive characteristic of these informational written registers is the extent to which they rely on grammatical phrases. When verbs are used, they are often non-finite forms, but the even more distinctive characteristic is the absence of verbs altogether. Thus, consider the first sentence from text Sample 7; there is only one finite verb (deal with) and one non-finite verb (including) in this long, complex sentence: Cultural and Social Anthropology deal with the many aspects of the social lives of people around the world, including our own society: their economic systems, legal practices, kinship, religions, medical
Speech and writing: linguistic styles enabled by the technology of literacy
145
practices, folklore, arts and political systems, as well as the interrelationship of these systems in environmental adaptation and social change. Instead of verbs and clauses, information is conveyed primarily through complex noun phrases, with extensive phrasal modifiers (attributive adjectives, pre-modifying nouns, and prepositional phrases). Classroom teaching is especially noteworthy in this regard, in that it does not make dense use of nouns and complex noun phrase structures, even though it has similar informational communicative purposes to textbooks. Thus, Text Sample 2 is dramatically different from Text Samples 6 and 7. Text Sample 2 is typical of most classroom teaching in that it uses comparatively few nouns overall, and few complex noun phrases (e.g. with embedded prepositional phrases or relative clauses). Rather, the linguistic complexity of this passage is expressed primarily through the dense use of finite dependent clauses – adverbial clauses with if or because, and complement clauses (WH or that). Thus, communicative purpose is secondary to mode in its influence on linguistic form. That is, within speech, communicative purpose has a surprisingly small effect on the use of grammatical characteristics. Whether a speech event is interactive and interpersonal (as in normal conversation), or primarily monologic and informational (as in classroom teaching), it is characterized by the same set of typical linguistic features: verbs, pronouns, finite adverbial and complement clauses, etc. And all of these speech events are characterized by the relative absence of nouns and complex noun phrase structures. In contrast, within the written mode, communicative purpose and audience have a major influence. Thus, popular written registers with non-informational purposes use clausal linguistic styles, similar to spoken registers. However, informational or specialist written registers have developed a distinctive grammatical style, relying predominantly on nouns and phrases rather than verbs and clauses. In summary, Tables 1 and 2 show three general distributional patterns: 1) linguistic features that are common in informational writing tend to be rare in the spoken registers, and vice versa; 2) spoken registers are surprisingly similar to one another in their typical linguistic characteristics, regardless of differences in communicative purpose, interactiveness, and pre-planning; and 3) in contrast, written registers have a much wider range of linguistic diversity. The linguistic uniformity among spoken registers can be attributed to their shared production circumstances. Spoken texts are normally produced in real time. As a result, spoken registers share a heavy reliance on finite clausal syntax. Conversely, it seems that the dense use of complex noun phrase structures – typical of some kinds of written prose – is simply not normally feasible given the production constraints of the spoken mode. At the same time, we find large linguistic differ-
146
Douglas Biber
ences among written registers, corresponding to differences in purpose, interactiveness, author involvement, etc. This variation is attributed to the production circumstances of writing, which give the author maximum flexibility to choose styles of linguistic expression very similar to those typical of speech, or to produce expressions that are apparently not feasible in speech. Multi-dimensional studies of spoken and written registers in English (e.g. Biber 1988, 1995; Conrad and Biber 2001) show similar patterns. Several general patterns and conclusions about spoken and written language have emerged from multi-dimensional studies: 1) Some dimensions are strongly associated with spoken and written differences; other dimensions have little or no relation to speech and writing; 2) There are few, if any, absolute linguistic differences between spoken and written registers; 3) However, there are strong and systematic linguistic differences between stereotypical speech and stereotypical writing, that is, between conversation and written informational prose; 4) The spoken and written modes differ in their linguistic potential; they are not equally adept at accommodating a wide range of linguistic variation. In particular, there is an extremely wide range of linguistic variation among written registers, because writers can choose to employ linguistic features associated with stereotypical speech. In contrast, there is a more restricted range of linguistic variation among spoken registers. As in the discussion above, this last pattern can be attributed to the real-time production circumstances of the spoken mode, making it difficult to employ many of the linguistic features associated with stereotypical informational writing.
3.
Historical evidence
Historical corpus-based studies help us to better understand the nature of these differences between the spoken and written modes. For example, Biber and Clark (2002) investigate historical change in the noun phrase structures commonly used in drama, fiction, and medical prose (from the ARCHER Corpus). The study shows how non-clausal ‘compressed’ types of noun modification – attributive adjectives, nouns as pre-modifiers, and prepositional phrases as post-modifiers – have increased dramatically in use over the past three centuries, while clausal noun modifiers, like relative clauses, have remained relatively constant in frequency across these periods. For example:
Speech and writing: linguistic styles enabled by the technology of literacy
147
Attributive adjectives: gradually expanding cumulative effect Nouns as pre-modifiers: baggage inspection procedures Prepositional phrases as post-modifiers: a high incidence of heavy alcohol consumption amongst patients … However, as Figures 1–3 show, the increase in non-clausal modifiers has occurred only in informational written prose (medical prose in this study; see also Biber 2003). In contrast, drama and fiction have shown only slight increases, if at all. These historical developments can be attributed to two influences: 1) an increasing need for written prose with dense informational content, associated with the ‘informational explosion’ of recent centuries, and 2) an increasing awareness among writers of the production possibilities of the written mode, permitting extreme manipulation of the text. Specifically, it seems that the extremely dense use of complex noun phrase constructions (described in section 2 above) is not normally feasible in speech, regardless of the communicative purpose. As a result, we did not have models for this style of linguistic expression in earlier centuries. As research specialists developed a communicative need for such styles, authors became increasingly aware that the written mode provided extended production possibilities, allowing dense phrasal embedding in ways that are not attested in the
Figure 1. Attributive adjectives across periods [adapted from Biber and Clark (2002)]
148
Douglas Biber
Figure 2. Noun-noun sequences across periods [adapted from Biber and Clark (2002)]
Figure 3. Prepositional phrases (excluding OF)as postnominal modifiers, across periods [adapted from Biber and Clark (2002)]
Speech and writing: linguistic styles enabled by the technology of literacy
149
spoken mode. Here again, it is important to note that the written mode does not necessitate these distinctive linguistic styles; thus written fiction has not changed historically to employ these complex noun phrase structures. Rather, it seems that the written mode provides possibilities for styles of linguistic expression not normally possible in speech, and that authors have only gradually come to exploit those possibilities over the past four centuries.
4.
Conclusion
Some individual speakers are especially gifted and can produce notably elaborate and fluent styles of discourse. Oral poets in traditional societies are examples of such gifted speakers. For example, Andrzejewski and Lewis (1964) describe the elaborated linguistic styles typical of oral poetry in Somali. A Somali poet can spend days working on a single poem, to produce a text governed by strict rules of alliteration, with a strong preference to avoid repetitions of words. The existence of such speakers shows that the production difference between the spoken and written modes is not an absolute one. That is, a few speakers are able to mentally compose dense, lexically elaborated texts, relying on memory without the aid of writing. Such texts go through multiple rounds of planning, revision, and editing, similar to the process of careful production described above for written registers. In this case, the process of careful production and revision relies heavily on an exceptional memory – the entire text is planned, revised, and edited over a period of weeks, relying on the powers of memory. The case of Somali oral poets show that such feats are humanly possible. However, these are truly exceptional spoken registers. The vast majority of speech, in any language, is not memorized and has not been mentally revised and edited. Rather, speech is normally produced spontaneously in real-time (even if it has been pre-planned, as in the case of university lectures). Corpus-based studies of spontaneous spoken registers have shown consistently that they differ from written registers in that they do not provide the possibility of extreme lexical diversity, or the dense use of complex noun phrase constructions. Rather, such linguistic styles require extensive planning, revision, and editing – processes that are normally possible only in writing. This does not represent an absolute or necessary difference between speech and writing. Rather, authors can exploit the written mode to produce texts that are very similar to the typical linguistic styles of speech. However, the converse is not true: that is, speakers are not normally able to revise and edit their texts because they are constrained by real-time production circumstances. As a result, some written registers have evolved to employ linguistic styles – with extreme lexical diversity and a dense use of complex noun phrase structures – that are not normally feasible in the spoken mode.
150
Douglas Biber
In sum, there are genuine linguistic consequences of literacy, but these consequences have to do with the linguistic potential of the two modes rather than the necessary linguistic characteristics of the two modes. In particular, the present chapter has shown that language production in the written mode enables styles of linguistic expression not normally attested in speech, even though writers often choose not to exploit that linguistic potential. References Andrzejewski, B.W. and I.M. Lewis 1964 Somali poetry: An introduction. Oxford: Oxford University Press. Beaman, Karen 1984 Coordination and subordination revisited: Syntactic complexity in spoken and written narrative discourse. In: Deborah Tannen (ed.), Spoken and written language: Exploring orality and literacy, 45–80. Norwood, NJ: Ablex Publishing. Biber, Douglas 1986 Spoken and written textual dimensions in English: Resolving the contradictory findings. Language 62: 384–414. Biber, Douglas 1988 Variation across speech and writing. Cambridge: Cambridge University Press. Biber, Douglas 1995 Dimensions of register variation: A cross-linguistic comparison. Cambridge: Cambridge University Press. Biber, Douglas 2003 Compressed noun phrase structures in newspaper discourse: The competing demands of popularization vs. economy. In: Jean Aitchison and Diana M. Lewis (eds.), New media discourse, 169–181. Routledge. Biber, Douglas 2006 University language: A corpus-based study of spoken and written registers. Amsterdam: John Benjamins. Biber, Douglas and Victoria Clark 2002 Historical shifts in modification patterns with complex noun phrase structures: How long can you go without a verb? In: Teresa Fanego, Mará José LópezCouso and Javier Pérez-Guerra (eds.), English historical syntax and morphology, 43–66. Amsterdam: John Benjamins Publishing. Biber, Douglas and Susan Conrad 2009 Register, genre, and style. Cambridge: Cambridge University Press. Biber, Douglas, Susan Conrad and Randi Reppen 1998 Corpus linguistics: Exploring language structure and use. Cambridge: Cambridge University Press. Biber, Douglas and Edward Finegan 1989 Drift and the evolution of English style: A history of three genres. Language 65: 487–517. Biber, Douglas and Edward Finegan 1997 Diachronic relations among speech-based and written registers in English. In: Terttu Nevalainen and Leena Kahlas-Tarkka (eds.), To explain the present:
Speech and writing: linguistic styles enabled by the technology of literacy
151
Studies in changing English in honor of Matti Rissanen, 253–76. Helsinki: Societe Neophilologique. [Reprinted in Conrad and Biber 2001. 66–83]. Biber, Douglas, Stig Johansson, Geoffrey Leech, Susan Conrad and Edward Finegan 1999 Longman grammar of spoken and written English. London: Longman. Bloch, Maurice 1993 The uses of schooling and literacy in a Zafimaniry village. In: Brian V. Street (ed.), Cross-cultural approaches to literacy, 87–109. Cambridge: Cambridge University Press. Chafe, Wallace L. 1982 Integration and involvement in speaking, writing, and oral literature. In: Deborah Tannen (ed.), Spoken and written language: Exploring orality and literacy, 35–54. Norwood, NJ: Ablex Publishing Co. Chafe, Wallace L. Jane and Danielewicz 1986 Properties of spoken and written language. In: Rosalind Horowitz and S. Jay Samuels (eds.), Comprehending oral and written language, 82–113. New York: Academic Press. Conrad, Susan and Biber Douglas (eds.) 2001 Variation in English: Multi-Dimensional studies. London: Longman. Halverson, John 1991 Olson on literacy. Language in Society 20: 619–640. Hornberger, Nancy H. 1994 Continua of biliteracy. In: Bernardo M. Ferdman, Rose-Marie Weber and Arnulfo G. Ramirez (eds.), Literacy across languages and cultures, 103–139. Albany: State University of New York Press. McEnery, T., R. Xiao and Y. Tono 2006 Corpus-based language studies. London: Routledge. O’Donnell, R. C. 1974 Syntactic differences between speech and writing. American Speech 49: 102–110. Olson, David 1977 From utterance to text: The bias of language in speech and writing. Harvard Educational Review 47(3): 257–81. Tannen, Deborah 1982 The oral/literate continuum in discourse. In: Deborah Tannen (ed.), Spoken and written language: Exploring orality and literacy, 1–16. Norwood, NJ: Ablex Publishing.
6.
Phonetics and the management of talk-in-interaction Gareth Walker
1.
Introduction
Since the late 1970s a body of work has built up in which analysts explore relationships between the local interactional context and social contingencies and the precise phonetic design features of the talk. Work in interactional phonetics proceeds on the basis that how something is said may have an impact on what that talk is doing, and what it is taken by a co-participant to have done. Consider someone at a meal table in an ordinary dining room bellowing “could you pass me the salt?” In the absence of extenuating circumstances (e.g. one of the participants has considerable hearing difficulty), this turn at talk is likely to be treated rather differently from that same utterance produced at a normal volume. It may, for instance, receive a response such as “don’t shout at me”. Such a response would be an accountable action had the preceding turn not been bellowed. In orienting to speech as having been bellowed, a recipient is orienting to the changes in the actions of the lungs and other vocal organs as significant. Fragment 1 shows an orientation by one participant to the phonetic design of the other participant’s talk in much the same way. Joel, a male in his late teens, is sitting in a car talking with his mother.1 (1) Dersley and Wootton (2001:616)
154
Gareth Walker
Joel’s pleas in lines 135–6, “DO:N’ shout mum. don’ shout,”, show an orientation to the design of Mum’s preceding talk. And there seems to be good grounds for Joel’s pleas. Mum’s talk is loud throughout this fragment. However, in line 131 Mum’s talk increases in loudness and increases in overall pitch. While perhaps not “shouting” in the usual sense of the word, Mum’s talk becomes demonstrably emphatic in line 131, and this emphasis is sustained through the remainder of the fragment. Support for these observations can be provided by inspection of the speech signal via computer software.2 Relevant fundamental frequency (F0) and intensity measures can be found in Table 1. F0 is the acoustic correlate of pitch: the higher the frequency, the higher the perceived pitch. Intensity, often measured in decibels (dB), is the acoustic correlate of loudness. Roughly speaking, a doubling of loudness corresponds to an increase of 10 dB (Laver 1994: 502). Providing defensible quantified measures of loudness is more difficult than pitch due to a range of confounding factors, such as segmental quality, background noise and overlapping talk, and distance from microphone. However, the measures given here do reflect auditory percepts. Table 1.
Frequency and intensity measures for parts of Fragment 1. Measures given in italics are those where only part of the relevant talk could be measured. Measures are given in Hertz (Hz), semitones (ST) and deciBels (dB).
There is a clear overall raising of pitch in Mum’s turn in line 131, the subsequent utterance in line 134 and beyond. With respect to loudness syllables in lines 131 and 134 have a higher mean peak intensity than in lines 126–7. These measures are indicative of audible differences. Joel is treating aspects of the delivery of Mum’s talk as important: in producing “DO:N’ shout mum. don’ shout,” (lines 135–6), Joel is showing a clear orientation to the changes in Mum’s voice. What can be seen in Fragment 1, in an especially plain fashion, is an orientation by one participant to the phonetic design of a co-participant’s talk. In other words, the phonetic design of one participant’s talk has occasioned a particular interactional
Phonetics and the management of talk-in-interaction
155
outcome. Work in interactional phonetics has identified various ways in which manipulations of the vocal organs – sometimes very subtle manipulations – are implicated in the management of talk-in-interaction, and principally everyday conversation. It is work in this methodological paradigm which will be the focus of this chapter. The work described in the sections which follow is arranged around three main tasks which the phonetic design of talk has been shown to handle. These are: 1. the management of turn-projection and overlap (section 3); 2. the marking of relationships between turns (section 4); 3. the marking of relationships within turns (section 5). Some concluding remarks are given in the final section, along with suggestions for future work in interactional phonetics and some discussion of challenges which will need to be met. Some methodological principles are discussed in the next section.
2.
Methodological principles
Interactional phonetics studies socially conditioned phonetic variation. It can therefore be seen as aligned with sociophonetics, though the focus there has mainly been on variability attributable to gender, age and class (Foulkes and Docherty 2006). Sociophonetics generally overlooks the social-interactional import of phonetic variability. Certain methodological principles – which in most cases are adapted from principles of conversation analysis (CA; see Drew 2004; Goodwin and Heritage 1990; Schegloff 2007) – make interactional phonetics a distinct analytic approach (see also Local and Walker 2005). First, analysts within interactional phonetics strive to demonstrate participants’ orientations to the analytic claims made. It is necessary to show that some phonetic feature or feature-cluster is not simply present in the interaction, but is treated as significant (oriented to) by the participants themselves. Second, only data drawn from talk-in-interaction are considered. Talk-in-interaction refers to talk produced such that there is some degree of interaction between the participants and includes, for instance, business meetings, unscripted lectures and interviews, as well as everyday conversation. Study is rightly not limited to everyday conversation, though everyday conversation is the principal form of talk-in-interaction. Using data from talk-in-interaction not only ensures the ecological validity of the data, but also provides analytic resources unavailable in non-interactive and scripted settings, including the demonstration of participant’s orientations. Third, although there is no requirement that the analysis should be qualitative, any analytic account must handle single cases as cogently as it does the aggregate. Constructing robust statistical analyses of sequential-interactional events is notoriously problematic (see Schegloff 1993); furthermore, no quanti-
156
Gareth Walker
tative measure of frequency alters the fact that an episode of interaction occurred in that way on that occasion and was dealt with by those participants (Wootton 1989).
3.
Turn projection and overlap
One key task for participants engaged in spoken interaction is managing the entry to and exit from talk. Fragment 2 is taken from a naturally occurring telephone conversation between two middle-aged sisters from North America, Emma and Lottie. Talk has been about Thanksgiving turkeys. (2) NB.IV.13, 10:16
In line 2 Emma is treating Lottie’s talk at line 1 as transition relevant. In other words, Lottie’s talk in line 1 ends at a point where speakership may legitimately change (Sacks, Schegloff and Jefferson 1974). A change in speakership may legitimately occur at this point as Lottie’s talk in line 1 constitutes a possibly complete turn-constructional unit, or TCU (Schegloff 1996). There are three factors which contribute to the status of the talk in line 1 as a possibly complete TCU. First, Lottie’s talk is grammatically complete: it is a complete sentence. Second, it implements a recognizably complete action: it is an enquiry. Third, the phonetic design of the talk is such that it is hearable as possibly complete. Figure 1 provides a visual representation of aspects of this turn at talk. Time runs along the x-axis. The figure shows a waveform (lower part), and in the upper part a F0 trace (dotted line, left y-axis) and an intensity trace (solid line, right y-axis). To give an indication of placement in the speaker’s range, the F0 trace is presented scaled to the speaker’s base- and topline pitches, which were established on the basis of a representative speech sample. The intensity trace is not scaled as this cannot be done in any meaningful fashion for these data. Word labels run along the top of the figure, with boundaries marked by vertical dotted lines.3 Among those features which contribute to the talk in line 1 being hearable as possibly complete are the following: 1. an overall declination in loudness across the speaking turn; the nuclear accent in this intonation phrase is on “gi:t” which has a peak intensity of 81.8 dB while the next accented syllable (the first syllable of “turkey.”) has a peak intensity of 75 dB.
Phonetics and the management of talk-in-interaction
157
2. an overall declination in pitch across the speaking turn; the nuclear accent (“gi:t”) has a peak F0 of 220 Hz, while the first syllable of “turkey.” has a peak F0 of 207 Hz. 3. a fall in pitch of 7.6 ST to low in the speaker’s pitch range over the last foot (“turkey.”; a foot consists of one stressed syllable and all following unstressed syllables up to, but not including, the next stressed syllable). 4. a slowing down towards the end of the TCU. Inspecting the waveform and boundaries in Figure 1 shows that the final foot has a longer duration than the preceding feet. The final foot has a duration of 503 ms, the penultimate foot 340 ms, and the antepenultimate 308 ms. 5. an absence of any final glottal or supra-glottal (oral) closures which can adumbrate more talk (Local and Kelly 1986; Local and Walker 2005). It is important to register that in this context “hearable as possibly complete” means “hearable as possibly complete by a co-participant”. We can see that Emma hears it – orients to it – as possibly complete by her electing to start up talk as Lottie’s TCU reaches its conclusion. Furthermore, there is nothing in Lottie’s subsequent conduct which suggests that Emma’s incoming was anything other than legitimate: in other words, Emma’s incoming in no way runs against Lottie’s view of the design of her own talk in line 1. There are, then, a range of phonetic features of the talk in line 1 of Fragment 2 which mark out the talk as possibly complete. In their account of turn-taking in
Figure 1. A visual representation of line 1 of Fragment 2.
158
Gareth Walker
conversation, Sacks, Schegloff and Jefferson (1974) claim a connection between the phonetic design of talk and the management of turn-taking. While this claim was not given any significant, direct treatment by Sacks, Schegloff and Jefferson (1974), since the publishing of their article a considerable amount of work in interactional phonetics has been dedicated to studying the role of phonetic detail in the management of turn-taking: see e.g. Auer, Couper-Kuhlen, and Müller (1999); Couper-Kuhlen (1993); Ford and Thompson (1996); Fox (2001); Local, Kelly, and Wells (1986); Local, Wells, and Sebba (1985); Obeng (1991); Ogden (2001); Selting (2000) and Wells and Peppé (1996). Although there is variation in the precise analytic focus of these studies, they routinely find that phonetic characteristics contribute to the status of an utterance as transition relevant, or not. Features described include prosodic (pitch, loudness, rhythm) and ‘non-prosodic’ (articulatory, phonatory) features. The most detailed parametric-phonetic study of turn-taking to date is provided by Local, Kelly and Wells (1986). They argue that in a particular variety of English (Tyneside English) two clusters of phonetic features routinely mark transition relevance, including (i) a slowing down in tempo, (ii) a loudness ‘swell’ on the last accented syllable, (iii) centralized vowel qualities in the last foot, (iv) marked aspirated release of plosives in final position, and (v) either a pitch step-up or dropping pitch at the end of the turn. Studies of turn-taking in Englishes spoken in the UK and the US are the most numerous. One thing which has become clear from this work on turn delimitation is that the phonetic details which mark transition relevance differ across UK varieties: in addition to Local, Kelly, and Wells (1986) on Tyneside English, see e.g. Local, Wells, and Sebba (1985) on London Jamaican, Wells and Peppé (1996) on Ulster English and Wells and Macfarlane (1998) on a variety spoken in the West Midlands. Work has also shown that phonetic detail has a role to play in the management of turn-taking across languages. On the basis of recordings of naturally occurring conversations, Obeng (1991) argues that speakers of Akan (a language of Ghana) can mark out talk as transition relevant by producing it with diminuendo (decreasing), piano (soft) or pianissimo (very soft) loudness accompanied by low or falling pitch at its end. Crucially, Obeng shows that co-participants orient to these features as marking talk as transition relevant, principally by starting up their talk soon after the occurrence of talk exhibiting these prosodic features. Working on turn-taking in another non-Indo-European language, Ogden (2001, 2004) shows how creak phonation and glottal stops figure in the management of turn-taking in Finnish (for accounts of different phonation types, see Catford 1977: 93–116, Laver 1994: 184–201). Working with a corpus of Finnish radio phone-in programs, Ogden argues that creak is associated with turn-yielding, and glottal stops with turn-holding. Fragment 3 is taken from Ogden (2004: 38). The incidence of creak phonation, along with other prosodic characteristics, is marked with C above the line of transcription; braces and dashes indicate the temporal extent of the relevant feature.
Phonetics and the management of talk-in-interaction
159
(3) Ogden (2004:38)
The talk produced by the presenter of the radio phone-in (P in the transcription) in lines 21–22 is a possibly complete TCU. This talk is syntactically possibly complete, and it implements a recognizable action: it is a ‘yes-no’ interrogative. Furthermore – and this point is central to Ogden’s argument – line 22 ends with creak phonation which also contributes to the status of this talk as transition relevant. Note also that the end of this turn is delivered with piano loudness characteristics, symbolised by p: cf. the observations made by Obeng (1991) described above and the discussion of the English example in Fragment 2. As was shown for Fragment 2, the claim that this talk is transition relevant gains support from what the co-participant does next. In Fragment 3 the caller (C) starts up talk without delay (symbolised by the equals signs, =, joining these two turns at talk), and in doing so displays an orientation to the talk in lines 21–22 as transition relevant. While creak phonation can mark talk out as transition relevant in Finnish, a glottal stop at the end of an otherwise possibly complete TCU marks that there is more to come from that speaker (cf. observations on the turn-holding function of glottal stops in English made by Local and Kelly 1986; Local and Walker 2005). Fragment 4 is taken from Ogden (2001: 147). The glottal stop of interest occurs in line 53, and is transcribed using symbols drawn from the International Phonetic Alphabet: ʔ. (4) Ogden (2001:147)
160
Gareth Walker
Ogden (2001) argues that the presence of the unreleased glottal stop, along with the absence of creak phonation and other intonational cues to transition relevance, marks out this stretch of talk as incomplete. That this talk is not transition relevant, even though it is complete in other ways (including it being syntactically possibly complete) is supported by P continuing talk following the glottal stop, and by the absence of any incoming talk from a co-participant which might suggest a treatment of the talk as transition relevant. By working on a non-Indo-European language, Ogden’s work on phonetics and the management of turn-taking in Finnish has widened the scope of interactional phonetics. Furthermore, while it is often features of intonation and rhythm which are inspected for their role in the marking of transition relevance, Ogden’s work shows how phonatory and articulatory details can be significant too. This is an important methodological point: that it is not only prosodic features which merit study. By taking a more rounded view of the speech signal it becomes possible to uncover regularities which would otherwise pass the analyst by (see also Local 2003; Local and Kelly 1986; Local and Walker 2005). In the preceding discussion the point has been made that participants can show their own orientation to features which mark transition relevance. So far the main form of evidence for such an orientation has come from the occurrence of smooth turn-transition occurring ‘in the clear’ (i.e. without simultaneous talk) after the occurrence of the relevant features. Evidence for an orientation to phonetic markers of transition relevance can also be sought in materials where talk occurs in overlap i.e. where more than one participant talks at once. Overlap has received considerable attention in CA, e.g. Jefferson (1973, 1986); Schegloff (2000), and the phonetic design of talk in overlap has been the object of study in interactional phonetics, too (French and Local 1983; Wells and Macfarlane 1998; Wells and Corrin 2004). The remarks which follow can only serve as the briefest of introductions to some of the relevant issues. Fragment 5 is taken from a telephone conversation involving the same participants as Fragment 2. The fragment comes from near the end of the call. They are making arrangements to meet up later in the day. (5) NB.II.5, 6:32
Phonetics and the management of talk-in-interaction
161
At first glance, the generally smooth organisation of turn-taking appears to have broken down: in lines 4 and 5 the participants appear to be talking over one another. One possibility is that Emma has already got the gist of what Lottie is saying in line 3, and decides to start up talk in line 5 without regard for how turn-taking is usually organised. Another possibility is that Emma has simply stopped listening to Lottie (though there is no evidence for this in Emma’s subsequent talk, which is fitted to the account Lottie has just given for the timing of her visit). A third possibility – which turns out to be correct – is that there is something in the way Lottie designs her talk in line 3 which makes Emma’s incoming talk at line 5 entirely legitimate. Notice that Emma’s incoming talk at line 5 is timed to begin just after a point of possible syntactic and pragmatic completion in Lottie’s talk: “I gotta wash out Gmy clo:thes” (line 3). Furthermore, the phonetic design of Lottie’s talk up to this point incorporates exactly those kinds of features associated with other designed-to-be- and treated-as-complete utterances. Lottie’s talk in line 3 is produced as a complete intonation phrase, with an overall declination of pitch and loudness which is characteristic of other possibly complete, transition relevant utterances. In addition, Lottie’s “clo:thes” exhibits falling pitch, with a fall of 3.8 ST. It is also elongated: a slowing down is a regular characteristic of other transition relevance places. Emma’s incoming at line 5, then, can be understood as entirely legitimate and a by-product not of failing to attend to Lottie’s talk at line 3, but of very close monitoring of its design. Overlapping talk following a point of transition relevance is quite common in naturally occurring conversation. However, overlapping talk may occur at other points relative to the ongoing talk. Fragment 6 – taken from a recording of an academic seminar – shows a case from French and Local (1983) where a current speaker, N, is clearly in the middle of a TCU at line 8 when a co-participant (K) starts up talk. (6) French and Local (1983:19)
162
Gareth Walker
K’s incoming talk at line 9 occurs at a point of non-completion in the ongoing talk and is hearable as competitive. That is, it directly challenges the ongoing speaker seeking sole turn occupancy at that point. This kind of competitive, in-overlap talk routinely exhibits certain phonetic features, most notably that it is higher and louder than usual for that speaker’s turn beginnings, higher and louder than the remaining unmarked portion of the talk, and loud relative to the ongoing talk. In the transcription, h and f mark out talk as high and loud respectively. These features of turn-competition are not always present in overlapping talk. For instance, Emma’s incoming talk in line 4 of Fragment 5 is designedly non-competitive: it is neither high pitched nor loud relative to her usual mode of speaking (in fact it is rather quiet and low pitched) and is both quieter and lower in pitch than the talk which Lottie produces simultaneously with it. It seems logical that things should be this way: that a speaker competing for the floor, and therefore talking at the same time as at least one other participant, would produce high pitched and loud talk. Conversely, producing talk in overlap which is low pitched and quiet would seem unsuited to the task of turn-competition. Incoming talk produced in overlap reveals connections between the design of the talk and the interactional goals which it seeks to accomplish. Furthermore, the co-participants orient to these two designs – competitive on the one hand, and noncompetitive on the other – rather differently in these cases. In Fragment 5 both speakers bring their turns to completion, whereas in Fragment 6 the current speaker (N) drops out in the face of K’s competitive incoming. More generally, these cases of talk produced in overlap reveal particular, and perhaps unexpected, orientations to indicators of transition relevance and the management of turn-taking. In addition to the placement of her incoming talk just after a point of possible turn completion, the design of Emma’s talk in line 4 of Fragment 5 as non-competitive embodies a claim as to the legitimacy of her (Emma) starting up at just that point: she need not do anything special to secure rights to the turn other than try to get in there first (Sacks, Schegloff and Jefferson 1974). On the other hand, in designing the talk in line 9 of Fragment 6 as turn-competitive, K is displaying an understanding that this incoming occurs at a point where a start-up is something other than legitimate or otherwise usual. Rather than undermining the case for the role of phonetic detail in the management of turn-taking in talk-in-interaction, the occurrence of talk in overlap – and the design features of that talk – strengthens it.
4.
Relationships between turns
An important function of the phonetic design of talk is to mark out how the current turn relates to preceding talk. In other words, the phonetic design of talk can mark out the syntagmatic relationships which occurs between turns. This typically involves marking out how what is being said now relates to what has come just be-
Phonetics and the management of talk-in-interaction
163
fore. On the basis of question-answer and telephone-call closing sequences, Goldberg (1978, 2004) argues that one way such relationships between utterances can be expressed is through shifts in amplitude, and by extension in perceived loudness. She argues that a downward shift (decrease) in the amplitude of successive utterances by the same speaker may affiliate the current utterance to a prior, and that an upward shift (increase) disaffiliates the current utterance from a prior. She argues that, sequence initiation is accompanied by raised amplitude, and that there is a tendency for the peak amplitude of utterances to descend over the course of a sequence. Goldberg is not the only analyst to describe ways in which the phonetic design of talk can mark sequential (dis)affiliation between successive utterances. Couper-Kuhlen (2004b) argues that increases in pitch and loudness are among the prosodic features which can mark out the current turn not as a continuation of the action which went before, but rather as the beginning of a new course of action. Exploring articulatory and phonatory aspects of speech as well as prosodic features, Local (2004) argues that a particular phonetic design of “and-uh(m)” marks out what follows not as connected to the immediate prior (cf. other uses of conjunctions in talk-in-interaction) but as a resumption of something occurring earlier in the interaction. These design features include the production of “and-uh(m)” with creaky voice and/or glottal stop at the start of “and”, a full (non-reduced) vowel quality in “and”, the production of the whole token with roughly level pitch in the middle of the speaker’s pitch range, a relatively slow rate, and no break between the two syllables. The phonetic design of talk can also be used to show syntagmatic relationships between utterances which are more complex than straightforward disjuncture. Consider Fragment 7, taken from a telephone call between Leslie and Ed, a music teacher. Ed has been telling Leslie about his having started rehearsals that day for a musical production at the end of the month. (7) Holt.O88.1.9, 3:28
At line 3 Leslie produces a first pair part to an adjacency pair (Schegloff 2007; Schegloff and Sacks 1973). In so doing, she makes a response from Ed expectable, or relevant. However, rather than provide the response Leslie’s first pair part made relevant, Ed continues his own previous line. Although timed to coincide with the end of Leslie’s first pair part, Ed’s “half term” (line 4) is clearly not the response which Leslie’s first pair part enquiry made relevant. Ed’s talk is hearable not as the expected second pair part response but as something related to – to use the locution
164
Gareth Walker
of Golberg (1978), as an affiliate of – his own earlier project. On completion of this talk which postpones the response made relevant by the first pair part, Ed produces the expected second pair part: “No:. I went to Spain.” (lines 4–5). Two features of the talk produced by Ed at the start of line 4 make it hearable as a postponement rather than the expected second pair part. First, Ed’s “half term” (line 4) is an account for why it is that he has just returned from holiday. In that respect, Ed’s talk connects his own prior talk at line 1 rather than to Leslie’s first pair part at line 3. Second, the phonetic design of Ed’s “half term” – particularly in terms of pitch and loudness characteristics – marks this out as a continuation of his previous project. In addition, the phonetic design of the talk which follows, along with its lexical content, (“.hh No:. I went to Spain.”, lines 4–5) mark out that talk as connecting not to Ed’s own first TCU, but to the sequence initiated by Leslie with her first pair part at line 3. Figure 2 provides a visual representation of aspects of Ed’s turn at (a) line 1 and (b) lines 4–5. It can be seen from the intensity trace (solid line, upper parts of the figures) that the postponing talk – Ed’s “half term” – has lower overall intensity (is quieter) than his talk in line 1. Table 2 presents certain measures of frequency and intensity. Table 2.
Frequency and intensity measures for parts of Fragment 7.
The mean peak intensity for the syllables in line 1 is 74.9 dB while the mean peak intensity for the syllables in “half term” is 66.3 dB. This is in line with the findings of Goldberg (1978, 2004): that broadly speaking, successive utterances in sequences have a lower amplitude than preceding ones. In addition to this decrease in loudness, Ed’s “half term” is produced with low pitch and a relatively narrow range of 3.8 ST (cf. Couper-Kuhlen 2004b’s reports of turns which continue sequences as having low pitch and being quiet relative to immediately prior talk from the same speaker). In contrast to the low pitched, quiet postponing talk, the postponed talk which follows in the same turn, “No:. I went to Spain.” (lines 4–5), is ‘stepped up’ in pitch and loudness. This part of the turn is produced with a much bigger pitch range than the postponing talk: 9.2 ST as opposed to 3.8 ST for the postponing talk. While each part of the turn has a similar lowest pitch, the postponed talk reaches much higher in the speaker’s pitch range. With respect to loudness, the mean peak intensity for syllables in this part of the turn is 73.9 dB: it is audibly considerably louder than the postponing talk.
Phonetics and the management of talk-in-interaction
(a) Line 1
(b) Lines 4–5 Figure 2. A visual representation of parts of Fragment 7.
165
166
Gareth Walker
What we are seeing in Fragment 7, then, is pitch and loudness being used to mark out syntagmatic relationships between utterances and parts of utterances. Features of pitch and loudness are being used to mark out the talk which immediately follows the second pair part as a continuation of a preceding sequence, rather than as the second pair part. The phonetic design of what follows in the same turn, now stepped up in pitch and in loudness, makes that talk hearable as a response – albeit a postponed one – to the first pair part. This consideration of syntagmatic relationships across speaking turns brings the discussion to the final task for phonetic design to be dealt with here: how the phonetic design of talk can signal relationships within turns.
5.
Relationships within turns
Line 1 of Fragment 2, “Whurdju gi:t th’turkey.” was described above as a turn consisting of a single TCU; it is grammatically complete, it implements a recognizable action and phonetically it is hearable as complete. One task handled by the phonetic design of that unit is marking it out as just that: a single, internally coherent unit. This internal coherence comes about as a result of a number of features of phonetic design. (A visual representation of aspects of this TCU is presented in Figure 1 above.) So, for instance, while there is an overall declination in loudness across the speaking turn, there are no disjunctive step-ups (or step-downs) in pitch or loudness (cf. the design of the multi-unit turn in Fragment 7; see also those multi-unit, multi-action turns involving an abrupt-join described in Local and Walker 2004). There are also no disjunctive changes in articulation rate; there is no dramatic speeding-up or slowing-down (again, cf. the abrupt-joins, and so-called rush-throughs whereby a speaker speeds up as s/he approaches a point of possible completion, Walker 2010). In articulatory terms, there are none of the kinds of oral or glottal cut-offs which can mark the occurrence of self-repair and the potential aborting of one TCU in favour of another (Jasperson 2002). In other words, this single unit is hearable as just that: a single unit. However, turns may have more complex internal structures than being composed of a single TCU. Speakers may draw upon phonetic resources to show how the various parts of such turns are related to one another. So, for instance, in Fragment 7, the increases in pitch and loudness on the postponed second pair part set that part of the turn apart from the postponing talk which preceded it. Another more complex turn-type is shown in lines 13–14 of Fragment 8, taken from a telephone conversation between Emma and Lottie; talk has been about dresses Lottie and Emma saw on a recent shopping trip.
Phonetics and the management of talk-in-interaction
167
(8) NB.IV.3, 2:45
Emma’s turn in lines 13–14 reaches possible grammatical and pragmatic completion at the end of “B’t I dLO:VE the b:one”; it is a complete sentence, and gives an assessment of one of the dresses they saw (“bone” refers to the colour of a dress). However rather than yield her turn at this point Emma continues her talk, using “the b:one” as the grammatical beginning of what follows: “the b:one w’z s:O:: …”. This shared element – the pivot – can therefore be interpreted as both the possible end of one grammatical unit, and the beginning of another (Walker 2007; see Scheutz 2005 and Norén 2007 for studies of related constructions in German and Swedish respectively). The pivot gains this possible and simultaneous leftwards and rightwards interpretation through a combination of its grammatical structure and its phonetic design. Figure 3 provides a visual representation of part of the turn in lines 13–14 of Fragment 8. A range of phonetic features bind the pivot to the pre-pivot which precedes it (allowing for leftwards interpretation of the pivot) and bind the post-pivot which follows to the pivot (allowing for rightwards interpretation of the pivot). For instance there are no disjunctive changes in pitch between the three components (pre-pivot, pivot, post-pivot). Instead, the pivot-initial “the” is produced within the (falling) pitch trajectory of final word of the pre-pivot (“LO:VE”); the first word of the post-pivot (“w’z”) is produced in the (falling) pitch trajectory of the pivot-final “b:one”. As well as an absence of pitch discontinuities, there are also no disjunctive changes in loudness: the mean peak intensity for syllables in the pre-pivot, pivot and post-pivot are 81.4 dB, 81.2 dB and 77 dB respectively. (In this case, then, the loudness characteristics are much the same as those for the single-TCU turn in line 1 of Fragment 2). There are no disjunctive changes of articulation rate within this part of the turn, and there is close temporal proximity of all components. In this particular case, the close temporal proximity manifests itself through continued phonation between the pre-pivot and the pivot, and between the pivot
168
Gareth Walker
Figure 3. A visual representation of part of Fragment 8.
and the post-pivot. As well as being audible this is evidenced in the figure by periodicity in the waveform around these points, and by the phonetic analysis software being able to find voiced frames across these ‘joins’. These features of pitch, loudness, articulation rate and temporal proximity all give this part of Emma’s turn a sense of coherence; of being produced as a single, albeit complex, unit. There is also a routine absence of phonetic discontinuity where a speaker produces an addition to a possibly complete turn. Fragment 9 is taken from a recording of two student friends engaged in a face-to-face interaction at a UK university; speaker G is British, speaker H, American. (9) Walker (2004:161)
Phonetics and the management of talk-in-interaction
169
In line 5 H brings her talk to possible completion: “English people don’t go to Germany on holiday”. In the absence of any uptake following her bringing talk to possible completion (line 126 extends her talk by producing a grammatically fitted continuation of it at line 7: “generally”. Such grammatically fitted continuations – increments (Ford, Fox, and Thompson 2002; Schegloff 1996) – routinely exhibit a range of phonetic features which mark out the increment as a recompletion of its host TCU.4 In this case, a range of phonetic features mark out “generally” (line 7) as fitted to “English people don’t go to Germany on holiday” (line 10). Figure 4 provides visual representations of the increment and its host. It can be seen from Figure 4 that the articulation rate of the increment is very similar to that of the end of the host; the host-final “holiday” measures 527 ms while “generally” (the increment) measures 548 ms. There is no disjunctive speeding up or slowing down, which can accompany some turn continuations (see e.g. Local and Walker 2004). The increment is produced in the same part of the speaker’s pitch range as the end of the host. The pitch contours are also the same in each case: they show low, falling-rising pitch. There are no disjunctive changes in loudness: the increment is neither significantly quieter, nor significantly louder, than the end of the host; “holiday” has a mean intensity of 75.8 dB while “generally” has a mean intensity of 76.9 dB. In this particular case there are also interesting similarities in the phonation types evident at the end of the host and at the end of the increment. The final syllable of the host (“day”) is produced with creaky voice, beginning at about 1.7 s and lasting for around 150 ms. This change in phonation type is evident in Figure 4(b), in both the waveform (where there is aperiodicity) and the spectrogram (the vertical striations which correspond to the vibrations of the vocal folds have become irregular). Note also the inability of the computer software to reliably detect periodicity in the waveform in Figure 4(a): there is a ‘gap’ in the F0 trace. This creaky voice is followed by a short period of breathy voice. The same changes in phonation are evident in the final syllable of the increment; creaky voice begins at around 2.6 s and last for approximately 85 ms, before a portion of breathy voice. These phonetic features of pitch, loudness, articulation rate and in some cases phonation and articulation all contribute to marking out the increment as fitted to its host, just as its syntactic make-up does. In other words, features of phonetic design and syntax mark out cohesive syntagmatic relationships between the different parts of the turn. The converse of the features observed in pivot constructions and increments to TCUs – disjunctive changes in pitch, loudness, articulation rate, ‘cut-offs’ and silences – are all features which can mark out that some new line is being taken (Couper-Kuhlen 2004b; Goldberg 1978, 2004; Jasperson 2002; Local 1992; Local and Walker 2004). In addition, in other sequential environments some of these features can mark out that a previous line is being resumed. Consider Fragment 10, taken from a broadcast radio interview between interviewer James Naughtie (JN) and Lord Falconer (Fal), then Secretary of State for Justice in the British Govern-
170
Gareth Walker
(a) Labelled F0 trace, intensity trace (upper part) and waveform (lower part)
(b) Spectrogram (upper part) and waveform (lower part) of “holiday” (left) and “generally” (right) Figure 4. Visual representations of parts of Fragment 9.
Phonetics and the management of talk-in-interaction
171
ment; questions have been about the government’s immigration policy and comments made by Alan Duncan MP, then Shadow Secretary of State for Constitutional Affairs. (10) Today 05/04/2004, 6:20
In line 4 JN makes a start on his next question with “do we need”, before breaking off and producing talk which is not a straightforward continuation of his “do we need”, but on a rather different line. His “do we need” is then recycled in line 5 and his question brought to completion. Embedded within the bounds of JN’s TCU, then, is talk (marked in the transcription with dashed underlining) which is inserted into, or parenthetical to, his question. Local (1992) deals with a number of similar sequences and shows among other things that parenthetical talk is regularly faster in rate, lower in pitch and greater in loudness than (i) the talk which immediately precedes and (ii) the talk which follows it. Mazeland (2007) deals with aspects of parenthetical sequences in Dutch, and reports similar findings concerning their phonetic design. And it is in part the phonetic design of JN’s turn in Fragment 10 which marks out this insertion as parenthetical to the main question. First, the phonetic design of the talk marks out where the parenthetical begins i.e. where JN begins to take a different line from that projected by his turn initial “do we need”. One audible change beginning with “just” is in terms of articulation rate: JN’s parenthetical talk is noticeably faster than the talk which precedes it and which follows it. His turn-initial “do we need” is produced at a rate of 5.4 syllables per second, his parenthetical talk is produced at a faster rate of 6.2 syllables per second, and his post-parenthetical talk up to the end of his turn is produced at a rate of 5.1 syllables per second. As well as overall speeding up relative to the surrounding talk, the production of this talk low in the speaker’s pitch range also signals that the speaker is in the midst of a parenthetical. Fundamental frequency minimums, maximums, means and ranges for the talk preceding the parenthetical, the parenthetical and the remainder of the turn are shown in Table 3. Figure 5 provides a visual representation of parts of JN’s turn. A second task handled by the phonetic design of the talk, with regard to the structuring of the turn, is to mark out where the parenthetical talk ends. There are at least two relevant phonetic features which mark out that the parenthetical is coming to an end. The first is that the “on” of JN’s “before I move on” (line 10) is produced with falling-rising pitch. Local (1992: 278) reports that what he labels
172
Gareth Walker
Table 3.
Frequency measures for parts of Fragment 10.
self-interrupting talk – of which that in Fragment 10 is one kind – is routinely characterised by rising pitch at its end. The second relevant feature is the inbreath which JN produces immediately after “on”. While an inbreath does not mark the end of a parenthetical per se, this particular inbreath has that sort of quality (including a noticeably high rate of ingressive airflow when compared with other inbreaths from JN, such as that later in this same turn at line 12) which is “characteristic of some disjunctive next move” (Drew and Holt 1998: 507–8). These features – final rising pitch and an ‘emphatic’ inbreath – coupled with the arrival at a point of possible syntactic completion for the parenthetical, project most strongly that the parenthetical has ended and the line taken earlier is about to be resumed. A third task for the phonetic design of the talk to handle is to mark out the talk which follows the parenthetical not as further parenthetical talk, but as a return to the talk begun earlier. As Table 4 shows, JN’s second, resumptive “do we need” is produced with markedly higher pitch than the end of parenthetical talk (“before I move on”), is louder and has a lower articulation rate i.e. is produced more slowly. Importantly, the pitch, loudness and articulation rate characteristics also mark out this resumptive “do we need” as picking up the earlier line started at the beginning of JN’s turn then postponed. Table 4.
Frequency, intensity and articulation rate measures for parts of Fragment 10.
In Fragment 10 and other cases like it there is first a break, then some talk on a slightly different line, and there is later a resumption. This break-and-resumption organisation is signalled by the phonetic design of the talk in concert with the lexis. There is one further point to note about this fragment, which speaks to the more
Phonetics and the management of talk-in-interaction
Figure 5. A visual representation of part of Fragment 10.
173
174
Gareth Walker
general issue of relationship between phonetic design and the lexical-grammatical format of talk. The point at which the parenthetical begins in Fragment 10 is not simply marked out by a break in the unfolding grammatical structure of the turn in progress, with phonetic design in some kind of supporting role. In fact, in Fragment 10 the lexical choices JN makes does not make an understanding of the parenthetical as a parenthetical necessary at all. He could be asking Lord Falconer the question “do we need just to pick up on another of Mr Duncan’s questions before I move on?” This is not the case, and Lord Falconer does not treat JN’s talk as such a question. However, it is only clear that this is not the case when phonetic design is considered, as it would be by the participants themselves, as an integral part of the unfolding turn and not as some sort of ‘optional extra’ to be drawn upon by the participants where necessary.
6.
Future directions
One aim in this article has been to show some of the ways in which the phonetic design of talk figures in the management of talk-in-interaction. It is important to note that the tasks described here – the management of turn-taking and overlap, the marking of relationships between turns, and the marking of relationships within turns – are not the only ways in which the phonetic design of talk figures in the management of talk-in-interaction. They are not necessarily even the most basic tasks for the phonetic design of talk to handle. One basic function of phonetic detail not discussed here is preserving the recognisability of lexical items, which must be done irrespective of other tasks which the phonetic design of talk is handling at any given point. Furthermore, empirical work has shown that there are other areas not discussed here where the phonetic design of talk makes a contribution, such as the marking of stance and affect (Couper-Kuhlen 2004a; Local and Walker 2008), contributing to the particular action being performed in a turn (CouperKuhlen 2001; Ogden 2006; Ogden, Hakulinen, and Tainio 2004; Selting 1996), and the assignation of blame for problems of understanding (Curl 2005). The three tasks described in detail here are however the tasks which at the time of writing we know most about, particularly in terms of their management across languages. Part of the enterprise of interactional phonetics to date has been to identify techniques of sequential and phonetic analysis and representation which provide for exploring the function of phonetic detail in the management of talk-in-interaction. There are therefore many possible future directions. Most obviously, attempts can be made to refine our understandings of how phonetic detail functions in the management of the three tasks outlined here. There are also more specialised research contexts where techniques of interactional phonetics could yield significant insights, and where promising starts have been made: see, for instance, work in interactional phonetics on the talk of young children (e.g. Corrin, Tarplee, and Wells
Phonetics and the management of talk-in-interaction
175
2001; Tarplee 1996; Wells and Corrin 2004). Clinical and other practical applications are also under-explored from the perspective of interactional phonetics (but cf. Auer and Rönfeldt 2004; Local and Wootton 1995; Wells and Local 1993 for promising starts on interactional phonetics in clinical contexts). Work by Goodwin and colleagues on interactions involving a man suffering post-stroke aphasia emphasises how a limited set of lexical items can be used with different phonetic designs in the co-construction of meaning (Goodwin 1995; Goodwin, Goodwin, and Olsher 2002). Goodwin’s work also draws attention to perhaps the most significant gap in our understanding of talk-in-interaction: how phonetic resources mesh with the visual (see also Walker in press). Interactional phonetics continues to develop. Over the last 20 years there has been a steady increase in the number of analysts working from the perspective of interactional phonetics, and consequently a steady growth in the amount of published empirical research. While the future looks very bright, there are a number of challenges which will need to be met as interactional phonetics diversifies and attracts new practitioners. First, as technological developments make increasingly large databases of audio-visual material available, it will be important not to lose sight of one of the core characteristics of interactional phonetics: that any analysis arises out of, and must account for, the details of single episodes of interaction. Second, relatively inexpensive but very powerful computers and software (which in some cases can be downloaded free of charge) make available a range of different kinds of acoustic analyses. However, one challenge is ensuring that those analyses are informed by a working knowledge of (at least basic) speech production and perception. In the context of interactional phonetics, acoustic analysis (such as that made possible by Praat and other acoustic analysis software), should not be seen as a substitute for careful auditory parametric analysis, but rather as its computer-based counterpart. A third challenge is for those working in, or close to, interactional phonetics, to look at the speech signal in its entirety, encompassing and attending equally to features at quality, duration, pitch and loudness (Laver 1994: 27). If these challenges can be met, interactional phonetics will surely continue to provide insights into the ways in which phonetic detail figures in the management of talk-in-interaction.
Notes 1. With the exception of minor typographical changes, transcriptions are included as they appeared either in transcriptions prepared by Gail Jefferson or in original publications (in which case citations are given in the fragment label). Whatever their source, all transcriptions broadly follow those conventions set out in Jefferson (2004). Any further relevant conventions are explained in the discussion. Audio samples to accompany this article are available via http://gareth-walker.staff.shef.ac.uk/.
176
Gareth Walker
2. Ladefoged (2003) provides a useful introduction to the acoustic properties of speech and their analysis. All acoustic analyses reported in this article were performed using the Praat analysis software: http://www.fon.hum.uva.nl/praat/. Praat was also used in the preparation of all figures. 3. For technical reasons, underlined text in the transcriptions appears as italicized text in the figure labels. 4. For accounts of comparable continuations in German with consideration of phonetic aspects, see Auer (1996); for consideration of increments in other languages, see the papers in Couper-Kuhlen and Ono (2007).
References Auer, Peter 1996 On the prosody and syntax of turn continuation. In: Elizabeth Couper-Kuhlen and Margret Selting (eds.), Prosody in Conversation: Interactional Studies, 57–94. Cambridge: Cambridge University Press. Auer, Peter, Elizabeth Couper-Kuhlen and Frank Müller 1999 Language in Time: The Rhythm and Tempo of Spoken Interaction. Oxford: Oxford University Press. Auer, Peter and Barbara Rönfeldt 2004 Prolixity as adaptation: prosody and turn-taking in German conversation with a fluent aphasic. In: Elizabeth Couper-Kuhlen and Cecilia E. Ford (eds.), Sound Patterns in Interaction, 171–200. Amsterdam: John Benjamins. Catford, J.C. 1977 Fundamental Problems in Phonetics. Bloomington: Indiana University Press. Corrin, Juliette, Clare Tarplee and Bill Wells 2001 Interactional linguistics and language development: A conversation analytic perspective on emergent syntax. In: Margret Selting and Elizabeth CouperKuhlen (eds.), Studies in Interactional Linguistics, 199–225. Amsterdam: John Benjamins. Couper-Kuhlen, Elizabeth 1993 English Speech Rhythm: Form and Function in Everyday Verbal Interaction. Amsterdam: John Benjamins. Couper-Kuhlen, Elizabeth 2001 Interactional prosody: High onsets in reason-for-the-call turns. Language in Society 30:29–53. Couper-Kuhlen, Elizabeth 2004a Analyzing language in interaction: The practice of never mind. English Language and Linguistics 8(2):207–237. Couper-Kuhlen, Elizabeth 2004b Prosody and sequence organization in English conversation. In: Elizabeth Couper-Kuhlen and Cecilia E. Ford (eds.), Sound Patterns in Interaction, 335–376. Amsterdam: John Benjamins. Couper-Kuhlen, Elizabeth and Tsuyoshi Ono (eds.) 2007 Turn continuation in cross linguistic perspective. Special issue of Pragmatics, 17 (4).
Phonetics and the management of talk-in-interaction
177
Curl, Traci S. 2005 Practices in other-initiated repair resolution: The phonetic differentiation of ‘repetitions’. Discourse Processes 35(1):1–43. Drew, Paul 2004 Conversation analysis. In: Kristine Fitch and Robert Sanders (eds.), Handbook of Language and Social Interaction, 71–102. New Jersey: Lawrence Erlbaum. Drew, Paul and Elizabeth Holt 1998 Figures of speech: Figurative expressions and the management of topic transition in conversation. Language in Society 27:495–522. Ford, Cecilia E., Barbara A. Fox and Sandra A. Thompson 2002 Constituency and the grammar of turn increments. In: Cecilia E. Ford, Barbara A. Fox and Sandra A. Thompson (eds.), The Language of Turn and Sequence, 14–38. New York: Oxford University Press. Ford, Cecilia E. and Sandra A. Thompson 1996 Interactional units in conversation: Syntactic, intonational, and pragmatic resources for the management of turns. In: Elinor Ochs, Emanuel A. Schegloff and Sandra A. Thompson (eds.), Interaction and Grammar, 134–184. Cambridge: Cambridge University Press. Foulkes, Paul and Gerard Docherty 2006 The social life of phonetics and phonology. Journal of Phonetics 34:409–438. Fox, Barbara A. 2001 An exploration of prosody and turn projection in English conversation. In: Margret Selting and Elizabeth Couper-Kuhlen (eds.), Studies in Interactional Linguistics, 287–315. Amsterdam: John Benjamins. French, Peter and John Local 1983 Turn competitive incomings. Journal of Pragmatics 7:701–715. Goldberg, Jo Ann 1978 Amplitude shift – a mechanism for the affiliation of utterances in conversational interaction. In: Jim Schenkein (ed.), Studies in the Organization of Conversational Interaction, 199–218. New York: Academic Press. Goldberg, Jo Ann 2004 The amplitude shift mechanism in conversational closing sequences. In: Gene H. Lerner (ed.), Conversation Analysis: Studies From the First Generation, 257–297. Amsterdam: John Benjamins. Goodwin, Charles 1995 Co-constructing meaning in conversations with an aphasic man. Research on Language and Social Interaction 28(3):233–260. Goodwin, Charles, Marjorie H. Goodwin and David Olsher 2002 Producing sense with nonsense syllables: Turn and sequence in the conversations of a man with severe aphasia. In: Cecilia E. Ford, Barbara A. Fox and Sandra A. Thompson (eds.), The Language of Turn and Sequence, 56–80. New York: Oxford University Press. Goodwin, Charles and John Heritage 1990 Conversation Analysis. Annual Review of Anthropology 19:283–307. Jasperson, Robert 2002 Some linguistic aspects of closure cut-off. In: Cecilia E. Ford, Barbara A. Fox and Sandra A. Thompson (eds.), The Language of Turn and Sequence, 257–286. New York: Oxford University Press.
178
Gareth Walker
Jefferson, Gail 1973 A case of precision timing in ordinary conversation: Overlapped tag-positioned address terms in closing sequences. Semiotica 9(1):47–96. Jefferson, Gail 1986 Notes on ‘latency’ in overlap onset. Human Studies 9(2–3):153–183. Jefferson, Gail 2004 Glossary of transcript symbols with an introduction. In: Gene H. Lerner (ed.), Conversation Analysis: Studies From the First Generation, 13–31. Amsterdam: John Benjamins. Ladefoged, Peter 2003 Phonetic Data Analysis: An Introduction to Phonetic Fieldwork and Instrumental Techniques. Oxford: Blackwell. Laver, John 1994 Principles of Phonetics. Cambridge: Cambridge University Press. Local, John 1992 Continuing and restarting. In: Peter Auer and Aldo di Luzio (eds.), The Contextualization of Language, 273–296. Amsterdam: John Benjamins. Local, John 2003 Variable domains and variable relevance: Interpreting phonetic exponents. Journal of Phonetics 31:321–339. Local, John 2004 Getting back to prior talk: and-uh(m) as a back-connecting device. In: Elizabeth Couper-Kuhlen and Cecilia E. Ford (eds.), Sound Patterns in Interaction, 377–400. Amsterdam: John Benjamins. Local, John and John Kelly 1986 Projection and ‘silences’: Notes on phonetic and conversational structure. Human Studies 9:185–204. Local, John, John Kelly and Bill Wells 1986 Towards a phonology of conversation: Turn-taking in Tyneside English. Journal of Linguistics 22:411–437. Local, John and Gareth Walker 2004 Abrupt-joins as a resource for the production of multi-unit, multi-action turns. Journal of Pragmatics 36(8):1375–1403. Local, John and Gareth Walker 2005 Methodological imperatives for investigating the phonetic organisation and phonological structures of spontaneous speech. Phonetica 62(2–4):120–130. Local, John and Gareth Walker 2008 Stance and affect in conversation: On the interplay of sequential and phonetic resources. Text & Talk 28(6):723–747. Local, John, Bill Wells and Mark Sebba 1985 Phonology for conversation: Phonetic aspects of turn-delimitation in London Jamaican. Journal of Pragmatics 9(2–3):309–330. Local, John and Anthony J. Wootton 1995 Interactional and phonetic aspects of immediate echolalia in autism: A case study. Clinical Linguistics and Phonetics 9:155–194. Mazeland, Harrie 2007 Parenthetical sequences. Journal of Pragmatics 39(10):1816–1869.
Phonetics and the management of talk-in-interaction
179
Norén, Niklas 2007 Apokoinou in Swedish talk-in-interaction: A family of methods for grammatical construction and the resolving of local communicative projects. Linköping: Linköping University, available from http://urn.kb.se/resolve?urn=urn:nbn:se: liu:diva-9869. Obeng, Samuel Gyasi 1991 Pitch, loudness and turn regulation in Akan conversation. York Papers in Linguistics 15:221–235. Ogden, Richard 2001 Turn transition, creak and glottal stop in Finnish talk-in-interaction. Journal of the International Phonetic Association 31(1):139–152. Ogden, Richard 2004 Non-modal voice quality and turn-taking in Finnish. In: Elizabeth CouperKuhlen and Cecilia E. Ford (eds.), Sound Patterns in Interaction, 29–62. Amsterdam: John Benjamins. Ogden, Richard 2006 Phonetics and social action in agreements and disagreements. Journal of Pragmatics 38(10):1752–1775. Ogden, Richard, Auli Hakulinen and Liisa Tainio 2004 Indexing ‘nothing much’ with stylization in Finnish. In: Elizabeth CouperKuhlen and Cecilia E. Ford (eds.), Sound Patterns in Interaction, 299–334. Amsterdam: John Benjamins. Sacks, Harvey, Emanuel A. Schegloff and Gail Jefferson 1974 A simplest systematics for the organization of turn-taking for conversation. Language 50(4):696–735. Schegloff, Emanuel A. 1993 Reflections on quantification in the study of conversation. Research on Language and Social Interaction 26(1):99–128. Schegloff, Emanuel A. 1996 Turn organization: One intersection of grammar and interaction. In: Elinor Ochs, Emanuel A. Schegloff and Sandra A. Thompson (eds.), Interaction and Grammar, 52–133. Cambridge: Cambridge University Press. Schegloff, Emanuel A. 2000 Overlapping talk and the organization of turn-taking for conversation. Language in Society 29:1–63. Schegloff, Emanuel A. 2007 Sequence Organization in Interaction: A Primer in Conversation Analysis 1. Cambridge: Cambridge University Press. Schegloff, Emanuel A. and Harvey Sacks 1973 Opening up closings. Semiotica 8:289–327. Scheutz, Hannes 2005 Pivot constructions in spoken German. In: Auli Hakulinen and Margret Selting (eds.), Syntax and Lexis in Conversation, 103–128. Amsterdam: John Benjamins. Selting, Margret 1996 Prosody as an activity-type distinctive cue in conversation: the so-called ‘astonished’ questions in repair initiation. In: Elizabeth Couper-Kuhlen and Mar-
180
Gareth Walker
gret Selting (eds.), Prosody in Conversation: Interactional Studies, 231–270. Cambridge: Cambridge University Press. Selting, Margret 2000 The construction of units in conversational talk. Language in Society 29:477–517. Tarplee, Clare 1996 Working on young children’s utterances: prosodic aspects of repetition during picture labelling. In: Elizabeth Couper-Kuhlen and Margret Selting (eds.), Prosody in Conversation: Interactional Studies, 406–435. Cambridge: Cambridge University Press. Walker, Gareth 2004 On some interactional and phonetic properties of increments to turns in talkin-interaction. In: Elizabeth Couper-Kuhlen and Cecilia E. Ford (eds.), Sound Patterns in Interaction, 147–169. Amsterdam: John Benjamins. Walker, Gareth 2007 On the design and use of pivots in everyday English conversation. Journal of Pragmatics 39(12):2217–2243. Walker, Gareth 2010 The phonetic constitution of a turn-holding practice: Rush-throughs in English talk-in-interaction. In: Dagmar Barth-Weingarten, Elisabeth Reber and Margret Selting (eds.), Prosody in Interaction, 51–72. Amsterdam: John Benjamins Walker, Gareth In press Coordination and interpretation of vocal and visible resources: ‘Trail-off’ conjunction. Language and Speech. Wells, Bill and Juliette Corrin 2004 Prosodic resources, turn-taking and overlap in children’s talk-in-interaction. In: Elizabeth Couper-Kuhlen and Cecilia E. Ford (eds.), Sound Patterns in Interaction, 119–144. Amsterdam: John Benjamins. Wells, Bill and John Local 1993 The sense of an ending: A case of prosodic delay. Clinical Linguistics and Phonetics 7:59–73. Wells, Bill and Sarah Macfarlane 1998 Prosody as an interactional resource: Turn-projection and overlap. Language and Speech 41(3–4):265–294. Wells, Bill and Sue Peppé 1996 Ending up in Ulster: Prosody and turn-taking in English dialects. In: Elizabeth Couper-Kuhlen and Margret Selting (eds.), Prosody in Conversation: Interactional Studies, 101–130. Cambridge: Cambridge University Press. Wootton, Anthony J 1989 Remarks on the methodology of conversation analysis. In: Derek Roger and Peter Bull (eds.), Conversation: An Interdisciplinary Perspective, 238–258. Clevedon: Multilingual Matters.
7.
Prosody and pragmatic effects Anne Wichmann
1.
Introduction
Tone of voice is an elusive aspect of spoken communication and yet it has the power to tell us not only much about the speaker but also about what a speaker is doing in a particular communicative context. There is as yet no coherent approach to the study of prosody in a sociopragmatic context, but by reviewing many disparate existing studies, we can assemble some parts of a complex picture. This chapter aims to show what we know already, and how such research might fruitfully develop. The disparateness of approaches to the study of prosody in its social context is due in part at least to the many different groups undertaking it. On the one hand, these include phoneticians (e.g. Fletcher and Harrington 2001, Foulkes and Docherty 2006), typologists (e.g. Jun 2006) and pragmaticians (e.g. Culpeper 2005, Culpeper et al. 2003), who are interested in spoken communication for its own sake. On the other hand there are speech technologists (e.g. Kousidis et al. 2009), clinicians (e.g. McCann and Peppé 2003), applied linguists (e.g. Mennen et al. 2010) and forensic scientists (e.g. Shriberg and Stolcke 2008), who are motivated by very different, practical applications. This wide variety of approaches and methods means that the results we have are often difficult to compare, and comparability is further impeded by differences between the models of prosody on which the studies are based. The social variables that have been correlated with prosodic phenomena are only a subset of those common in the field of sociolinguistics. To our knowledge, there is, for example, no research that attempts to correlate social class with prosody. There are, however, studies of prosody in relation to other social variables, including more stable, speaker-related ones such as age, gender and region of origin, and also more dynamic, context-specific variables such as speaking mode, social distance, politeness and power. Most of these studies consider the sociopragmatic meanings generated by utterance-internal speaker choices in prosody. These choices include gradient prosodic phenomena, such as higher or lower pitch over a stretch of speech, and also categorical phonological choices, such as the choice of a final rising contour rather than a final falling contour. Recent work, however, has begun to see the potential for creating sociopragmatic meaning generated not by any inherent characteristics of an utterance itself but by the sequential relationship between an utterance and that of another speaker. This is an aspect of prosodic meaning that has received little attention,
182
Anne Wichmann
and has important theoretical consequences for our understanding of prosodic behaviour. This chapter will begin by outlining the prosodic resources available to speakers, and some of the methods used to study them. It will then provide an overview of prosody research in relation to speaker-related parameters. Finally I will review some of the research into how prosody reflects and constructs more dynamic meanings such as power and social distance. However, before doing so, I will make a brief excursus to describe the prosodic phenomenon of pitch concord, and give an account of Social Accommodation Theory, in order to provide a theoretical framework for this aspect of prosodic behaviour. 1.1.
Prosodic resources
The vocal resources that are collectively known as prosody include four main components: pitch, loudness, tempo and voice quality. These features contribute, in the first instance, to the expression of paralinguistic meaning, such as aggression, solidarity, affection and distance, and they also convey the speaker’s emotional state: a voice can be raised in anger, whispered in complicity or lowered in sadness. Such features tend to be gradient or scalar, and the phonetic continuum maps closely onto the semantic continuum – a raised voice may be angry, and a very raised voice may be very angry. The same prosodic components, and pitch in particular, also create the linear sequence of elements that constitute the “grammar” of intonation, or intonational phonology. The most extensive study of the linguistic structure of intonation is by Ladd, who shows that “intonation has a categorical linguistic structure, consisting of a linear sequence of phonological events that occur at well-defined points in the utterance” (1996: 41). The great difficulty in the study of prosody, and especially when studying prosodic variation, is that the same channel is used for both paralinguistic and linguistic meaning, and it is not always easy to separate the two. In Ladd’s words, “(t)he close acoustic and semiotic connection between intonation and paralinguistic cues is unquestionably the most important problem in studying intonation” (1996: 40). Many aspects of variation concern differences in global phonetic characteristics of speech – such as whether a voice is generally high or generally low, or whether the speech is fast or slow. Other kinds of variability are found in the phonetic realisation of individual phonological categories, analogous to variation in phoneme realisation in regional accents. Therefore, if we are to study sociopragmatic variation, we need to consider not only the broad gradient characteristics of how people speak, but also the “grammatical” (i.e. categorical phonological) choices they make. Unfortunately, while there is general agreement that such systems exist (Crystal 1969, Cruttenden 1997 [1986], Ladd 1996), there is no universal agreement as to how they should be modelled (see e.g. Hirst and Di Cristo 1998).
Prosody and pragmatic effects
183
There are broadly speaking two main approaches to intonational phonology: the British system of “nuclear” tones (e.g. Crystal 1969, Cruttenden 1997 [1986]) and the American autosegmental system of pitch targets or levels (Pierrehumbert 1980, Ladd 1996). In the British tradition of intonation analysis, within each intonation domain (a.k.a. tone group, tone unit, intonational phrase, intonation domain) the nucleus, normally the last prominent syllable in a tone group, is the only obligatory element. It is optionally preceded by the prehead (any unstressed syllables preceding the head) and head (the first accented syllable in the tone group), and followed by the tail (any unstressed syllables following the nucleus up to the end of the intonation domain). In tone groups containing only one prominent syllable, this syllable is simultaneously head and nucleus. It may be preceded and/or followed by unstressed syllables. The structure of these components is as follows: Intonation Domain: (prehead) (head) nucleus (tail) (Crystal 1969: 208, optional components in brackets). The major patterns of pitch movement (also referred to as “tones” or “nuclear tones”) identified in the British tradition include the fall (from a high accented syllable, cf. Figure 1a), rise (from a low accented syllable, Figure 1b), fall-rise (Figure 1c), and rise-fall (Figure 1d). Nuclear tones normally begin on the nuclear syllable and extend to the end of the intonational domain. The American AutosegmentalMetrical (AM) system also identifies intonational domains, but the holistic contours of the British system are decomposed into underlying pitch targets, or “tones”, H (high) or L (low). A starred pitch accent (H* or L*) indicates its association with a prominent (accented) syllable. A nuclear fall, for example, is thus the interpolation between a High starred pitch accent (H*) and a Low target (L); thus H* L (see Figure 1). This autosegmental model has now become the international standard in intonational phonology (Ladd 1996; Gussenhoven 2004) and for typological comparison (Jun 2006; Grabe and Post 2002). An important distinction between the two traditions is that the British system was developed primarily for auditory, impressionistic analysis, while the AM model is more often associated with a mixture of auditory and instrumental analy-
Figure 1. Stylised nuclear tones and AM equivalents, based on Wichmann (2000:11)
184
Anne Wichmann
sis. Binary pitch targets lend themselves better to the annotation of the Fundamental Frequency (F0) curve1, and of course for any computational operations. Annotation, even with the visual help of the F0 curve, is a time-consuming business (see Wichmann 2008) and therefore costly. It lends itself best to small studies of preidentified linguistic phenomena (e.g. the realization of discourse markers (e.g. Astruc-Aguilera 2005, Aijmer 2002, Lam 2009) and the nuclear tones associated with speech acts, e.g. Stenström 1994: 84–96, Hedberg et al. 2006). Larger quantitative studies such as Shriberg et al. (1998) tend therefore to rely on automatic analysis, which can be applied over longer stretches of the speech signal, for example to identify the average maximum pitch over a stretch of speech, or the average pitch range, and frequency of change of direction. Studies motivated by technological applications – e.g. automatic speech recognition and text-to-speech synthesis – are more likely to explore global gradient features in this way than to carry out close manual annotation of the text focusing on phonological structure (see e.g. Hirschberg et al. 2004, Shriberg et al. 1998). Given that the majority of work on prosody is currently in the technical domain, most recent studies are of global, gradient and automatically analysable characteristics of the voice rather than of choices within the phonological system. To talk about the use of prosody in a social context means that we are not only faced with various notions of social variables, with varying degrees of stability, but with any one or more of a range of prosodic phenomena. An example might be the frequency and distribution of certain prosodic events, such as a rising or falling final tone on a question or statement. One can also examine the phonetic realisation of such events; expressed in terms of holistic contours, these could be steep or shallow falls, low or high rises, or, expressed in autosegmental terms, the pitch value or timing of individual pitch targets such as H* or L. Finally, one can relate social and contextual parameters to the more global, gradient aspects of speech – overall high pitch, greater or lesser exploitation of pitch range (wide or narrow fluctuations), slow or fast speech. 1.2.
Methodological approaches
Analysing prosody is not straightforward. Some long-term features, such as average pitch, or speech and articulation rate (based on pause identification), can be analysed automatically, but the identification of phonologically significant points in the pitch contour requires a knowledge of the language, and generally relies on both auditory and acoustic analysis. Given the extreme variability of the human voice, analogous to that of handwriting, it is not surprising that there are cases where even experts do not agree, or are unsure about how to analyse certain prosodic events. The kind of data used varies greatly. Some studies, or parts of studies, are carried out under laboratory conditions, where volunteer speakers read aloud written
Prosody and pragmatic effects
185
sentences (e.g. Grabe et al. 2001). Others, also laboratory-based, are designed to elicit unscripted but highly controlled speech. The best known of these elicitation techniques is the Map Task, as used for example by Shobbrook and House (2003) for a study of the high rising terminal, and Grabe et al. (2001) for the study of regional variation in British intonation. In this task, two participants work together, each with a stylised map. One participant has a route marked, and is required to describe this route so that the other participant can draw it into their map. They do not know initially that the maps are not identical (they are not able to see each other’s maps). While this task allows strict control of speakers (age, gender etc.) and of the words used (referring to the items on the map, such as “windmill”, “house”, “forest”), it also elicits a very limited range of speech acts – some questions, but mainly a string of instructions and responses. This limits the usefulness of such data for prosodic research. Those wishing to analyse the prosody of naturally-occurring speech, especially casual conversation, tend to use their own specially-collected recorded data, such as radio phone-in programmes or other publicly available data. This kind of data is used by those working in the Conversation Analytic (CA) framework, where there is a long tradition of close auditory analysis, captured, partially at least, in the transcription conventions developed by Jefferson (e.g. 1985). More recently, modifications to these transcriptions have been suggested (see Walker 2004) and stretches of speech that are of particular interest are subjected in addition to instrumental analysis, using freely available speech software, e.g. PRAAT (Boersma and Weenink 2005). Few studies of prosody now rely entirely on auditory analysis, but it is important to note that auditory and acoustic analysis are complementary, and speech software can assist but not replace the human listener, except in the calculation of long-term trends such as pitch range. Studies of conversation, such as Couper-Kuhlen and Ford (2004), are typical of this approach. They are concerned less with the effects of social variables and more with the local management of the unfolding interaction. However, CA methods have also been used to identify patterns of inter-speaker relationships, in their terms “affiliation” and “disaffiliation”, and this is discussed more fully in section 3.2 below. Finally, for those interested in the patterns of occurrence and distribution of prosodic events (e.g. Wichmann 2004) in naturally occurring speech, whether scripted (e.g. broadcast speech) or spontaneous (e.g. casual conversation), there are several spoken corpora available. Given the time-consuming nature of prosodic transcription, it is not surprising that there are very few prosodically-transcribed corpora available. For British English these include the London-Lund Corpus (Svartvik 1990) and the Lancaster/IBM Spoken English Corpus (Knowles et al. 1996), and for American English the Corpus of Spoken American English (CSAE) (see e.g. Du Bois et al. 2000). A more recent collection of spoken English, which is tagged and parsed2 but not prosodically transcribed, is the British component of the International Corpus of English (ICE GB) (Nelson et al. 2002). (For a detailed
186
Anne Wichmann
overview of available corpora, see Wichmann 2008). Prosodic research using ICE GB (e.g. Dehé and Wichmann 2010, Wichmann 2001, 2004, 2005, 2009, Wichmann et al. 2010) is based on a preliminary search for lexical items such as please, sorry, of course, or grammatical structures such as comment clauses or parenthetical clauses. An auditory analysis is then carried out by close listening to each of the selected tokens3. Subsequently, those examples that are of high-enough sound quality4 are analysed instrumentally. Future research into the sociopragmatics of conversational interaction will need to take account not only of prosodic information but also body-language, including gaze, proximity and gesture. Multimodal corpora are currently being compiled with the chief aim of modelling human behaviour for speech recognition (see e.g. Martin et al. 2008) but will also provide important insights for those less interested in technological applications and more in the nature of human communication in itself.
2.
Prosody in a social context – speaker-related phenomena
Some prosodic features of the human voice are of course physiologically determined. Women generally have higher voices than men, and there are known effects of aging on the voice (e.g. Zellner Keller 2006, Nishio and Niimi 2008, Reubold et al. 2010), of mental state (e.g. Bard et al. 1996) and of physical condition (e.g. Braun and Künzel 2003). Other features are geographically or ethnically determined – different accents of English are often characterised not only by segmental differences but also by their melody (e.g. Fletcher and Harrington 2001, Grabe 2004). Even relatively stable characteristics, however, can be subject to cultural influence, and to conscious or unconscious adaptation in order to convey specific meaning. There are, for example, recognisable “speaking styles” (see section 2.3. below) typical of certain situational contexts (Wichmann 1996). The way we speak in casual conversation, for example, is generally different from the way we speak in a court of law or when reading poetry aloud. A subject of particular relevance for the pragmatic effects of prosody is the way that some variation reflects, or constructs, the power relationship between participants (e.g. Wennerstrom 2001, 192–198, Roth and Tobin 2010). However, these studies are among the few that recognise the significance of sequential information – the meaning generated by the prosodic relationship of one utterance to another (both within and across speaker turns) rather than by anything inherent in an individual utterance. This will be discussed further in sections 3.2 and 3.2. of this chapter. The social variables considered in the following sections are of course not exhaustive, but have been selected on the basis of available prosodic research and the variables addressed by individual studies.
Prosody and pragmatic effects
2.1.
187
Age, gender and sexual orientation
An intonation feature that has received much attention of late is the use in some varieties of English of so-called “uptalk”. This is the tendency for especially younger speakers, and according to some studies, female speakers, to mark statements with a final rising contour so that they sound like questions (1). (1) A: What are you doing at the weekend? B: We’re going to Dorset? (after Cruttenden 1997 [1986]: 129) This high rising tone (sometimes referred to as the high rising terminal – HRT) has received so much attention (e.g. Warren and Daly 2000), possibly because it is one of those language phenomena that causes older people to complain bitterly and to express their displeasure in letters to newspapers. It is frequently “deplored”, which is generally a good cue to language change in progress, and also a cue to the current limits of its accepted usage. According to Cruttenden (1997 [1986]: 129–130) it was first reported in Australia, then New Zealand and North America. Studies suggested that it was used more often by female speakers (e.g. Warren and Daly 2000). Cruttenden (1997 [1986]: 130) reports that it had spread to Britain by 1995 but was mainly restricted to London. It is now ubiquitous in the UK but seems to be still a feature more typical of young female speakers. It is also associated more with the middle class than the working class, as was reported for Australia and New Zealand. Some suggestions that the rise tone was realised differently if associated with a declarative, in other words different from an ordinary question, were not confirmed in a study of British English by Shobbrook and House (2003). Other gender-specific usage of individual phonological choices is hard to find. According to Brend (1975, cited in Smith 1979: 125) women use “a ‘polite’ pattern of assertive intonation … while men use a more ‘deliberate’ pattern.” To judge from the transcription in the example used to illustrate these patterns (yes I know), it seems that the “polite” fall is a high fall, while the “deliberate” fall is a low fall.5 Why one is “polite”, and the other “deliberate” is not immediately obvious, nor is it explained. It could be linked to the difference in perceived finality between a high and low fall: a high fall has been found to signal “more to come”, and a low fall greater finality (cf. Wichmann 2000). The latter could thus signal “the last word” and appear to close off the interaction. Leaving the interaction open, on the other hand, could easily be heard as showing concern for the interlocutor and hence politeness. Clearly a high falling nucleus is not inherently “polite”, but, as with most such attitudinal labels, becomes so only in context. There is a view, however, that there is a certain non-arbitrariness in prosody, even at the phonological level. One source of the attitudes described above, therefore, may also be a degree of iconicity, or sound symbolism. A very obvious difference between male and female speech is the result of a biological difference in
188
Anne Wichmann
size. The largest members of the group are usually male; women are generally smaller, and children are the smallest. The difference in size determines the size of the larynx, which in turn determines the pitch of the voice. The average pitch for women is around 200 Hz and for men around 100–150 Hz (with a tendency to rise slightly in old age) (e.g. Helfrich 1979, Beck 1997, Daly and Warren 2001). In human communication, there is a perception of low pitch as more dominant and assertive, and high pitch as submissive or lacking in confidence. This was confirmed by an experiment carried out by Ohala (1982) in which speech was digitally manipulated to control all variables except for pitch height. Low pitch was judged to be more dominant. The meanings that are conveyed by these differences in pitch have been explained in terms of sound symbolism. “A higher pitch is associated with a smaller sound-producing object, one that is not a threat to the members of the same or other species and is therefore entitled to claim protection”, while the male “advertises his role of potential aggressor by means of a vocal mechanism that exaggerates the low frequencies” (Bolinger 1989: 21) According to Ohala, high pitch might even be “a form of infant mimicry” which might be effective in “pacifying a would-be aggressor, since … natural selection has left most species with a very strong inhibition against harming conspecific infants” (1994: 330). Ohala calls this sound symbolic motivation for F0 variation the “frequency code”, which he believes is innate. The notion of biological “codes” has been taken up and extended by Gussenhoven (e.g. 2004), who argues that the frequency code, and other related codes, operate not only at a paralinguistic level but have also become grammaticalised in intonational phonology. He uses this to explain near universal features such as low endpoints relating to statements and high endpoints relating to questions, although as we will see in section 2.2, there are variants of British English intonation that do not conform to this apparent universal. Ohala and others (e.g. Smith 1979, Bolinger 1989) do not deny that some of the observed differences in humans may be learned; although the basic difference is biological, “anatomy does not account for all of this difference” (Smith 1979: 123). Ohala suggests that “speakers are aware of the sex-determined differences in speech and that they may choose to emphasize their masculinity or femininity by producing speech which exaggerates these differences” (1994: 344 n.4), and Bolinger, too, observes that “both sexes cultivate the difference” (1989: 22). How they cultivate it, however, depends on the social environment. In some situations, for example, women try to sound “small” and vulnerable, possibly child-like, while in others they make an effort to sound “larger” and therefore more powerful. A wellknown study (Beattie et al. 1982) reports that Margaret Thatcher, a former British Prime minister, was trained to speak at a lower pitch in order to sound more authoritative. This had the unfortunate side effect that her low pitch was sometimes mistaken for a cue to the end of her turn at speaking (also generally low in pitch), and she was frequently interrupted.
Prosody and pragmatic effects
189
A study by Van Bezooijen (1995) suggests that pitch differences may be differently constructed across cultures. She found that the average pitch of adult female speakers varied across different countries; in those countries where a more childlike demeanour is seen to be more “feminine”, it seems that women have a higher average pitch than in countries with greater gender equality. As Giles et al. (1979) put it, some of our speech markers are evolutionarily inherited and biologically determined; in other words intrinsically derived. … However, the origins of many speech markers are best represented by an interaction of biological (intrinsic), and extrinsic factors at the psychological and sociocultural levels. … for instance, it would seem reasonable to suppose that in most cultures, women … are expected to speak in particular ways (1979: 364).
Some acculturation can occur at an early age. For example, children of the age of 10 have been found to vary in their use of pitch in gender-specific ways, despite the fact that physiological differences have not yet emerged (Whiteside and Hodgson 1998), while Ainsworth 1994 (cited in Daly and Warren 2001) found gender-specific differences in the use of intonation patterns among children as young as four. As with much prosody research however, methods and scales of measurement vary, and results are not necessarily reliable or comparable. Different speaking tasks can lead to different results, and some discrepancies between studies may be due to the sex and age of the interviewer, with children adapting their voices to match that of the interlocutor. As I will discuss in sections 3.2 and 3.3 below, such pitch convergence, or pitch accommodation, is a phenomenon that can affect the results of these studies and has a large part to play in communication. In addition to overall pitch height, it has long been maintained that women exhibit greater pitch variability or “dynamism” (rate of change) than men (Smith 1979). Whatever the reason for this, it contributes to how men and women are judged. A dynamic use of pitch is characteristic of the expression of strong emotion, and women may thus be judged to be intrinsically more “emotional”. Alternatively, it may indicate that in some cultures women are socially conditioned to display emotion, while men are conditioned to conceal it. McConnell-Ginet, taking a strong feminist standpoint, argues that this is the source of the idea that women are deemed to be less stable than men: “masculine speech melodies (are heard as) metaphors for control, for “coolness”, and feminine melodies as uncontrolled, untamed by culture” (1983: 82). Ultimately, she argues, the problem lies in society: “women’s tunes probably can be interpreted to keep her in her place: on her back and out of power. But views of women’s intonational styles as uncontrolled (uncontrollable) and ineffectual (lacking in authority) can be challenged once the androcentric origins of these views are clearly understood” (1983: 83). From an ethological perspective this view is not entirely consistent. Animals choose to exaggerate their smallness or largeness for strategic reasons, usually to avoid confrontation that could be harmful, and research suggests that gender-specific behav-
190
Anne Wichmann
iour may also be (unconsciously) strategic. The feminist view seems to be that women have a naturally greater pitch dynamism but that it is falsely interpreted as a weakness, while the ethological view would suggest that this is a strategic display of “weakness”, and thus must be of some advantage in a given social or cultural context, just as a female politician’s strategic display of “strength” (lowered pitch) is presumably an advantage in the context in which she operates. Closely related to the study of prosody in relation to gender is that related to sexual orientation, since it essentially looks for traces of “feminine” speech in gay men and “masculine” speech in gay women. There have been few studies of lesbian prosody. These include Moonwomon-Baird (1997) and Lutzross (2010).6 Moonwomon-Baird failed to find any clear prosodic characteristics of lesbian speech, while Lutzross has identified certain stereotypical features, mainly “masculine” low pitch and narrow pitch range. However, an experiment showed that “these perceptions in no way correspond with the actual orientation of the speaker” (2010: 4). There have been more studies of the prosodic correlates of homosexual orientation in men, e.g. Gaudio (1994), Smyth et al. (2003), Levon (2006) and Podesva (2007). Not surprisingly, given what is reported as being characteristic of women’s speech, the use by men of wider pitch range is judged effeminate (Terango 1966). As with other attempts to relate prosody and gender, however, these studies are problematic in assuming stable categories. Podesva et al. (2001) claim that the search for a “gay” speaking style “limits style to a single dimension and does not explain how situational and interactional factors, such as social power, mode of interaction, topic, setting … contribute to intraspeaker variation” (2001: 178). Contrary, perhaps, to some expectations, Podesva (2007) finds that F0 does not correlate with gay orientation per se. He suggests, in his study of one gay man’s use of falsetto, that high F0 is associated with “flamboyance” (in itself a label open to interpretation), and that this in turn can be a way of constructing a gay identity. Most importantly, gay men may, like any other speakers, choose to display different personae in different situations. His subject, a doctor, when spending time with friends, “sometimes distinguishes himself as a flamboyant diva, a persona markedly different from the personae he takes on in other situations, such as the phone conversation (with his father) and the meeting with his patient” (2007: 492). The approach in this study “treats gay identity not as directly observable in the speech stream, but rather as emergent in the ways that phonetic features are employed in particular interactional contexts” (2007: 497). Much of the research reported in this section was carried out, at least implicitly, within a sociolinguistic variationist framework, treating social variables as relatively fixed. As we see, there are nonetheless signs of work that looks at the way in which speakers adapt to, or co-construct their communicative context by means of prosody. In other words, we see the beginnings of a sociopragmatic approach that focuses on how speakers adjust their speech strategies according to the identities or relationships they wish to convey.
Prosody and pragmatic effects
2.2.
191
Comparative intonation – regional and national differences
Early work on accent differences, e.g. in the UK, was mainly segmental. Even now, with the wide availability of instrumental techniques for analysis, most work is segmental or subsegmental7. An early study of Liverpool English (known locally as “Scouse”) (Knowles 1978) was the first to discuss intonation as part of the description of the accent, but it is only recently that intonational aspects of some UK accents have been systematically studied. An important contribution to this was the IViE project (Grabe et al. 2001), based on a corpus of speech from nine different areas in the British Isles, including Belfast, Bradford (bilingual Punjabi/English speakers), Cambridge, Cardiff (bilingual Welsh-English speakers), Dublin, Liverpool, Leeds, London and Newcastle. There have also been studies of Manchester intonation (Cruttenden 2001) and of Tyneside intonation (Local 1986). Even these few studies however are hard to compare because they use different models of intonation as a basis. The IViE project, heavily influenced by work in the AM framework, uses a tiered transcription system in the ToBI8 tradition, while Knowles (1978), Local (1986) and Cruttenden (2001) are influenced by the British system of holistic nuclear contours (fall, rise etc). The studies are for the most part focussed on the intonation contours associated with certain speech acts (question, statement etc.), or, in the case of Local’s study, on conversational functions such as “understanding checks” (1986: 193). These are echo questions, i.e. cases when one speaker repeats the other speaker’s word or phrase in order to query it (e.g. A: I’m hungry. B: Hungry?). Local claims that in Tyneside speech these are realised with falling contours rather that with the rise common in RP. The most marked feature of some non-standard British accents is the use of an apparently rising contour in utterance-final position (for example in Belfast English) where the standard accent would use a fall. The contours identified by Grabe and Post (2002) include the truncated rise, the low rise, the rise-plateau fall and the fall-rise (see Figure 2). The Manchester intonation as described by Cruttenden (2001), also differs from the standard – its equivalent to a final fall in RP is a “rise slump”, which is a rising contour followed by “a fall from a relatively high pitch to a relatively mid pitch” (2001: 57). The authors of both studies observe that the ToBI inventory of tones is inadequate for this kind of variation, especially for indicating pitch targets that are neither high nor low, but mid-range. Thus the transcriptions in Figure 2 are to some extent ad hoc solutions. It would be beyond the scope of this chapter to discuss the theoretical issues raised here, but it is important to note that the fundamental ToBI constraint of having only two tones, H and L, while convenient in its simplicity, does not capture some important contour differences that are so typical of regional varieties of English. Grabe and Post’s (2002) study compares the contours for declaratives across different varieties. They find that the falling tone (H*L) in the standard variety contrasts with a very common rise L*H for the Belfast variety. The contours as-
192
Anne Wichmann
Figure 2. Stylised representations of pitch contours in different varieties of English. Contours include those referred to by Grabe and Post (2002) and Cruttenden (2001)9.
sociated with yes-no questions (the authors refer to these as “inversion questions”) include the fall, the fall-rise and the low rise for the Cambridge variety, while the Belfast accent uses almost exclusively a truncated rise. With the spread of English as a lingua franca and the increased awareness that prosody must be a part of pronunciation teaching, there has been some study of non-native prosody (Trouvain and Gut 2007). While foreign accents in general may lead to misunderstandings, those caused by inappropriate intonation, as opposed to segmental differences, often give rise to unintended pragmatic effects. How do these errors arise? According to Mennen “it is important for teaching purposes to distinguish between phonological and phonetic errors, so that the source of the problem can be addressed in teaching” (2007: 71). Some non-native effects may be phonetic implementation errors rather than phonological errors. In other words, the correct phonological choice is made but the way it is realized phonetically (“implemented”) creates an unintended effect. For example, the percept of an incorrect choice of contour (e.g. rise instead of fall) may actually be the result of different alignment of the contour in relation to the associated syllable. Similarly, “incorrect” prominence placement may be the result of using different prominence
Prosody and pragmatic effects
193
cues …” (Learners) may not be producing stress with the same cues as native speakers do” (2007: 63). Both of these are implementation errors, i.e. mistakes in how a phonological category is realised phonetically, rather than phonological errors. Some misunderstandings may lead to incorrect affective or personality judgements, especially if they derive from the cultural differences in the use of pitch height and pitch range. Such differences have been described above in relation to gender differences, but overall differences between languages have also been observed (e.g. Braun 1994, Gfroerer and Wagner 1995, reported in Mennen 2007). These studies have generally been motivated by forensic applications such as voice identification, but also have relevance for L2 acquisition, language teaching and cross-cultural pragmatics. As Mennen points out, “we derive much of our impression of a speaker’s attitude and disposition towards us from the way they use intonation in speech” (2007: 54). In particular, differences in pitch range may give rise to unintended impressions. For speakers of languages with a narrow pitch range, such as German, English is perceived as exaggerated, while English speakers may consider a narrower range as indicating lack of rapport (Grabe 1998 reporting Trim 1988). Comparative studies of intonation are therefore clearly important for the teaching of languages, but can also provide an insight into pragmatic miscommunication that can have far-reaching effects. 2.3.
Speaking styles
In prosodic research, the notion of “speaking style” is ill-defined and has been used to mean many different things. Early interest in stylistic variation was generally focused on very marked activity types – performing the liturgy, scripted public oratory and unscripted broadcast commentary (Crystal 1969, Crystal and Davy 1969). We also find studies of poetry reading (Wichmann 1996, Tsur 2006), horse racing commentary (Trouvain and Barry 2000), railway station announcements and street vendors’ calls (Zollna 2003), race-calling and cattle auctions (Kuiper 2004). Apart from these isolated genre studies, the majority of recent work claiming to examine prosodic style has in fact been restricted to the mode of speaking rather than other contextual parameters. In the main, it focuses on the distinction between reading aloud and speaking spontaneously, or occasionally on the gradient distinction between casual and careful speech (see Hirschberg 2000). These studies are typically motivated by applications in speech technology, including the need to optimise automatic speech recognition, e.g. Weintraub et al. (1996), who compare the vocabulary recognition rate of their system under different conditions. Speech recognition has a number of applications, including human-machine interaction, and systems need to be robust enough to accommodate speech disfluencies, varying speech rates and contours that are not necessarily those of read-aloud laboratory speech on which many synthesis systems are based. In addition, when speakers are asked to repeat something the machine has not understood, they tend to switch to a
194
Anne Wichmann
more careful mode of speaking, and the system must adapt to this (Levow 1999). Of course, even such dimensions as read vs. spontaneous are in practice gradient rather than binary – there are degrees of preparation for both scripted and unscripted speech which influence performance (cf. Wichmann 2000: 19–22). A large scale corpus-based10 study of prosodic differences between read and spontaneous speech is reported by Hirschberg (2000). She examines disfluencies, speech rate, pause duration, choice of tonal contour and pitch range and their roles in distinguishing speech type. She found considerable differences in speech rate – read speech was generally faster than spontaneous speech, although it was not clear whether this was simply due to the presence of fewer pauses than in spontaneous speech or whether the articulation rate was also faster.11 Disfluencies, such as hesitations, filled and unfilled pauses and self repairs are common in spontaneous speech, but their absence is not necessarily an indication of read speech, and for automatic detection of spontaneous speech other cues are clearly necessary. Average F0 and maximum F0 were higher for read speech, which was also observed by Blaauw (1991) and Koopmans-van Beinum (1991) and is consistent with Eskenazi’s (1993) claim that F0 maxima are higher and F0 range greater in “clear” (or read) rather than “casual” speech. Finally, Hirschberg found that the tone choice typically associated with sentence type (declarative, wh-interrogative, yes-no interrogative) was more consistent in read speech, with spontaneous speech displaying more variability. For example, the association of falling nuclei on declaratives and rising nuclei on yes/no-questions was found to be much stronger in read than spontaneous speech. Wh-questions, with a final falling contour occurred 84 per cent of the time in read speech but only 62 per cent in spontaneous speech. As with studies of other aspects of variation, comparability of results is not guaranteed. A study by Llisteri (1992) of the papers submitted to a special workshop on speaking styles (1991 Barcelona) reveals a lack of consistency in the type of data analysed, the approaches taken and the labels used. Even within the fairly narrow dimension of comparison, i.e. read vs. spontaneous, Llisteri observes that “what is labelled as ‘spontaneous speech’ has been in most cases obtained in a quite constrained situation, i.e. in a laboratory, the speaker being taken out of his natural context to produce speech samples for an experimental study” (1992: 3). Just as with the study of the prosodic correlates of social categories, a uni-dimensional approach to speaking styles has been criticised. Podesva et al. (2001) claim that “this methodology routinely elicits pre-defined ‘styles’ from speakers (such as Casual Style, Reading Style, etc.), … limits style to a single dimension and does not explain how situational and interactional factors, such as social power, mode of interaction, topic, setting … contribute to intraspeaker variation” (2001: 178; my emphases).
Prosody and pragmatic effects
3.
195
Social distance and power relations
Some of the studies reported so far have taken a variationist approach, for example by examining correlations between prosodic features and social categories such as age, sex and region, or contextual constraints such as activity type or speech mode. In this section I will examine studies of more fluid contexts, such as the social distance between speakers, the power relationship between speakers, and the way in which prosody reflects and constructs these relationships in the course of the interaction. First I will describe a study (Wichmann 2004) that attempts to correlate prosodic choices (choice of final contour) with power relations between participants. I will then move on to consider the socio-pragmatic meanings generated not by the features of an individual utterance but by the sequential relationship between utterances, including those of different speakers. This follows the view underlying Conversation Analysis that the sequential organisation of actions is the relevant unit of analysis when examining the way in which actions “produce and reproduce social life (e.g., Goffman 1983; Goodwin and Goodwin 2000)” (Roth and Tobin 2010; cf. also Svennevig and Johansen, this volume). 3.1.
Tone choice and power relationships in please-requests.
Wichmann’s (2004) study of please-requests in the ICE GB corpus (Nelson et al. 2002), compares the intonation contours used in “private” and “public” situations. Private refers to situations where social distance is small and the relationship between participants is symmetrical. Public, on the other hand, refers here to situations where social distance is greater and the power relationhsip asymmetrical. Please-requests spoken in private situations tend to be realized with a rising tone on an utterance-final please, e.g. Could you open the \ door /please while a similar request spoken by the more powerful of participants is more likely to be realized with a final falling contour of which please constitutes the low nuclear “tail”, e.g. Could you open the \ door please. The latter, while maintaining an element of courtesy, has more affinity prosodically with a command, while the former is more like a question. This clearly reflects the relative rights and obligations of the participants; an equal has the right to turn down a request, while a subordinate does not. In the first case, the high endpoint of the rising contour is associated with an “open” meaning, or if we return to the ethological perspective, with “submissiveness” and “smallness”, while the low endpoint of the falling contour is associated with “closure” or “authority”. What is not possible to say at this point is the extent to which the choice of tone is the product of the speaker-hearer relationship or actually constructs it. According to Podesva et al. “as with any correlational approach, it does not reveal whether categories shape linguistic practice or are themselves derivative of language use. It
196
Anne Wichmann
also assumes the stability of both style and social categories” (2001: 178). It seems highly likely that prosody can play at least some part in creating, manipulating or controlling context, and some further evidence for this will be presented in section 3.3 below. However, in order to do this, it is first necessary to give a brief account of an aspect of prosody that has been paid little attention hitherto but is crucial for what follows. 3.2.
Prosodic matching – a sequential resource for intonational meaning
Sequential relationships have already been found important for the phonetic analysis of speech within turns. Local (1992), for example, uses sequential patterns to identify whether a turn at speaking is new or the continuation of a previous turn. Walker (2004) takes a similar look at “increments” – these are utterance continuations that can occur after a potential (syntactic, prosodic, pragmatic) completion point, as in, for example “English people don’t go to Germany on holiday [pause] generally”. They can also occur after another speaker turn, as in (2). (2) A: I’m working on Thursday. B: Oh yes. A: all day. S The question Walker asks is whether the “all day” in this exchange is a new turn or a continuation of (incremental to) the speaker’s previous turn. With these and similar examples, Walker shows that a stretch of talk cannot be recognised as an increment by any inherent prosodic characteristics but only by the relationship it bears to the preceding talk of the same speaker. There is not a paradigmatic phonetics of increments: some piece of grammatically incomplete talk cannot be recognised as an increment when examined out of context, away from its host, by virtue of its phonetic constitution. Rather, in the phonetic details of the increment there are the phonetic exponents of a syntagmatic relationship with its host. (Walker 2004: 166).
Sequential organisation has also been shown to be important between the turns of different speakers. This is the case, for example, in relation to rhythm or timing, such as with response particles, also known as backchannels or “continuers” (Schegloff 1982: 81) such as m-hm, uh huh, right, yes and ok. While these responses are generally assumed to be supportive (hence “continuer”), this is so only if they fit rhythmically with the speech of the other speaker. If the prevailing rhythm is disturbed, they can also subtly indicate the opposite, as described in Müller (1996), who notes that the absence of rhythmic integration is a sign of “disaffiliation” i.e. backchannels that are not fully supportive. We see, then, that sequential patterning – how an utterance relates prosodically to that of another speaker – is important, and can reveal something about the inter-
Prosody and pragmatic effects
197
personal relationship between speakers, such as the degree of co-operativeness or compliance. This has most commonly been observed in relation to pitch, and is variously referred to as pitch concord, pitch matching, register matching or pitch accommodation. Brazil (1985) observed that a feature of compliant behaviour was what he called “pitch concord”. This is expressed in relation to his notion of “key”. He suggests that there are three significant pitch levels, high medium and low and that “(t)he key of ‘termination’ (the starting point of a (falling) nuclear tone) puts constraints on the key of the next turn. In compliant behaviour, Brazil assumes a pressure towards pitch concord, at least after mid and high terminations” (Wichmann 2000: 141). This means that a turn that ends in a mid fall (i.e. a fall starting at a point in the middle of the speaker’s range) favours a new turn starting also in the speaker’s middle range. Since each person’s pitch range is different, the concord is assumed to be relative; that is, the second speaker adopts the same pitch relative to his or her own range. While Brazil’s notion of pitch concord is defined in relation to categorical pitch height differences (high, mid and low “key”), other researchers have looked at the phenomenon of pitch matching from a more gradient perspective. In other words, they have simply observed whether or not two speakers use a similar pitch height, or register (again relative to each speaker’s own range), in consecutive utterances, where “register” generally refers to the position of an utterance in the speaker’s voice range. Couper-Kuhlen (1996), for example, looks at register matching during lexical repetition in a radio phone-in programme. Callers are asked to guess the answer to a riddle, and the radio host frequently repeats the answer before saying whether it is right or wrong, as in (3). (3) Host: it is complete though it seems it isn’t (.) what d’you reckon Caller: heart Host: heart S Caller: yeah Host: er no (adapted from Couper-Kuhlen 1996: 374) The lexical repetition is usually realised in the same (relative) pitch range as that of the caller. However, Couper-Kuhlen shows that if the radio host attempts to mimic the caller’s pitch in absolute rather than relative terms, the effect is no longer a positive one of compliance but a negative one of mimicry or mockery. Schegloff (1998), like Couper-Kuhlen, sees register matching as a gradient phenomenon. He describes a telephone opening sequence in which one participant initially uses a high pitch in greeting “Hi”, but then adjusts to the lower pitch of the caller. Schegloff compares “the “enthusiasm” of the greeting with the “restraint” of the other speaker, and suggests that “the two young women have thus taken up different stances toward the occasion-so-far” (1998: 246) (see Figure 3). Wennerstrom (2001) discusses this same sequence, and argues that this is evidence of a negotiation of power relationships between participants. Pitch concord, she claims is
198
Anne Wichmann
Figure 3. The beginning of a telephone conversation. From Schegloff (1998: 244). Copyright © 1998 by Kingston Press Services Ltd. Reproduced by permission of SAGE.
“a valuable tool to study power and solidarity moves in conversation” (2001: 185; see also section 3.3). Theoretical support for the concept of register matching can be found in Communication Accommodation Theory, which provides “a wide-ranging framework aimed at predicting and explaining many of the adjustments individuals make to create, maintain, or decrease social distance in interaction” (Giles et al. 2006: 10). These adjustments may include acoustic and kinesic convergence, noted for example by Gregory et al. (1993). A study of segmental accommodation (in her terms “entrainment”) between speakers (increased similarity in pronunciation) by Pardo (2006) suggests similarly that it plays a part in achieving mutual goals, in other words segmental accommodation has a social function, and that previous explanations in terms of, for example, automatic priming, are inadequate. The same is likely to be true of prosodic accommodation, which has been less widely studied. A speaker who exhibits pitch concord is in effect accommodating to the previous speaker, and there is evidence that the accommodation of pitch (and also of other prosodic features) occurs not only at certain points in the interaction – beginnings of turns or short responses, but over longer stretches of interaction. Kousidis et al. (2009) (see Figure 4) observed that over the course of a conversation, two speakers broadly accommodated to each other’s pitch register. The
Prosody and pragmatic effects
199
Figure 4. Time plot pitch for a conversation between two speakers. Pitch points are 10 seconds apart and normalized for the speakers’ voices. From Kousidis et al. (2009: 5). Reproduced by kind permission of the authors.
various fluctuations of pitch height are fairly closely aligned; when one raises his voice the other does too. This study was carried out under laboratory conditions, but similar behaviour has been observed in uncontrolled classroom situations. An important study by Roth and Tobin (2009) observes the prosodic, gestural and language patterns used in classrooms in the interaction between teachers, co-teachers and students. They found that “successful and unsuccessful lessons – as indicated by outcome measures such as quizzes and tests – are associated with the production and reproduction of prosodic alignment and misalignment” (Roth and Tobin 2009: 808). In particular, they note that “in classes and teacher arrangements where we observe alignment in prosody, participants report feeling a sense of solidarity, whereas where the prosodies of speakers are misaligned, participants report the presence of conflict” (Roth and Tobin 2009: 808). They conclude that “prosodic alignment and misalignment are resources that are pragmatically deployed to manage face-toface interactions that have solidarity and conflict as their longer-term outcomes” (Roth and Tobin 2009: 807). 3.3.
The role of prosodic accommodation in managing conflict
Roth and Tobin’s (2009) study has implications for a number of situations where there is a need to negotiate relationships, in particular where there are asymmetries
200
Anne Wichmann
of power, as suggested by Wennerstrom (2001). The authors see their work as “contributing descriptions of prosody as a transactional resource that is deployed pragmatically (rather than deterministically and causally) and subject to the purposes at hand” (Roth and Tobin 2009: 842). In other words they take the view that linguistic behaviour is not necessarily determined by the context of situation but is deployed to create it. They suggest for example “that in the service professions (teaching, healthcare), providers may increase their effectiveness when they attune themselves to the recipients of the service, who, having the sense that the service provider really listens (exhibiting empathy or solidarity), better respond to the service” (Roth and Tobin 2009: 828). People who are severely depressed, for example, generally speak with a low and narrow pitch range. Police negotiators in the UK, when being trained in suicide intervention techniques, are encouraged to “lower their tone whenever possible if they are trying to portray a caring emotion in their voice whilst trying to build up rapport and trust” (Detective Chief Superintendent Ian Kennedy, West Yorkshire Police, personal communication). Such an alignment, or display of “solidarity” through pitch matching is, however, not always appropriate in asymmetrical situations. Roth and Tobin observed that in some cases it was important for the teachers in their study to defuse conflict by lowering their voices, rather than raising them in alignment with a student. Since, as we have seen in the discussion of gender differences in pitch behaviour, high pitch tends to be associated with loss of control and weakness, it is often suggested that authority figures (teachers, police) should indicate their control of a situation by maintaining a low pitch. In the following, I will examine some previous research on the prosody of conflict and close by presenting unpublished research on prosodic accommodation, and failure to accommodate, in conflict situations. Within the framework of linguistic impoliteness, especially in relation to “face”, a study by Culpeper et al. (2003) examines communicative strategies designed to attack face and cause social conflict. The data used is from a BBC documentary television series The Clampers (1998) containing recordings of disputes between traffic wardens and car owners, between bailiffs and those who have defaulted on parking fines, and between officers and appellants at appeal hearings against parking fines. The study is of the linguistic strategies used on both sides in these disputes, and includes an analysis of prosodic features of impoliteness. These include the exploitation of global phonetic resources, such as shouting (“invading auditory space”) and also the absence of pitch accommodation, in other words maintaining a pitch level that is clearly divergent from that of the other speaker. This can mean, for example, keeping a consistently low pitch even when the other speaker has raised his or her voice. The strategic denial of pitch accommodation appears to be a means of increasing the distance between the interlocutors. In order to further investigate this strategy of pitch (non-)accommodation, sections of this data were revisited for instrumental analysis. In an unpublished pilot study, Ellison (2008) recorded the pitch behaviour of participants in a number of
Prosody and pragmatic effects
201
Figure 5. F0 accommodation followed by non-accommodation in a dispute. From Ellison 200813. Figures 5–7 are reproduced with kind permission of the author.14
scenes12. In Figure 5 we find an instrumental analysis of an exchange previously discussed by Culpeper et al. (2003). It is an example of conflict at an appeals tribunal, between the presiding officer and a couple who are appealing against what they perceive to be an unfair parking fine. Until about 45 seconds into the clip, the husband and the officer are using a similar low pitch and narrow range. When it becomes clear that their appeal will be rejected, both husband and wife become angry and raise their voices. The tribunal officer, however, (designated “lawyer” in the figure) maintains broadly the same low pitch and narrow range. This lack of accommodation is a powerful act, and the tribunal officer’s intention not to engage in further discussion, and his right to control the proceedings, is made explicit in the latter part of the dialogue, reproduced in (4). S1 is the tribunal officer. S2 and S3 are husband and wife respectively. (The transcript begins approximately 50 seconds into the clip shown in Fig 5, as indicated by accompanying numbers.) (4) (50) S1: a I’m sorry Mr S2: happens if a sign changes over night what is then the situation S3: (55) S1: Coup I’m I’m not here to um discuss the a a S2: no no what is the situation what us our S3 (60) S1: ah well I have I’m finished S2: situation I’m not finished yet as far am I’m concerned S3 S1: and I’m running this tribunal and that the end of it
202
Anne Wichmann
Figure 6. Non-accommodation of F0 in a dispute. From Ellison 2008.
The same pattern of non-accommodation is observable in a dispute between a traffic warden and an angry driver. The warden (S1) is confronted by a market trader (S2) after his van has been immobilised for being parked in a loading area for longer than the allotted twenty minute time limit. The pitch pattern alone is sufficient (Figure 6) to show the way in which the traffic warden attempts to maintain control. The angry trader uses a relatively high pitch (over 400 Hz is high for a man) and wide fluctuations from high to low. The warden, on the other hand, maintains a low pitch and a narrow range. His pitch is mainly around 200 Hz, only dropping occasionally to near 100 Hz. However, not all moments of conflict are marked by divergent pitch register. One exchange (5) is an altercation between the traffic warden (S1) and his driver (S2). Despite the fact that the traffic warden has seniority over the driver, the relationship is a fairly symmetrical one as they are colleagues. The traffic warden has received an alert that a colleague may be in danger. They are travelling together in the clamping van and are attempting to reach the given address. However, as the driver has forgotten the A-Z (London street map) they are lost, and tempers are rising, evident not least in the expletives used. (5) (0) S1: We’re having a problem at the moment we’re going to a code red and we S2: (5) S1: don’t have an Ay to Zed and we don’t have the location cos the driver’s not S2: S1: brought his Ay to Zed today have you the one that I gave him15 what do you S2: (10) S1: think I bought it for and gave it to you Miguel S2: listen I told you because I forgot
Prosody and pragmatic effects
203
(15) S1: well don’t forget it’s mine I bought S2: the fucking Ay to Zed man . for fuck’s sake (20) S1: it so that you could get yourself familiar with the S2: what did you give it me then (25) S1: area cos you I bought it to help you well it’s not S2: I don’t know I don’t know the bloody road that one as well S1: gonna help you sitting in the house S2:
just because I forgot the fucking Ay to Zed (30) S1: druid street it cant be this way this is long lane S2: it was in another street sorry (35) S1: we will have to turn round and go back S2: I know
During the argument, voices gradually rise in tandem to a peak around 17 seconds into the episode and fall together as agreement is reached (see Figure 7). There is no evidence here that one participant is trying to maintain control over the other, and there seems to be an element of co-operativeness (in the Gricean sense) in their willingness to argue with each other. Conflict tends to arouse strong emotions and these are often reflected in the voice. The prosodic correlates of emotion are not easy to establish, but a fairly uncontroversial observation is that “active” emotions (cf. Cowie et al. 2000) whether positive or negative, (i.e. very happy or very angry) cause speech to be louder and higher in pitch than neutral, unemotional speech or “passive” emotions (contentment, sadness). In conflict talk between participants with a symmetrical power relationship, we may find that despite the conflict there is an underlying co-operativeness, in that participants raise their voices in equal measure. Where the power relationship is not symmetrical, however, we can expect a different effect on prosody. As Roth and Tobin (2010) observed, and as Ellison’s (2008) data shows, the more powerful participant (police, parent, teacher etc.) often wishes to appear to be in control, and this is generally reflected in a low pitch; if the less powerful participant is raising his/her voice in anger, the result is a prosodic misalignment. While it may be important for figures of authority to maintain the appearance of control (and hence power), the problem is that they cannot do this and at the same time display solidarity with the other, thus potentially exacerbating the conflict.
204
Anne Wichmann
Figure 7. A “co-operative” dispute during which the F0 of both participants rises and falls.
4.
Conclusion
The socio-pragmatic effects of prosody have only just begun to be widely explored. As I have shown, prosody has the potential to convey meaning both through phonological choice, e.g. of rising or falling tone, and also by exploiting the more gradient features including pitch range and overall height, such as a higher voice to sound more vulnerable or a lower voice to sound more authoritative. Most importantly, in my view, and so far only sporadically investigated, is the role of prosody in the well-known phenomenon of social accommodation and its potential to be exploited strategically. Research described in this chapter suggests that a speaker can express stance (power or solidarity) by means of accommodating, or not, to the prosody of an interlocutor. Studies must therefore consider not only the possible prosodic correlates of social context within individual utterances, but pay more attention to the syntagmatic relationship across utterances. The cooperative principle in interaction has been well-attested (e.g. Clark 1996) but the role of prosody remains elusive.
Notes 1. Fundamental Frequency, measured in Hertz (Hz), is what we hear as pitch. It is displayed by speech analysis software on a screen as a (broken) line. Some breaks indicate pauses, but most occur at points where there is no voicing, such as fricatives (e.g. [s] [f]). The human ear compensates for these latter breaks and hears a continuous melody. 2. Each word is tagged according to its part of speech, i.e. noun, pronoun, adverbial etc., and the text is also parsed syntactically. For more details see Nelson et al. (2002). 3. The sound files are linked to the transcription and therefore directly accessible. 4. The disadvantage of recording casual conversation in its natural habitat is that extraneous noise sometimes makes instrumental analysis impossible. There is thus a trade-off be-
Prosody and pragmatic effects
5. 6. 7.
8.
9. 10.
11.
12. 13.
14. 15.
205
tween quality of sound and naturalness of speech. Laboratory studies, on the other hand, sometimes have to sacrifice the latter for the former. A high fall is one starting relatively high in the speaker’s range, while a low fall begins lower. Both are assumed to have the same low endpoint. See also footnote 8. I am very grateful to Robert Podesva and Paul Baker for these references. Segmental refers to discrete speech units, broadly speaking individual vowels and consonants, while sub-segmental refers to smaller acoustic components of these segments, such as the release phase of a plosive such as [t]. ToBI is an annotation system based on the AM model of intonation, and stands for Tones and Break Indices. The “tones” are the pitch targets H and L. The Break Indices indicate the strength of a prosodic break, with a silent pause being the strongest. The symbol !H*, as seen in Cruttenden’s transcription of the “rise slump”, represents a socalled downstepped pitch accent, a high pitch target that is lower than the preceding one. This study draws on three separate studies using the TIMIT database of read speech (Garofolo et al.1993), the ATIS0 (air travel information systems) corpus of spontaneous task-directed speech, and the DARPA Resource Management database (http://www.ldc. upenn.edu/Catalog). These corpora were all compiled in the U.S. for the development of speech recognition systems. Speech rate is the time taken, expressed in syllables per second, over an entire utterance or spoken text, including periods of silence. Articulation rate is the rate in syllables per second of vocalisation only, i.e. disregarding pauses. See also the discussion in Wichmann 2010. In Figures 5, 6 and 7 the graphs plot F0 samples approximately 0.5 seconds apart depending on speech rate. This gives a very rough impression of the F0 level for each speaker. No normalisation was carried out. Julie Ellison died in childbirth in Dec 2010. I pay tribute to her contribution. Up to this point the speaker is mainly addressing the camera. We cannot therefore exclude a certain amount of “performance” in this and other episodes.
References Aijmer, Karin 2002 English discourse particles: evidence from a corpus. Amsterdam: John Benjamins. Ainsworth, Helen 1994 The emergence of the high rising terminal contour in the speech of New Zealand children. Te Reo 37: 3–20. Astruc-Aguilera, Maria-Lluisa 2005 The intonation of extra-sentential elements in Catalan and English. Cambridge University, UK. PhD thesis. Bard, Ellen G., Catherine Sotillo, Anne H. Anderson, Henry S. Thompson and M.M. Taylor 1996 The DCIEM Map Task Corpus: spontaneous dialogue under sleep deprivation and drug treatment. Speech Communication 20(1–2): 71–84. Beattie, Geoffrey W., Anne Cutler and Mark Pearson 1982 Why is Mrs Thatcher interrupted so often? Nature 300: 744–747.
206
Anne Wichmann
Beck, Janet M. 1997 Organic variation of the vocal apparatus. In: William J. Hardcastle and John Laver (eds.), The Handbook of Phonetic Sciences, 256–297. Oxford: Blackwell. Blaauw, Eleonora 1991 Phonetic characteristics of spontaneous and read-aloud speech. In: Proceedings of the ESCA Workshop ‘Phonetics and Phonology of Speaking Styles: Reduction and Elaboration in Speech Communication’. Paper 012, 1–5. Barcelona, Catalonia, Spain, 30 September – 2 October. Boersma, Paul and David Weenink 2005 Praat: Doing Phonetics by Computer (version 4.3.04). [Computer program] http://www.praat.org (Accesssed 3 June 2010). Bolinger, Dwight 1989 Intonation and its Uses. London: Edward Arnold. Braun Angelika 1994 Sprechstimmlage und Muttersprache. Zeitschrift für Dialektologie und Linguistik LXI (2): 170–178. Braun, Angelika and Hermann J. Künzel 2003 The effect of alcohol on speech prosody. In: Proceedings, 15th International Congress of Phonetic Sciences, Barcelona, 3–9 August. Brazil, David 1985 Phonology: Intonation in Discourse. In: Teun Van Dijk (ed.), Handbook of Discourse analysis, Vol2 Dimensions of Discourse, 57–75. London: Academic Press. Brend, Ruth 1975 Male-female intonation patterns in American English. In: Barrie Thorne and Nancy Henley (eds.), Language and sex: difference and dominance, 84–87. Rowley, Mass. Clark, Herbert H. 1996 Using language. Cambridge: Cambridge University Press. Couper-Kuhlen, Elizabeth 1996 The prosody of repetition: on quoting and mimicry. In: Elizabeth CouperKuhlen and Magaret Selting (eds.), Prosody in Conversation, 366–405. Cambridge: Cambridge University Press. Couper-Kuhlen, Elizabeth and Cecilia E. Ford (eds.) 2004 Sound Patterns in Interaction. Amsterdam: Benjamins. Cowie, Roddy, E. Douglas-Cowie, Susan Savvidou, Edelle McMahon, Martin Sawey and Marc Schröder 2000 ‘FEELTRACE’: an instrument for recording perceived emotions in real time. In: Roddy Cowie, Ellen Douglas-Cowie and Marc Schröder (eds.), Proceedings, ISCA Workshop on Speech and Emotion, 19–24. Newcastle, N. Ireland 5–7 September. Cruttenden, Alan 1997[1986] Intonation. 2nd edn. Cambridge: Cambridge University Press. Cruttenden, Alan 2001 Mancunian Intonation and Intonational Representation. Phonetica 2001(58): 53–80.
Prosody and pragmatic effects
207
Crystal, David 1969 Prosodic Systems and Intonation in English. Cambridge: Cambridge University Press. Crystal, David and Derek Davy 1969 Investigating English Style. London: Longman. Culpeper, Jonathan 2005 Impoliteness and entertainment in the television quiz show: The Weakest Link. Journal of Politeness Research 1: 35–72. Culpeper, Jonathan, Derek Bousfield and Anne Wichmann 2003 Impoliteness revisited: with special reference to dynamic and prosodic aspects. Journal of Pragmatics 35: 1545–1579. Daly, Nicola and Paul Warren 2001 Pitching it differently in New Zealand English: Speaker sex and intonation patterns. Journal of Sociolinguistics 5(1): 85–96. DARPA Air Travel Information System (ATIS0). Spontaneous Speech Pilot Corpus and Relational Database. http://www.ldc.upenn.edu/Catalog/readme_files/atis/sspcrd. readme.html (Accessed 21. 5. 10) Dehé, Nicole and Anne Wichmann 2010 Sentence-initial I think (that) and I believe (that): Prosodic evidence for use as main clause, comment clause and discourse marker. Studies in Language 34(1): 36–74. Du Bois, John W., Wallace L. Chafe, Charles Meyer and Sandra A. Thompson 2000 Santa Barbara corpus of spoken American English, Part 1. Philadelphia: Linguistic Data Consortium. ISBN 1–58563–164–7. Ellison, Julie 2008 A study of the prosody of conflict. University of Central Lancashire, Preston, UK, undergraduate dissertation. Eskenazi, Maxine 1993 Trends in speaking styles research. In: Proceedings EUROSPEECH’93, 501–509. Berlin, Germany, 22–25 September. Fletcher, Janet and Jonathan Harrington 2001 High Rising Terminals and fall-rise tunes in Australian English. Phonetica 58: 215–229. Foulkes, Paul and Gerard Docherty 2006 The social life of phonetics and phonology. Journal of Phonetics 34(4): 409–438. Garofolo, John S., Lori F. Lamel, William M. Fisher, Jonathan G. Fiscus, David S. Pallett, Nancy L. Dahlgren and Victor Zue 1993 TIMIT Acoustic-Phonetic Continuous Speech Corpus. Philadelphia: Linguistic Data Consortium. Gaudio, Rudolf P. 1994 Sounding Gay: Pitch Properties in the Speech of Gay and Straight Men. American Speech 69(1): 30–57. Gfroerer, Stefan and Isolde Wagner 1995 Fundamental frequency in forensic speech samples. In: Angelika Braun and Jens-Peter Köster (eds.), Studies in Forensic Phonetics, 41–48. Trier, Wissenschaftlicher Verlag Trier.
208
Anne Wichmann
Giles, Howard, K. R. Scherer and D. M. Taylor 1979 Speech markers in social interaction. In: K. R. Scherer and H. Giles (eds.), Social markers in speech, 342–381. Cambridge: Cambridge University Press. Giles, Howard, Jennifer Fortman, René M. Dailey, Valerie Barker, Christopher Hajek, Michelle Chernikoff Anderson and Nicholas O. Rule 2006 Communication Accommodation: Law Enforcement and the Public. In: R.M. Dailey and B.A. Le Poire (eds.), Applied Research in Interpersonal Communication: Family Communication, Health Communication and Communicating Across Social Boundaries, 241–269. New York: Peter Lang. Goffman, Erving 1983 The Interaction Order. American Sociological Association, 1982 Presidential Address. American Sociological Review 48: 1–17. Goodwin, Marjorie H. and Charles Goodwin 2000 Emotion within Situated Activity. In: Alessandro Duranti (ed.), Linguistic Anthropology: A Reader, 239–257. Malden, Mass: Blackwell. Grabe, Esther 1998 Comparative Intonational Phonology: English and German. University of Nijmegen PhD thesis. Published in Max Planck Institute Series in Psycholinguistics. Grabe, Esther 2004 Intonational variation in urban dialects of English spoken in the British Isles. In: Peter Gilles and and Jörg Peters. (eds.), Regional Variation in Intonation, 9–31. Linguistische Arbeiten, Tuebingen: Niemeyer. Grabe, Esther, Brechtje Post and Francis Nolan 2001 The IViE Corpus. Department of Linguistics, University of Cambridge. http://www.phon.ox.ac.uk/IViE. (accessed 3 June 2010). Grabe, Esther and Brechtje Post 2002 Intonational Variation in English. In: Bernard Bel and I. Marlin (eds.), Proceedings of Speech Prosody, 343–346. Laboratoire Parole et Langage, Aix-enProvence, France. 11–13 April. Gregory, Stanford, Stephen Webster and Gang Huang 1993 Voice pitch and amplitude convergence as a metric of quality in dyadic interviews. Language and Communication 13: 195–217. Gussenhoven, Carlos 2004 The Phonology of Tone and Intonation. Cambridge: Cambridge University Press. Hedberg, Nancy, Juan M. Sosa and Lorna Fadden 2006 Tonal Constituents and Meanings of Yes-No Questions in American English. In: Rüdiger Hoffman and Hansjörg Mixdorff (eds.), Proceedings of Speech Prosody, 3rd International Conference, On CD-Rom. Dresden. Helfrich, Hede 1979 Age Markers in Speech. In: Klaus R. Scherer and Howard Giles (eds.), Social Markers in Speech, 66–107. Cambridge: Cambridge University Press. Hirschberg, Julia 2000 A corpus-based approach to the study of speaking style. In: Merle Horne (ed.), Prosody: Theory and Experiment. Festschrift in Honor of Gösta Bruce. 335–350. Dordrecht: Kluwer Academic Press.
Prosody and pragmatic effects
209
Hirschberg, Julia, Diane Litman and Marc Swerts 2004 Prosodic and other cues to speech recognition failures. Speech Communication 43(1–2):155–75. Hirst Daniel and Albert Di Cristo (eds.) 1998 Intonation Systems. Cambridge, CUP Jefferson, Gail 1985 An exercise in the transcription and analysis of laughter. In: Teun A van Dijk (ed.), Handbook of Discourse Analysis, Vol 3, 25–34. London: Academic Press. Jun, Sun-Ah 2006 Prosodic Typology; The phonology of intonation and phrasing. Oxford: Oxford University Press. Knowles, Gerry 1978 The nature of phonological variables in Scouse. In: Peter Trudgill (ed.), Sociolinguistic patterns in British English, 80–90. London: Edward Arnold. Knowles, Gerry, Anne Wichmann and Peter Alderson (eds.) 1996 Working with Speech: perspectives on research into the Lancaster/IBM Spoken English Corpus. London: Longman. Koopmans-van Beinum, Florien 1991 Spectro-temporal reduction and expansion in spontaneous speech and read text: Focus words versus non-focus words. In: Proceedings of the ESCA Workshop ‘ Phonetics and Phonology of Speaking Styles: Reduction and Elaboration in Speech Communication’. Paper 36, 1–5. Barcelona, Catalonia, Spain, 30 September – 2 October. Kousidis, Spyros, David Dorran, Ciaran McDonnell and Eugene Coyle 2009 Time Series Analysis of Acoustic Feature Convergence in Human Dialogues. Dublin Institute of Technology, Digital Media Centre Conference papers. Kuiper, Koenraad 2004 Formulaic performance in conventionalised varieties of speech. In: Norbert Schmitt (ed.), Formulaic sequences: acquisition, processing and use. Amsterdam: John Benjamins. Ladd, D. Robert 1996 Intonational phonology. Cambridge: Cambridge University Press. Lam, Phoenix 2009 What a difference the prosody makes: the role of prosody in the study of discourse particles. In: Dagmar Barth-Weingarten, Nicole Dehé and Anne Wichmann (eds.), Where prosody meets pragmatics, 107–126. Bingley, UK: Emerald. Levon, Erez 2006 Hearing ‘gay’: Prosody, interpretation, and the affective judgments of men’s speech. American Speech 81: 56–78. Levow, Gina-Anne 1999 Understanding recognition failures in spoken corrections in human-computer dialogue. In: Proceedings Dialogue and prosody, ESCA Tutorial and Research Workshop (DIAPRO-1999), 193–198. Veldhoven, The Netherlands, 1–3 September.
210
Anne Wichmann
Llisteri, Joaquim 1992 Speaking Styles in Speech Research. In: Proceedings ELSNET/ESCA/SALT Workshop on Integrating Speech and Natural Language. Dublin, Ireland, 15–17 July. Local, John 1986 Patterns and problems in a study of Tyneside intonation. In: Catherine JohnsLewis (ed.), Intonation in Discourse, 181–198. London: Croom Helm. Local, John 1992 Continuing and Restarting. In: Peter Auer and A. di Luzio (eds.), The Contextualization of Language, 273–296. Amsterdam: Benjamins. Lutzross, Auburn 2010 You sound like a Lesbian: Sociophonetic Stereotypes of Lesbian Speech. Hampshire College, Amhurst, MA: undergraduate dissertation. Martin, Jean-Claude, Patrizia Paggio, Peter Kuehnlein, Rainer Stiefelhagen and Fabio Pianesi 2008 Introduction to the special issue on multimodal corpora for modeling human multimodal behavior. Language Resources and Evaluation 42(2): 253–264. Mennen, Ineke 2007 Phonological and phonetic influences in non-native intonation. In: Jürgen Trouvain and Ulrike Gut (eds.), Non-native prosody: Phonetic description and teaching practice 53–76. Berlin: Mouton de Gruyter, Mennen, Ineke, Aoju Chen and Fredrik Karlsson 2010 Characterising the internal structure of learner intonation and its development over time. In: Proceedings of New Sounds 2010: The sixth international symposium on the acquisition of second language speech. University of Poznan´. 1–3 May. McCann, Joanne and Sue Peppé 2003 Prosody in Autism Spectrum Disorders: A Critical Review. International Journal of Language and Communication Disorders 38 (4): 325–350. McConnell-Ginet, Sally 1983 Intonation in a man’s world. In: Barry Thorne, Cheris Kramarae and Nancy Henley (eds.), Language, gender and society, 69–88. Rowley, Mass: Newbury House. Moonwomon-Baird, Birch 1997 Toward the Study of Lesbian Speech. In: Anna Livia and Kira Hall (eds.), Queerly Phrased: Language, Gender, and Sexuality, 202–213. NY: Oxford University Press. Müller, Frank Ernst 1996 Affiliating and disaffiliating with continuers: prosodic aspects of recipiency. In: Elizabeth Couper-Kuhlen and Margret Selting (eds.), Prosody in Conversation, 131–176. Cambridge: Cambridge University Press. Nelson, Gerald, Sean Wallis and Bas Aarts 2002 Exploring natural language: working with the British Component of the Interantional Corpus of English. Amsterdam: John Benjamins. Nishio, Masaki and Seiji Niimi 2008 Changes in Speaking Fundamental Frequency Characteristics with Aging. Folia Phoniatrica and Logopaedica 60(3): 120–127.
Prosody and pragmatic effects
211
Ohala, John J. 1982 The voice of dominance. Journal of the Acoustical Society of America 72: (1) S66-S66 (November). Ohala, John J. 1994 The frequency code underlies the sound-symbolic use of voice pitch. In: L. Hinton, J. Nichols and J.J. Ohala (eds.), Sound Symbolism, 325–347. Cambridge: Cambridge University Press. Pardo, Jennifer S. 2006 On phonetic convergence during conversational interaction. Journal of the Acoustical Society of America 119(4): 2382–2393. Pierrehumbert, Janet B. 1980 The phonetics and phonology of English intonation. MIT PhD thesis. Podesva, Robert J. 2007 Phonation type as a stylistic variable: The use of falsetto in constructing a persona. Journal of Sociolinguistics 11(4): 478–504. Podesva, Robert J., Sarah J. Roberts and Kathryn Campbell-Kibler 2001 Sharing Resources and Indexing Meanings in the Production of Gay Styles. In: Kathryn Campbell-Kibler, Robert J. Podesva, Sarah J. Roberts and Andrew Wong (eds.), Language and Sexuality: Contesting Meaning in Theory and Practice, 175–189. Stanford, CA: CSLI Publications. Reubold, Ulrich, Jonathan Harrington and Felicitas Kleber 2010 Vocal aging effects on F0 and the first formant: A longitudinal analysis in adult speakers. Speech Communication 52: 638–651. Roth, Wolff-Michael and Kenneth Tobin 2010 Solidarity and conflict: Prosody as a transactional resource in intra- and intercultural communication involving power differences. Cultural Studies of Science Education, Volume 5, Issue 4: 849–857. Schegloff, Emanuel A. 1982 Discourse as an interactional achievement: some uses of ‘uh huh’ and other things that come between sentences. In: Deborah Tannen (ed.), Analyzing Discourse: Text and Talk. Georgetown University Round Table on languages and Lingusitics 1981, 71–93. Washington DC: Georgetown University Press. Schegloff, Emanuel, A. 1998 Studying prosody in talk-in-interaction. Language and Speech 41(3–4): 235–263. Shobbrook, Katherine and Jill House 2003 High Rising Tones in Southern British English. In: Proceedings, 15th International Congress of Phonetic Sciences, 1273–1276. Barcelona, 3–9 August. Shriberg, Elizabeth, Rebecca Bates, Andreas Stolcke, Paul Taylor, Daniel Jurafsky, Klaus 1998 Ries, Noah Coccaro, Rachel Martin, Marie Meteer and Carol van Ess-Dykema Can Prosody aid the automatic classification of Dialog Acts? In: Language and Speech 41: 443–492. Shriberg, Elizabeth and Andreas Stolcke 2008 The case for automatic higher-level features in forensic speaker recognition. In: Proceedings, INTERSPEECH 2008 – 9th Annual Conference of the International Speech Communication Association, Brisbane, Australia, 22–26 September. ISCA Archive, http://www.isca-speech.org/archive/interspeech_2008 (accessed 3 June 2010).
212
Anne Wichmann
Smith, Philip M. 1979 Sex markers in speech. In: Klaus R. Scherer, Howard Giles (eds.), Social markers in speech, 109–146. Cambridge: Cambridge University Press. Smyth, Ron, Greg Jacobs and Henry Rogers 2003 Male voices and perceived sexual orientation: An experimental and theoretical approach. Language in Society 33: 329–350. Stenström, Anna-Brita 1994 An Introduction to Spoken Interaction London: Longman. Svartvik, Jan (ed.) 1990 The London-Lund Corpus of Spoken English: description and research. (Lund Studies in English 82). Lund: Lund University Press. Terango, Larry 1966 Pitch and duration characteristics of oral reading of males on a masculinityfemininity dimension. Journal of Speech and Hearing Research 9: 590–95. Trim, John L. M. 1988 Some contrastive intonated features of British English and German. In: J. Klegraf and D. Niels (eds.), Essays on the English Language and applied linguistics on the occasion of Gerhard Nickel’s 60th birthday, 235–49. Heidelberg: Julius Groos. Trouvain, Jürgen and William J. Barry 2000 The prosody of excitement in horse race commentaries. Proceedings, ISCA Workshop on Speech and Emotion, 86–91. Newcastle, N. Ireland 5–7 September. Trouvain, Jürgen and Ulrike Gut (eds.) 2007 Non-native prosody: Phonetic description and teaching practice. Berlin: Mouton de Gruyter. Tsur, Reuven 2006 Kubla Khan: Poetic Structure, Hypnotic Quality and Cognitive Style. Amsterdam: John Benjamins. Van Bezooijen, Renée 1995 Sociocultural aspects of pitch differences between Japanese and Dutch women. Language and Speech 38: 253–265. Van Bezooijen, Renée and Charlotte Gooskens 1999 Identification of Language Varieties: The Contribution of Different Linguistic Levels. Journal of Language and Social Psychology 18(1): 31–48. Walker, Gareth 2004 On some interactional and phonetic properties of increments to turns in talkin-interaction. In: Elisabeth Couper-Kuhlen and Cecilia E. Ford (eds.), Sound Patterns in interaction: cross-linguistic studies from conversation, 147–169. Amsterdam/Philadelphia: John Benjamins. Warren, Paul and Nicola Daly 2000 Sex as a factor in rises in New Zealand English. In: Janet Holmes (ed.), Gendered speech in social context: perspectives from gown and town, 99–115. Wellington: Victoria University Press. Weintraub, Mitch, Kelsey Taussig, Kate Hunicke-Smith and Amy Snodgrass 1996 Effect of Speaking Style on LVCSR Performance. In: Proceedings of the 4th International Conference on Spoken Language Processing (ICSLP ’96), 16–19. Philadelphia, PA, USA, October 3–6.
Prosody and pragmatic effects
213
Wennerstrom, Ann 2001 The Music of Everyday Speech. Oxford: Oxford University press. Whiteside, Sandra P. and C. Hodgson 1998 The development of fundamental frequency in 6- to 10-year old children: A brief study. Journal of the International Phonetic Association 28(1–2): 55–62. Wichmann, Anne 1996 Prosodic style: a corpus-based approach. In: Gerry Knowles, Anne Wichmann and Peter Alderson (eds.), Working with Speech: perspectives on research into the Lancaster/IBM Spoken English Corpus, 168–188. London: Longman. Wichmann, Anne 2000 Intonation in Text and Discourse. London: Longman. Wichmann, Anne 2001 Spoken Parentheticals. In: Karin Aijmer (ed.), A Wealth of English, 177–193. (Festschrift for Göran Kjellmer). Gothenburg: Gothenburg University Press. Wichmann, Anne 2004 The intonation of please-requests: a corpus-based study. Journal of Pragmatics 36(9): 1521–1549. Wichmann, Anne 2005 ‘Please’ – from courtesy to appeal: the role of intonation in the expression of attitudinal meaning. English Language and Linguistics 9(2): 229–253. Wichmann, Anne 2008 Speech Corpora. In: Anke Lüdeling and Merya Kytö (eds.), Corpus Lingustics: An International Handbook. Volume 1. (Handbooks of Linguistics and Communication Science 29.1–2), 187–207. Berlin: Mouton de Gruyter. Wichmann, Anne 2009 ‘Sorry’: Prosodic Evidence for Grammaticalisation in Progress. In: Rhonwen Bowen, Mats Mobärg and Sölve Ohlander (eds.), Corpora and Discourse – and Stuff. Papers in Honour of Karin Aijmer, 327–337. (Gothenburg Studies in English 96.) Gothenburg: Acta Universitatis Gothoburgensis. Wichmann, Anne 2010 Intonational meaning in institutional settings: the role of syntagmatic relations. Cultural Studies of Science Education 5(4): 849–857 Wichmann Anne, Anne-Marie Simon-Vandenbergen and Karin Aijmer 2010 How prosody reflects semantic change: a synchronic case study of ‘of course’. In: Hubert Cuyckens, Kristin Davidse and Lieven Vandelanotte (eds.), Subjectification, intersubjectification and grammaticalisation, 103–154. (Topics in English Linguistics). Berlin: Mouton de Gruyter. Zellner Keller, Brigitte 2006 Ageing and Speech Prosody. In: Proceedings Speech Prosody 2006. Dresden, Germany. 2–5 May. http://www.isca-speech.org/archive/sp2006/papers/ sp06_001.pdf (Accessed 3 June 2010) Zollna, Isabel 2003 Stimmen der Distanz: professionnelle monologische Sprechstile. Tübingen: Stauffenburg Verlag.
III. Pragmatic markers and the notion of speaker attitude
8.
Pragmatic markers in a sociopragmatic perspective Ad Foolen
1.
Introduction
More often than not, the interpretation of an utterance is not fully given with its explicit content. The hearer has to interpret the utterance in the context in order to understand the intended communicative goal. Speakers often add signals to their utterance which guide the hearer in this interpretation process. These signals are of various types: non-verbal signals, intonation, and special words or phrases. It is such special words and phrases which are the focus of the present overview. Pragmatic markers (PMs) will be used here as the general cover term for these items. Speech acts, implicatures and politeness are pragmatic topics which each have their classic references. For pragmatic markers, this is less pronounced, although books like Schourup (1985) and Schiffrin (1987) for research on English, Weydt (1969) for German “Partikelforschung” and Ducrot et al. (1980) for research on French “particules énonciatives” were influential pioneering works. In the past 15 years, quite a lot overview articles on PM research have been published. As the present volume takes a sociopragmatic perspective, I will pay only restricted attention (section 2) to general issues that typically occur in such overviews, like the variable terminology that is used for PMs, the different theoretical frameworks in which PMs have been studied, their syntax (cf. Urgelles-Coll 2010) and their notorious polysemy or polyfunctionality, which makes them so hard to analyze.1 And I will neglect the diachronic study of PMs, which thrived in recent years in the context of grammaticalization (cf. Brinton 1996) and pragmaticalization (cf. Günthner and Mutz 2004) research. Instead, I will focus on social aspects, in particular the study of PMs in a sociolinguistic context (variation and contact studies, see section 3) and the study of PMs in applied, socially relevant, perspectives like foreign language teaching and translation (section 4). Section 5 concludes this overview. But first, in section 2, I will give some general background information on PM research.
218
Ad Foolen
2.
Terminology, theories and methods
2.1.
Terminology
Among the different terms for PMs, discourse markers is probably used most frequently in the literature (cf. Schiffrin 2001, Maschler 2009, Fraser 2009, Fox Tree 2010, Dér 2010), followed by pragmatic markers (Fraser 1996, Andersen and Fretheim eds. 2000, Feng 2008, Norrick 2009), pragmatic particles (Foolen 1996, Cook 1999), discourse particles (Aijmer 2002, Fischer 2006, Misˇkovic´-Lukovic´ and Dedaic´ 2010), and just particles (van der Wouden, Foolen and van de Craen eds. 2002, Thüne and Ortu eds. 2007). Less used terms are, for example, phatic connectives (Bazzanella 1990) and discourse landmarks (Celle and Huart eds. 2007). In French research, particules énonciatives is the most viable term, see Fernandez (1994), but different terms are used here as well: mots du discours, particules discursives (see Vincent 1993). In German studies, besides modal particles the term Abtönungspartikel (Weydt 1969) is used. This terminological variation does not necessarily involve theoretical differences. In the present overview, the term “pragmatic marker” is used without a specific theoretical stance. It is a practical term, as it combines two rather neutral parts: “marker” is a functional notion which is neutral with respect to formal part of speech divisions like adverb, conjunction, or interjection (see Norrick, this volume). It is also neutral with respect to the phonological-morphological properties of the marker: “particle” would suggest small size, and although PMs are indeed typically “little words”, there are adverbs like moreover and phrases like you know and by the way which fulfill PM functions no less than particles. The first part of the term “pragmatic marker” is neutral with regard to medium (written text and spoken discourse) and also covers non-verbal aspects of social interaction like the social relation between speaker and hearer. The latter point is illustrated by Gómez de García, Axelrod and Luz Garcia’s (2010) analysis of the Ixil Maya word vet. They call this item a discourse marker but in their analysis, it becomes clear that some of the functions of this polyfunctional item involve social relations that go beyond discourse. According to the authors, vet can mark the interlocutors’ roles in particular socio-cultural activities. As was stated above, there are no phonological-morphological or part of speech restrictions on PMs. Neither are there syntactic restrictions. Many studies deal with PMs in English, which typically occur at the left periphery of utterances, but there is no principled reason to exclude markers that occur in other positions, like the Wackernagel position (after the first constituent or after the first stressed word, see, for example Ferraresi 2005: ch. 4 on discourse particles in Gothic), the middle field, or the utterance-final position. German, Dutch and Scandinavian languages have adverbial particles in the middle field, so called modal particles.2 Finnish, Hungarian and classical Greek have similar elements. Waltereit and
Pragmatic markers in a sociopragmatic perspective
219
Detges (2007) argue that modal particles should be distinguished from discourse markers in terms of their scope. According to them, discourse markers function at discourse level, whereas modal particles function at speech-act level (see also the discussion in Traugott 2007). This may very well be the case, but from a more general perspective, they are both signals to the hearer about how (aspects of) the utterance should be interpreted. I thus agree with Diewald’s (2006: 423) view that “[d]iscourse particles and modal particles are two distinct classes in German. These classes are, however, both related to discourse, so that it is reasonable to treat them together.” On the same ground, sentence-final particles in East Asian languages like Chinese and Japanese should be considered as belonging to the field of study of PMs; for some recent studies see Lee (2007), Strauss and Xiang (2009), and McGloin and Konishi (2010).3 Sentence-final particles occur in other languages too. In Norwegian, the use of sentence-final modal particles is rather widespread, see Andvik (1992: 15–18) and Fretheim (2010: 316–322) and Dutch has particles like hoor ‘hear’, hè ‘isn’t it’, zeg ‘say’ and joh ‘you, mate’ in final-position, see Kirsner and van Heuven (1996) and van der Wouden and Foolen (2010). Why languages differ in their preference for putting PMs in specific structural positions, is an interesting (typological) question, which, however, is beyond the scope of the present overview. 2.2.
Theories
PMs are studied from a range of theoretical perspectives and methodologies, see the contributions in Fischer (ed.) (2006). One of them is Relevance Theory, which has been a rather productive framework for PM analysis, see, among many others, Andersen and Fretheim (eds.) (2000), Blakemore (2002) and Pons Bordería (2008). According to Relevance Theory, PMs encode procedural meaning, i.e. instructions to the hearer how to treat the propositional meaning of the utterance. Another framework is the “Geneva model” (see Roulet 2006), which has been applied in particular on French PMs.4 Ethnomethodological conversation analysis (CA) is an approach that originated in sociology, which is the reason that it deserves special attention in a sociopragmatic overview. A strong feature of this approach is its close attention to the details of interactional processes. One would expect that this unavoidably leads to analyses of the function of PMs in conversation. However, PM analyses are not abundantly present in CA publications. The explanation might be that conversational researchers typically take as a point of departure the question what speakers do in interaction (cf. “doing other-attentiveness”, Bolden 2006), and PMs come in only as one of the means of realizing those “doings”. See however Jucker and Smith (1998), Heritage (1998), Mazeland and Huiskes (2001) and Bolden (2006) as examples of conversation analytic work which focusses more directly on PMs.
220
Ad Foolen
Although different theories highlight different aspects of the functioning of PMs, the empirical findings stated in terms of a specific approach can, in my opinion, rather easily be transferred to a different framework. In other words, the field of PM studies looks, from outside, as rather divided into different schools and theories, but in practice, this is not too much of a problem for communication between researchers. In an overview of PMs from a sociopragmatic perspective, it seems appropriate to devote special attention to a notion that was introduced by John Gumperz in the context of his sociolinguistic studies, namely “contextualization cue”. Gumperz (2001: 221) defines the notion as follows: I use the term contextualization cue to refer to any verbal sign which, when processed in co-occurrence with symbolic grammatical and lexical signs, serves to construct the contextual ground for situated interpretation and thereby affects how constituent messages are understood. Code-switching is one such contextualization cue. Others include pronunciation along with prosody (i.e. intonation and stress), rhythm, tempo, and other such suprasegmental signs. Contextualization cues, when processed in co-occurrence with other cues and grammatical and lexical signs, construct the contextual ground for situated interpretation and thereby affect how particular messages are understood. As metapragmatic signs (Lucy 1993), contextualization cues represent speakers’ ways of signaling and providing information to interlocutors and audiences about how language is being used at any one point in the ongoing exchange.
From this quote, we can infer that Gumperz distinguishes suprasegmental signs plus code-switching from grammatical and lexical signs. As a third group of signals, Gumperz distinguishes “other cues”, by which he probably means non-verbal cues (gesture, posture, facial expression). Gumperz seems to consider “metapragmatic sign” as a cover term for all signs and signals that serve the contextual grounding function. Contextualization cues function in an implicit and context-dependent way, which makes them hard to perceive (cf. Auer 1992). A specific code-switching (say from English into Italian) has no inherent meaning. What it means in a specific discourse, can only be established on an inferential basis, and for this understanding, one has to be an “insider”, with specific background knowledge. In Gumperz’ view, communication problems between different social groups and cultures often go back to the inability to perceive and interpret contextualization cues. The awareness of the implicit nature of contextualization cues may have been a reason for Gumperz to keep them apart from “grammatical and lexical signs” as the latter look more explicit, can’t be missed as such by “outsiders”, even if their meaning will not always be clear. Levinson (2003), however, does include these more conventionalized coding devices in the class of contextualization cues, besides more implicit signals like prosody and kinesics. Then a principled distinction between contextualization cues and pragmatic markers isn’t useful anymore.
Pragmatic markers in a sociopragmatic perspective
221
To round off this discussion of contextualization cues, I would like to point out another aspect of contextualization cues that has been stressed by Gumperz, namely their “context-evoking” character. What the relevant context for the interpretation of an utterance is, is not fixed in advance. It partly depends on the utterance and the contextualization cues themselves, which (aspects of a) context should be used in the interpretation process. At the same time, the function of a contextualization cue is dependent on the specific context. Contextualization cue and context are, thus, involved in a “bi-directional” interpretative process which is never conclusive. In this respect, traditional PM research could learn something from the micro-sociolinguistic research on contextualization cues. Too often, PMs are considered as markers that only explicate the already implicitly given relation between context and utterance. In this “simple” view, PMs could as well be left out, they are only an extra service to the hearer. On the basis of Gumperz’ work, PM research could become more aware of the fact that what counts as PM and as context is often not given in advance. 2.3.
Methodology
Besides terminology and theories, there is the methodological issue, about which we can, however, be short. In recent years, PM research has profited enormously from the increasing availability of corpora. In the 1970s and 1980s, it was acceptable to do research on PMs based on intuitions only. This is unthinkable now. As Mey (2004) stated “the study of the pragmatic aspects of language use need not preclude a rigorous quantification of linguistic data” (as quoted in Romero-Trillo 2008: 6). The minimal methodological requirement in present-day research is that an analysis of a PM in a specific language is based on a substantial set of “real” uses of the marker, whereby not only isolated utterances but also their context is taken into consideration. Better still is the exhaustive analysis of a corpus, in which all occurrences of a PM are accounted for. Modern corpora make possible a further step, as they are constructed on the basis of sociolinguistic and stylistic variables. Such corpora invite PM studies from a variational and quantitative perspective, to which we will turn in the next section.
3.
Variation within languages
In Foolen (1996), less than a page was devoted to PM research from a variational perspective (section 4.1 and 4.2). Since then, variational pragmatics has gained ground, see Schneider and Barron (eds.) (2008), and the variational study of PMs is part of this new field of study.
222
Ad Foolen
Not all varieties of a language (national varieties, dialects, sociolects, etnolects, etc.) use the same pragmatic markers and, more difficult to discover, one and the same pragmatic marker can be used in different ways in different varieties. As Romero-Trillo (2006: 640) rightly states, functions are more stable than forms: “[T]he phenomenon of discourse markers shows that spoken interaction needs to have a pragmatic skeleton, consisting of such discourse slots, that holds the communicative force of the interaction together. The slots are filled by elements that may vary according to regional, idiolectal, or sociolinguistic features within one and the same language.” In the following subsections, I will consider PMs in relation to some of the sociolinguistic parameters along which varieties have been distinguished in sociopragmatic research. 3.1.
Regional and national varieties of a language
Nonstandard varieties often show a richer inventory of pragmatic markers than standardized languages. Pusch (2002: 104), for example, observed that “Gascony Occitan is unique among Romance languages in having developed an elaborate system of highly recurrent preverbal particles.” In his paper, Pusch focuses on the “enunciative” particle que ‘that’, which signals topic-continuity (where the absence indicates discontinuity). He also points out (Pusch 2002: 107) that the historical development of such items is hard to reconstruct “as the use of enunciative particles is and probably always has been considered as a typically oral feature and therefore was banned from written texts.” More recent developments can, however, be based on sociolinguistic recordings of the past decades. Vincent (2005) traced the development of non-standard discourse markers in Québec French (par example, ‘for example’, mettons ‘let’s put’, disons ‘let’s say’, comme ‘like’, genre ‘kind of’, and style ‘way’, based on interviews carried out in 1971, 1984 and 1995. PMs in Québec French are also the topic of Dostie (2009). Butler and King (2008) analyze the evolution and use of mais dame ‘but madam’, which is “frequently encountered, in one particular variety of North American French, that spoken on western Newfoundland’s Port-au-Port Peninsula” (Butler and King 2008: 63). According to their analysis (Butler and King 2008: 81), “mais dame establishes contextual coordinates – reaffirmation of stance, narrative evaluation, return to a prior concern – relative to which the emerging discourse is to be understood.” Beeching (2009) studied the extension of the uses of bon ‘well’ in modern spoken French. The rise in frequency of the compound expressions mais bon ‘but well’ and parce que bon ‘because well’ suggests a shift towards increased intersubjectivity. Like French, English has received considerable attention with regard to the use of PMs across varieties of English. O’Keeffe and Adolphs (2008) found that listener response token activity differed between British and Irish English, both in form and frequency (“British English conversations contained far more”, O’Keeffe
Pragmatic markers in a sociopragmatic perspective
223
and Adolphs 2008: 92). Grant (2010) studied the use of I don’t know and I dunno by New Zealand speakers of English. She found that New Zealand speakers use these phrases with different frequency and for different reasons than British speakers (NZ speakers use both phrases most often as a hedge or marker of uncertainty). Pichler (2009) analyzed I don’t know and I don’t think and local variants of these expressions in a northern English dialect (Berwick-upon-Tweed). The use of the local variants correlates with age and gender. Some varieties of English contain PMs that do not occur in standard British or American English. Gupta (2006) and Ler (2006) analyzed sentence-final PMs in Singapore English. According to Ler, “[i]n a contact variety of English in Singapore, there is a group of discourse particles, such as lah, lor, meh, hah, and hor, which give Singapore Colloquial English (SCE) its special flavor. These DPs belong to the most frequent words used in conversations” (Ler 2006: 149). To get an idea of their use and function see http://en.wikipedia.org/wiki/Singlish. Sentence-final particles are primarily associated with East-Asian languages (see section 2), but recently, researchers have become aware of the relevance of the final position in other languages. Miller (2006: 216) observed that “[t]he Glasgow variety of Scottish English has a construction with but in final position, in which but is equivalent in discourse function to though. In she’ll do the job for us – she’s no very big but, the speaker makes an assertion about the person’s abilities but concedes a point concerning the person’s size.” Mulder, Thompson and Penry Williams (2009) found a similar use of but in Australian English. And Miller (2009) showed that in Australian and New Zealand English, like does not only occur clause-initially and clause-medially, but also clause-finally. In final position, like has to do with explanations and preventing hearers making incorrect inferences. 3.2.
Age
The variable “age” has led to studies which focus mainly on the language of adolescents. The Corpus of London Teenage Language (COLT) has been a fruitful basis for PM research. Andersen (2001) analyzed the use of like, and of innit/is it as invariant tags and follow-ups (invariant across the grammatical categories person and number) in the COLT and compared it to adult talk in the British National Corpus (BNC). His extensive study showed (Andersen 2001: 301) “that these phenomena are indeed teenage-specific; invariant use of innit/is it is not found in the adult reference material studied, while like as a marker occurs only to a little extent in that material, and with a much narrower set of functions than in the adolescent corpus.” Like is typically associated with adolescent speech, but Levey (2006) showed that new uses of like can be found in preadolescent speech as well.5 Two studies on the use of PMs by young Spanish speakers can be found in Stenström and Jørgensen (eds.) (2009). Stenström (2009) contributed an analysis of the various prag-
224
Ad Foolen
matic functions of pues nada ‘anyway’ in Spanish teenage conversation, which serves as an organizer and monitor on the discourse level, and Jørgensen (2009) analyzed the use of en plan ‘like’ as a hedge in Spanish teenage language. If PMs are more frequent or show more functions in adolescent speech in comparison with adult speech, then this can be interpreted as an indication of a grammaticalization (or pragmaticalization) process, cf. Erman (2001) on you know and Aijmer (2008b) on obviously and definitely. In her study of London teenage language, Aijmer observed that adolescents used obviously and definitely more frequently, involving intensification and affective meaning, than adults. The relation between frequency of use, range of functions, age and grammaticalization is a complex one. In combination with diachronic studies, the study of variation across age groups can contribute to theoretical progress in this field. 3.3.
Genres in spoken and written discourse
Hentschel (1986) and other studies on German modal particles found that these PMs occur primarily in informal spoken discourse. It would, however, be wrong to generalize this finding to all PMs. According to Miller (2006: 215), “Both writers and speakers use discourse particles, but different particles occur in unplanned speech/conversation and in various types of written text.” He observes that checking and backchannel particles are particularly frequent in task-related dialogues. As to written text, Miller (2006: 215) refers to Biber et al. (1999: 880; cf. also Biber, this volume), when he states that [L]linking adverbs are most frequent in academic writing (7100 per one million words), slightly less frequent in conversation (5900 per one million words), and relatively infrequent in fiction and newspapers (2200 and 1800 per one million words, respectively). The higher figure for conversation results from the intensive use of so and then, which are infrequent in academic writing. In contrast, academic writing offers a wide range of linking particles, such as for example, in addition, nevertheless, furthermore, and hence.
Verikaite˙ (2005) compared the use of additive, adversative, causal and temporal markers in academic textbooks and research articles. Causal and temporal markers occurred more in research articles, and within the group of adversative markers (Verikaite˙ 2005: 74), “corrective conjunctions prevailed in the genre of research articles, whereas contrastive conjunctions were more characteristic of the textbook.” Liu (2008) found variation in the frequency and usage patterns of English linking adverbials (however, in contrast, meanwhile, so, then, etc.) in five registers (spoken English, academic writing, fiction, news writing and other writing) in the British National Corpus (BNC). Findings like those of Verikaite˙ (2005) and Liu (2008) can probably be generalized in the sense that variation in function between text types will correlate with variation in the types and frequencies of PMs.
Pragmatic markers in a sociopragmatic perspective
225
Aijmer (2008a) analyzed commentary evaluative markers, a group of PMs which refer to expectations, as well as to hopes and wishes that something will take place or not take place. Examples are surprisingly, astonishingly, amazingly, sadly, fortunately, and unfortunately. These markers were generally more frequent in writing than in speech (the only exception being unfortunately). Aijmer (2008a: 17) suggests as an explanation for this difference “that it is possible that involvement is simply expressed differently in spoken and written language.” The study of the distribution of pragmatic markers across different genres can have consequences for the functional analysis of the markers themselves. A case in point is provided by Hasegawa (2010) who found that Japanese ne occurs in soliloquy as frequently as it does in dialogue, while yo does not. Both particles have traditionally been described almost exclusively in terms of information sharing, or a lack thereof, between speaker and addressee. “If we postulate that the primary function of ne is matching pieces of information and that of yo is as preparation, or a trigger, for inference, this highly skewed distribution [frequent use of ne in soliloquy] becomes common sense.” (Hasegawa 2010: 86). 3.4.
Gender and politeness
Brown and Levinson (1987: 146–162) showed that particles are often part of politeness strategies. With particles, speakers can hedge assumptions related to speech acts, with a softening effect. Vismans (1994) analyzed modal particles in Dutch directives and showed that they can be used for mitigation and for reinforcement. Some other studies on PMs from a politeness perspective are Meyerhoff (1994), Ouafeu (2006) and Chodorowska-Pilch (2008), who argues that Spanish verás ‘you will see’ can encode positive and negative politeness in the same utterance. From a variational perspective, it can be observed that gender differences in the use of PMs have been studied specifically in relation to politeness, see Holmes (1995: ch. 3) on English you know, I think, sort of, of course and Beeching (2002) on French c’est-à-dire ‘that means’, enfin lit. ‘finally’, hein ‘what’, and quoi ‘what’. Whereas Holmes found more tentative (polite) phrasings in women than in men, Beeching did not find confirmation for this, concluding that in France there is not much social asymmetry between the genders. Nevalainen (2006: 361) reports research on women’s language in 17th century discourse in which “women were found to use I think more in all recipient categories investigated: when writing to their nuclear family members (parents, children, and spouses), to other family members, to close friends and to other, more distant recipients.”
226 3.5.
Ad Foolen
Culture and identity
Maschler (2003) relates the use of the discourse marker nu ‘so?’ in Israeli Hebrew to the often (positive) perception of impatience in Israeli culture, see also Maschler (2009), in which she relates nu and other Hebrew PMs (bekitgsur ‘anyway’, ke’ilu ‘like, and tov ’good‘) to Israeli society, identity, and culture. Kita and Ide (2007) offer a discussion of possible socio-cultural motivations for nodding and final particles that shape Japanese conversation in a characteristic way. It would be interesting to see whether more cases of such relations between pragmatic markers and specific cultures can be discovered in future PM research. This overview did not pay attention to the topic of individual variation in the use of PMs. However, each researcher in this field will easily recognize the observation that people can differ quite a lot in their preference for specific PMs. Foolen et al. (2006: 142) observe that there is individual (instead of social) variation in the frequency of use of the Dutch quotative marker van ‘like’. It would be interesting to study, whether this individual variation has any “meaning”. Hellerman and Vergun (2007) suggest that adult learners of English in the U.S. use more discourse markers if they are more acculturated. Liao (2009) found that Chinese L1 graduate students who worked as teaching assistants in the U.S. differ individually in their use of PMs like yeah, oh, you know, like, well, I mean, ok, right, and actually. She shows (Liao 2009: 1324) “how their multifaceted social identities relative to their language attitudes and participation in the local community might contribute to their individual repertoires and frequency of DM [discourse marker] use.” The role of PM use as a marker of identity was also stressed by Schlieben-Lange (1979), who analyzed particles in the Bavarian variety of German. One of her observations was that Munich students in the sixties used modal particles like eh, halt, and f˜ei as a way to indicate an authentic way of life. Trester (2003) shows that in Costa Rican Spanish the expression pura vida, litt. ‘pure life’, has discourse structuring functions, but she stresses that it simultaneously plays a role in identity construction, cf. Trester (2003: 67): “A person choosing to use this phrase thus is not only alluding to this shared ideology and identity, he/she is at the same time constructing that identity by means of expressing it. Language is a very important tool of self-construction.” This constructivist view on PMs fits, in spirit, Gumperz’ view that contextualization cues construct the contextual ground for situated interpretation. 3.6.
New media
A new field of research is provided by novel settings of language use, in particular “the new media” (see Fox Tree 2010: section 4). Fox Tree et al. (2009) studied PMs in instant messaging and found that I mean, you know, well, oh, I dunno, and like are used like in dialogue, but the rates of use differ (see Fox Tree 2010: 277). As PMs are sensitive to contextual factors, they are useful indicators for establishing
Pragmatic markers in a sociopragmatic perspective
227
the parameters that determine language use in new genres like email, chat, and internet fora. It is to be expected that the new media will be a fruitful field for future research on PMs. Student papers like Sewell (2009) and Wong (2009) show that this is an attractive and challenging field for young researchers who have easy access to the relevant data. 3.7.
Bilingualism and borrowing: Contact linguistics
Contact linguistics is not a strictly sociolinguistic discipline. Besides the aspect of social constellations which lead to language contact of various types (in which power and prestige play a role), there is a psycholinguistic dimension to it (bilingual speakers) and a system linguistic aspect, in the sense that not all parts of the system are equal in “borrowability” (cf. Matras 1998 and Matras 2009: chapter 6). According to Matras, discourse markers are rather high in the borrowability hierarchy (they come close after nouns and verbs), which explains the easy borrowing of PMs between the languages of bilingual speakers and eventually between the languages themselves, cf. Fuller (2001), Sakel (2007) and the contributions in a special issue of The International Journal of Bilingualism, in particular Maschler (2000) and Matras (2000). On the basis of their study of Texas German discourse markers borrowed from English, like you know, Boas and Weilbacher (2009) critically discuss Matras’ pragmatic detachability scale, which states that within the group of PMs those that are less content-oriented (more “operational”) can be borrowed more easily in language contact. Torres and Potowski (2008) studied the use of Spanish and English discourse markers among first, second and third generation Spanish speakers in Chicago, comparing Mexican, Puerto Rican, and “MexiRican” speakers. There are differences in the use of PMs between these groups, but so turns out to be a core borrowing for all groups. The authors predict that the use of so will increase at the expense of Spanish entonces with subsequent generations. Peterson (2008) looked at the incorporation of English PMs in Finnish in Helsinki. In particular younger speakers use jees ‘yes’, enivei ‘anyway’, cool and pliis ‘please’. She also observed changes in the functions of existing Finnish PMs like kiitos ‘please’ and tyyliin ‘in the style of, like’ under the influence of English as a contact language in the age of globalization. Östman (2006: 255) raises the question “why Finland-Swedish dialects like the Solv dialect have developed a complex system of modals and particle-verbals which is lacking in standard (Finland) Swedish.” One answer that suggests itself (Östman 2006: 255) is “the close contact of the Swedish dialects to Finnish, which has a rich inflectional verbal morphology that can express many more modalities than what we typically find in Germanic languages. The similarities between Solv and Finnish (dialects) are not primarily found on the surface, but in the pragmatics of the two languages.”
228
Ad Foolen
Dittmar (2000) studied changes in the speech habits in the former German Democratic Republic after the fall of the wall in 1989. One change had to do with particles. Originally, the evidential particles halt and eben had a regional distribution: halt in the south of Germany, eben in the north, including the GDR. Halt spread, however, to the north, being used there as a more subjective (personal, friendly) variant of eben. This use did not spread to the GDR until after 1989. Early adopters were the ones who were open for a “western” life style.
4.
Pragmatic Markers in an applied perspective
4.1.
Learning and teaching pragmatic markers
Contrastive analysis is often seen as conditional for an appropriate way to teach a foreign language. This applied perspective also holds for the contrastive analysis of PMs. Modern contrastive analysis is often based on the analysis of language use in corpora instead of looking at the differences between languages on the system level. An example of such a “usage based” contrastive analysis is Gonzàlez (2004), who analyzed the use of pragmatic markers in oral narratives of personal experiences in Catalan and English. In the narratives, she found a high concentration of PMs in the action and evaluation segments, but more in Catalan than in English. PMs belong to the more difficult aspects of foreign language learning and teaching (see Fox Tree 2010: section 3). Romero-Trillo (2002) observed “pragmatic fossilization” of discourse markers in non-native speakers of English. As Romero-Trillo (2006: 641) concludes, this fossilization “reveals the shortcomings of learning a foreign language in a non-natural environment, especially as to the use of the elements that contribute to the pragmatics of interaction.” It is then no surprise that several studies have been devoted to the question how non-native speakers of different languages use PMs and how their teaching can be improved. House (2009) studied the use of you know by speakers of English as a lingua franca (ELF speakers). It turns out that they use you know differently from native speakers (House 2009: 190): “ELF speakers make use of you know when – at points of transitional relevance so to speak – they want to make salient coherence relations and focus on, or boost connections in discourse production and planning difficulties.”; cf. also House, this volume. Buysse (2009: 90) found that in specific contexts, (Belgian) non-native speakers of English use so “much more frequently than their native peers, which may be a part of an avoidance strategy towards other discourse markers such as I mean.” Müller (2005) analyzed the PMs like, well, you know and so in the interlanguage of German learners of English on the basis of comparable spoken corpus data. Differences between native and non-native speakers are attributed to various factors, including transfer from the mother tongue and teaching method (see also the review by Gilquin 2008).
Pragmatic markers in a sociopragmatic perspective
229
Möllering (2001, 2004) studied German modal particles in a didactic perspective. Wenzel (2002: ch. 6) analyzed the use of Dutch particles like maar ‘only, just’, alleen ‘only’, nou ‘now’, wel ‘yes, for sure’ by German learners of Dutch. As Dutch and German both have a class of modal particles, learners show transfer but also ‘fear of transfer’, i.e. avoidance strategies. Van Balen, Caspers and van der Wouden (2010) studied the acquisition of Dutch by native speakers of Spanish. They found a certain order in the acquisition of modal particles: maar ‘just’ was acquired first, followed by wel ‘yes, for sure’, dan ‘then’, dus ‘so’, nu/nou ‘now’, nog ‘still’, even ‘a moment’, ook ‘also’, eens ‘one time’, toch ‘nevertheless’, and misschien ‘perhaps’. It would be interesting to see follow up studies on the acquisition of PMs in other languages from this perspective. If more cases of regularity in the order of acquisition of PMs can be found, then these findings are potentially relevant for the construction of teaching material. And findings of this type can also be theoretically interesting, as they might indicate differences in the semantic complexity of the PMs involved. De la Fuente (2009) is an experimental study on different ways to teach Spanish discourse markers (o sea ‘that is’, entre tanto ‘meanwhile’, en cuanto ‘regarding’, and puesto que ‘because’) to English learners of Spanish. Two methods of instruction were compared: a conscious-raising task and an input enrichment task. The first task implied an explicit focus on form, the second task a more implicit focus on form. The learning results turned out to be better for the first type of task. Foolen (2010) analyzed different course books for teaching Dutch as a foreign language. In general, the courses provided enough input of particles, in particular in the dialogues which were provided as learning material, but explicit attention to the particles, in the form of grammatical information or focused exercises, was missing. 4.2.
Pragmatic markers in translation
In recent years, parallel and comparable corpora have become available. Parallel corpora consist of original and translated versions of the same text; comparable corpora are multilingual corpora that contain original texts in two or more languages which are matched by genre, time of publication, etc. (cf. Johansson 2007). Parallel corpora have been used to check how polyfunctional PMs are translated. If translators, as experienced language users, systematically choose the same translation for different pragmatic markers, then this can be taken as evidence for semantic closeness of the markers. If a specific marker is translated in different ways, this might be (additional) evidence for proposals for a polyfunctional analysis of a marker, based on monolingual data. Aijmer, Foolen and Simon-Vandenbergen (2006: section 6) present a general exposition of the method. Applications can be found in Simon-Vandenbergen and Aijmer (2007), Degand (2009), Niemegeers (2010), Vandeweghe (2010) and Nølke (2007), who looked at translations of French donc ‘so, therefore’ into a number of languages. Results of studies based on
230
Ad Foolen
parallel corpora can also be used for the improvement of bilingual dictionaries. An impressive piece of lexicographical work is Métrich and Faucher (2009) in which German particles are described in relation to their French equivalents. Comparable corpora come into play when one wants to compare original and translated varieties of a language. The experience of reading a text with the feeling that you are reading translated text is not uncommon. The translator may have transferred features of the source language into the target language, or he might have done his best to let the target text sound natural, to a degree that certain typical features of the target language are overused. Both effects have been observed with regard to pragmatic markers. Niemegeers (2010: chapter 10) analyzed the frequencies of the particles maar ‘only, just’, eens ‘once’, wel ‘yes, for sure’ and toch ‘nevertheless’ in a corpus of original Dutch in comparison to Dutch as translated from English. While there was a tendency to use the first three particles more in translated than in original Dutch, toch was used more frequently in original Dutch. Looking at English just, Niemegeers found that English, translated from Dutch, contained more occurrences than original English. Her conclusion (Niemegeers 2010: 304) is “that no generalizations can be made with respect to the use of Dutch modal particles in Dutch-English (and English-Dutch) translations.” More research, based on bigger comparable corpora, seems necessary to find out whether general translation patterns for PMs can be found. 4.3.
Computer-mediated natural language processing
In the field of natural language processing, PMs are often called ‘cue phrases’, see Hirschberg and Litman (1993). Cue phrases can play a role in natural language understanding systems (Heeman and Allen 1999) and in systems for human-machine-interaction. Suppose, one wants to construct a natural language processing device (for example a question-answering system) which makes a natural and friendly impression. If such a system can process PMs (understand and use them appropriately), then this will certainly contribute to its human-like behavior. Green (2006) tested this in a small pilot study with a beta version of the Restaurant Finder, an interactive internet application to find a restaurant along specific criteria (location, type of food, etc.). Green (2006: 134) concludes her paper with the following reflection: “Whether there is actually a market for interfaces that display the kind of emotional intelligence that particle use simulates is for others to determine. It is certain that further testing of the sort described here will tell us a lot more about particles, both individually and in interaction with each other.”
Pragmatic markers in a sociopragmatic perspective
5.
231
Conclusion
As the present overview has shown, PMs constitute a dynamic field of research. The research lines that were sketched in the previous sections should be continued. Further, the field needs integration of different sorts. First, theoretical and terminological integration would make the field more coherent and communication between researchers easier (cf. Galera Masegosa 2010, who combines Relevance Theory and the Lexical Constructional Model). Secondly, there are many descriptions available which deserve integration. Descriptive grammars of languages typically contain a chapter on “particles”. Integration of this material would make theorizing on PMs more adequate from a general linguistic point of view. Thirdly, there are phenomena like hedges, quotative markers, modal and evidential markers which have functions that overlap with the functions that are fulfilled by PMs (cf. Brinton 2008, Abraham 2009). An integrated theory on how linguistic information is contextualized should cover such items as well. Pragmatic markers are not the odd characters in language that occupy a place on the sideline of linguistic research. They provide insight in the relation between utterance and context, thus linking syntax and semantics to pragmatics. Such a link is not only beneficial for descriptive and theoretical linguistics but also for linguistics as applied in different fields of socially relevant practice.
Notes 1. Polysemy networks as proposed in Hansen (2008: 225–227) for the French adverbs déjà, encore, toujours and enfin are, in my view, an adequate way to state the relations between the different meanings and uses of a PM. 2. Many of the publications on modal particles are in German or Dutch, but for those who prefer English, I refer to contributions in Abraham (ed.) (1991) and some of the contributions in Weydt (ed.) (1989). Andvik (1992) and Fretheim (2010: section 5) analyze Norwegian modal particles, and van der Wouden (2002) studies the collocational behavior of Dutch modal particles. 3. This is not to suggest that Japanese or Chinese only have sentence-final PMs, see for example Onodera (2004) on Japanese demo, dakedo, ne and na, and Wang, Tsai and Yang (2010) on qishi ‘actually’ and shishishang ‘in fact’ in spoken Chinese, which appear in other positions. 4. Many of the studies on French PMs are in French. An overview article in English on research findings in French (and non-English Germanic languages, see note 2) would be a welcome enterprise. And probably more is available in other languages. The often very specific meanings of PMs invite publications in the language itself, which is understandable with regard to practical purposes like lexicography, reference grammar writing and developing teaching material. At the same time, international accessibility of the findings is important, as general theorizing on PMs should be based on data and descriptions from as many languages as possible.
232
Ad Foolen
5. Counterparts of like as a new quotative marker can be found in many languages, cf. Foolen (2008). Its use is typically associated with young speakers, although this generalization does not hold for Dutch van, which is used by speakers of all ages (see Foolen et al. 2006).
References Abraham, Werner (ed.) 1991 Discourse Particles. Amsterdam: Benjamins. Abraham, Werner 2009 Die Urmasse von Modalität und ihre Ausgliederung. Modalität anhand von Modalverben, Modalpartikel und Modus. In: Werner Abraham and Elisabeth Leiss (eds.), Modalität. Epistemik und Evidentialität bei Modalverb, Adverb, Modalpartikel und Modus, 251–302 (Studien zur deutschen Grammatik 77). Tübingen: Stauffenburg Verlag. Aijmer, Karin 2002 English Discourse Particles. Evidence from a Corpus. (Studies in Corpus Linguistics 10). Amsterdam: Benjamins. Aijmer, Karin 2008a At the interface between grammar and discourse – a corpus-based study of some pragmatic markers. In: Jesús Romero-Trillo (ed.), 11–36. Aijmer, Karin 2008b Modal adverbs in interaction – obviously and definitely in adolescent speech. In: Terttu Nevalainen, Irma Taavitsainen, Päivi Pahta and Minna Korhonen (eds.), The Dynamics of Linguistic Variation. Corpus Evidence on English Past and Present, 61–83. (Studies in Language Variation 2). Amsterdam: Benjamins. Aijmer, Karin, Ad Foolen and Anne-Marie Simon-Vandenbergen 2006 Pragmatic markers in translation: A methodological approach. In: Kerstin Fischer (ed.), 101–114. Andersen, Gisle 2001 Pragmatic Markers and Sociolinguistic Variation. A Relevance-theoretic Approach to the Language of Adolescents. Amsterdam: Benjamins. Andersen, Gisle and Thorstein Fretheim (eds.) 2000 Pragmatic Markers and Propositional Attitude. Amsterdam: Benjamins. Andvik, Erik E. 1992 A Pragmatic Analysis of Norwegian Modal Particles. Dallas, TX: The Summer Institute of Linguistics. Auer, Peter 1992 Introduction: John Gumperz’ approach to contextualization. In: Peter Auer and Aldo Di Luzio (eds.), 1–37. Auer, Peter and Aldo Di Luzio (eds.) 1992 The Contextualization of Language. Amsterdam: Benjamins. Bazzanella, Carla 1990 Phatic connectives as interactional cues in contemporary spoken Italian. Journal of Pragmatics 14(4): 629–647.
Pragmatic markers in a sociopragmatic perspective
233
Beeching, Kate 2002 Gender, Politeness and Pragmatic Particles in French. Amsterdam: Benjamins. Beeching, Kate 2009 Sociolinguistic factors and the pragmaticalization of bon in contemporary spoken French. In: Kate Beeching et al. (eds.), 215–230. Beeching, Kate, Nigel Armstrong and Françoise Gadet (eds.) 2009 Sociolinguistic Variation in Contemporary French. Amsterdam: Benjamins. Biber, Douglas, Stig Johansson, Geoffry Leech, Susan Conrad and Edward Finegan 1999 Longman Grammar of Spoken and Written English. London: Longman. Blakemore, Diane 2002 Relevance and Linguistic Meaning. The Semantics and Pragmatics of Discourse Markers. Cambridge: Cambridge University Press. Boas, Hans C. and Hunter Weilbacher 2009 The unexpected survival of German discourse markers in Texas German. CLS 42–1: The Main Session. Papers from the 42nd Annual Meeting of the Chicago Linguistics Society, 1–15. Bolden, Galina B. 2006 Little words that matter: Discourse markers “So” and “Oh” and the doing of other-attentiveness in social interaction. Journal of Communication 56: 661–688. Brinton, Laurel J. 1996 Pragmatic Markers in English: Grammaticalization and Discourse Functions. (Topics in English Linguistics 19). Berlin: Mouton de Gruyter. Brinton, Laurel J. 2008 The Comment Clause in English. Syntactic Origins and Pragmatic Development. Cambridge: Cambridge University Press. Brown, Keith (ed.) 2006 Encyclopedia of Language and Linguistics (Second Edition). Oxford: Elsevier. Brown, Penelope and Stephen C. Levinson 1987 Politeness. Some Universals in Language Usage. (Studies in Interactional Sociolinguistics 4). Cambridge: Cambridge University Press. Butler, Gary R. and Ruth King 2008 The French discourse marker mais dame: Past and present functions. Canadian Journal of Linguistics 53(1): 63–82. Buysse, Lieven 2009 So as a marker of elaboration in native and non-native speech. In: Stef Slembrouck et al. (eds.), 79–91. Celle, Agnès and Ruth Huart 2007 Connectives as Discourse Landmarks. Amsterdam: Benjamins. Chodorowska-Pilch, Marianna 2008 Verás in Peninsular Spanish as a grammaticalized discourse marker invoking positive and negative politeness. Journal of Pragmatics 40: 1357–1372. Cook, Haruko, M. 1999 Particles. Journal of Linguistic Anthropology 9(1–2): 181–183. Degand, Liesbeth 2009 On describing polysemous discourse markers. What does translation add to the picture? In: Stef Slembrouck et al. (eds.), 173–183.
234
Ad Foolen
de la Fuente, María José 2009 The role of pedagogical tasks and focus on form in acquisition of discourse markers by advanced language learners. In: Ronald P. Leow, Héctor Campos and Donna Lardiere (eds.), Little Words. Their History, Phonology, Syntax, Semantics, Pragmatics, and Acquisition, 211–221. Washington, DC: Georgetown University Press. Dér, Csilla Ilona 2010 On the status of discourse markers. Acta Linguistica Hungarica 57(1): 3–28. Diewald, Gabriele 2006 Discourse particles and modal particles as grammatical elements. In: Kerstin Fischer (ed.), 403–425. Dittmar, Norbert 2000 Sozialer Umbruch und Sprachwandel am Beispiel der Modalpartikeln halt und eben in der Berliner Kommunikationsgemeinschaft nach der ‘Wende’. In: Peter Auer and Heiko Hausendorff (eds.), Kommunikation in gesellschaftlichen Umbruchsituationen, 199–234. Tübingen: Niemeyer. Dostie, Gaétane 2009 Discourse markers and regional variation in French: A lexico-semantic approach. In: Kate Beeching et al. (eds.), 201–214. Ducrot, Oswald et al. 1980 Les Mots du Discours. Paris: Les Éditions de Minuit. Eerdmans, Susan L., Carlo L. Prevignano and Paul J. Thibault (eds.) 2003 Language and Interaction. Discussions with John J. Gumperz. Amsterdam: Benjamins. Erman, Britt 2001 Pragmatic markers revisited with a focus on you know in adult and adolescent talk. Journal of Pragmatics 33: 1337–1359. Feng, Guangwu 2008 Pragmatic markers in Chinese. Journal of Pragmatics 40: 1687–1718. Fernandez, M.M. Jocelyne 1994 Les Particules Énonciatives. Paris: Presses Universitaires de France. Ferraresi, Gisella 2005 Word Order and Phrase Structure in Gothic. Leuven: Peeters. Fischer, Kerstin 2006 Towards an understanding of the spectrum of approaches to discourse particles: Introduction to the volume. In: Kerstin Fischer (ed.), 1–20. Fischer, Kerstin (ed.) 2006 Approaches to Discourse Particles. (Studies in Pragmatics 1). Oxford Elsevier. Foolen, Ad 1996 Pragmatic particles. In: Jef Verschueren, Jan-Ola Östman, Jan Blommaert and Chris Bulcaen (eds.), Handbook of Pragmatics. Amsterdam: Benjamins (24 pages). Foolen, Ad 2008 New quotative markers in spoken discourse. In: Bernt Ahrenholz, Ursula Bredel, Wolfgang Klein, Martina Rost-Roth and Romuald Skiba (eds.), Empirische Forschung und Theoriebildung, 117–128. Frankfurt am Main: Peter Lang.
Pragmatic markers in a sociopragmatic perspective
235
Foolen, Ad 2010 Partikels volgens de schijf van vijf. [Particles according to the circle of five] Internationale Neerlandistiek 48(2): 41–51. http://www.internationaleneerlandistiek.nl/ Foolen, Ad, Ingrid van Alphen, Eric Hoekstra, Henk Lammers, Harrie Mazeland and Esther Pascual 2006 Het quotatieve van. Vorm, functie en sociolinguïstische variatie. [The quotative van. Form, function and sociolinguistic variation]. Toegepaste Taalwetenschap in Artikelen 76(2): 137–149. Fox Tree, Jean E. 2010 Discourse markers across speakers and settings. Language and Linguistics Compass 4(5): 269–281. Fox Tree, Jean E., Sarah A. Mayer and Teresa E. Betts 2009 At the crossroads of speaking and writing. Paper presented at the Society for Text and Discourse, July 26–28. Rotterdam, The Netherlands. Fraser, Bruce 1996 Pragmatic markers. Pragmatics 6(2): 167–190. Fraser, Bruce 2009 An account of discourse markers. International Review of Pragmatics 1(2): 293–320. Fretheim, Thorstein 2010 The metarepresentational use of main clause phenomena in embedded clauses. Linguistics 48(2): 301–324. Fuller, Janet M. 2001 The principle of pragmatic detachability in borrowing: English-origin discourse markers in Pennsylvania German. Linguistics 39(2): 351–369. Galera Masegosa, Alicia 2010 Discourse markers in the lexical constructional model: The case of the ‘So What X’ construction. Revista de Linguïstica y Lenguas Aplicadas 5: 41–52. Gilquin, Gaëtanelle 2008 Review of Müller (2005). Linguistics 46(3): 667–677. Gómez de García, Jule, Melissa Axelrod and María Luz García 2010 Sociopragmatic influences on the development and use of the discourse marker vet in Ixil Maya. In: Andrea L. Berez, Jean Mulder and Daisy Rosenblum (eds.), Fieldwork and Linguistic Analysis in Indigenous Languages of the Americas, 9–31. (Language Documentation & Conservation Special Publication 2). Gonzàlez, Montserrat 2004 Pragmatic Markers in Oral Narrative. The Case of English and Catalan. Amsterdam: Benjamins. Grant, Lynn E. 2010 A corpus comparison of the use of I don’t know by British and New Zealand speakers. Journal of Pragmatics 42(8): 2282–2296. Green, Georgia 2006 Discourse particles and the symbiosis of natural language processing and basic research. In: Betty Birner and Gregory Ward (eds.), Drawing the Boundaries of Meaning, 117–135. Amsterdam: Benjamins.
236
Ad Foolen
Gumperz, John 2001 Interactional sociolinguistics: A personal perspective. In: Deborah Schiffrin et al. (eds.), 215–228. Günthner, Susanne and Katrin Mutz 2004 Grammaticalization vs. Pragmaticalization? The development of pragmatic markers in German and Italian. In: Walter Bisang, Nikolaus Himmelmann and Björn Wiemer (eds.), What Makes Grammaticalization? A Look from its Fringes and Components, 77–107. Berlin and New York: Mouton de Gruyter. Gupta, Antea Fraser 2006 Epistemic modalities and the discourse particles of Singapore. In: Kerstin Fischer (ed.), 243–263. Hansen, Maj-Britt Mosegaard 2008 Particles at the Semantics/Pragmatics Interface: Synchronic and Diachronic Issues. A Study with Special Reference to the French Phasal Adverbs. (Current Research in the Semantics/Pragmatics Interface 19). Oxford: Elsevier. Hasegawa, Yoko 2010 The sentence-final particles ne and yo in soliloquial Japanese. Pragmatics 20(1): 71–89. Heeman, Peter A. and James F. Allen 1999 Speech repairs, intonational phrases and discourse markers: Modeling speakers’ utterances in spoken dialogue. Computational Linguistics 25(4): 1–45. Hellerman, John and Andrea Vergun 2007 Language which is not taught: The discourse marker use of beginning adult learners of English. Journal of Pragmatics 39(1): 157–179. Heritage, John 1998 Oh-prefaced responses to inquiry. Language in Society 27: 291–334. Hentschel, Elke 1986 Funktion und Geschichte deutscher Partikeln: Ja, doch, halt und eben. Tübingen: Niemeyer. Hirschberg, Julia and Diane Litman 1993 Empirical studies on the disambiguation of cue phrases. Computational Linguistics 19(3): 501–530. Holmes, Janet 1995 Women, Men and Politeness. Harlow, Essex: Longman. House, Juliane 2009 Subjectivity in English as a Lingua Franca discourse: The case of you know. Intercultural Pragmatics 6(2): 171–193. Johansson, Stig 2007 Seeing through Multilingual Corpora. On the Use of Corpora in Contrastive Studies. (Studies in Corpus Linguistics 26). Amsterdam: Benjamins. Jørgensen, Annette Myre 2009 En plan used as a hedge in Spanish teenage language. In: Anna-Brita Stenström and Annette Myre Jørgensen (eds.), 95–115. Jucker, Andreas H. and Sara Smith 1998 And people just you know like ‘wow’. Discourse markers as negotiating strategies. In: Andreas H. Jucker and Yael Ziv (eds.), 171–201.
Pragmatic markers in a sociopragmatic perspective
237
Jucker, Andreas H. and Yael Ziv (eds.) 1998 Discourse Markers. Theory and Descriptions. Amsterdam: Benjamins. Kita, Sotaro and Sachiko Ide 2007 Nodding, aizuchi, and final particles in Japanese conversation: How conversation reflects the ideology of communication and social relationships. Journal of Pragmatics 30: 1242–1254. Kirsner, Robert and Vincent van Heuven 1996 Boundary tones and the semantics of the Dutch final particles hè, hoor, zeg, and joh. In: Crit Cremers and Marcel den Dikken (eds.), Linguistics in the Netherlands 1996, 133–146. Amsterdam: Benjamins. Lee, Duck-Young 2007 Involvement and the Japanese interactive particles ne and yo. Journal of Pragmatics 39: 363–388. Ler, Soon Lay Vivien 2006 A relevance-theoretic approach to discourse particles in Singapore English. In: Kerstin Fischer (ed.), 149–166. Levey, Stephen 2006 The sociolinguistic distribution of discourse marker like in preadolescent speech. Multilingua 25(4): 413–441. Levinson, Stephen C. 2003 Contextualizing “contextualization cues”. In: Susan Eerdmans et al. (eds.), 31–39. Liao, Silvie 2009 Variation in the use of discourse markers by Chinese teaching assistants in the US. Journal of Pragmatics 41(7): 1313–1328. Liu, Dilin 2008 Linking adverbials. An across-register corpus study and its implications. International Journal of Corpus Linguistics 13(4): 491–518. Lucy, John 1993 Reflexive Language: Reported Speech and Metapragmatics. Cambridge: Cambridge University Press. Maschler, Yael 2000 What can bilingual conversation tell us about discourse markers?: Introduction. The International Journal of Bilingualism 4(4): 437–445. Maschler, Yael 2003 The discourse marker nu: Israeli Hebrew impatience in interaction. Text 23(1): 89–128. Maschler, Yael 2009 Metalanguage in Interaction. Hebrew Discourse Markers. Amsterdam: Benjamins. Matras, Yaron 1998 Utterance modifiers and universals of grammatical borrowing. Linguistics 36, 281–331. Matras, Yaron 2000 Fusion and the cognitive basis for bilingual discourse markers. The International Journal of Bilingualism 4(4): 505–528.
238
Ad Foolen
Matras, Yaron 2009 Language Contact. Cambridge: Cambridge University Press. Mazeland, Harrie and Mike Huiskes 2001 Dutch but as a sequential conjunction. Its use as a resumption marker. In: Margret Selting and Elizabeth Couper-Kuhlen (eds.), Studies in Interactional Linguistics, 141–169. Amsterdam: Benjamins. McGloin, Naomi and Yumiko Konishi 2010 From connective particle to sentence-final particle: A usage-based analysis of shi ‘and’ in Japanese. Language Sciences 32(5): 507–588. Métrich, René and Eugène Faucher 2009 Wörterbuch deutscher Partikeln: Unter Berücksichtigung französischer Äquivalente. Berlin: Walter de Gruyter. Mey, Jacob 2004 Introduction to special issue on Corpus Linguistics, Part III. Journal of Pragmatics 36: 15–19. Meyerhoff, Miriam 1994 Sounds pretty ethnic, eh?: A pragmatic particle in New Zealand English. Language in Society 23(3): 367–388. Miller, Jim 2006 Particles in spoken discourse. In: Keith Brown (ed.), 214–217. Miller, Jim 2009 Like and other discourse markers. In: Pam Peters et al. (eds.), 315–336. Misˇkovic´-Lukovic´, Mirjana and Mirjana N. Dedaic´ 2010 South Slavic discourse particles: Introduction. In: Mirjana N. Dedaic´ and Mirjana Misˇkovic´-Lukovic´ (eds.), South Slavic Discourse Particles, 1–22. Amsterdam: Benjamins. Möllering, Martina 2001 Teaching German Modal Particles: A Corpus-Based Approach. Language Learning & Technology 5(3): 130–151. Möllering, Martina 2004 The Acquisition of German Modal Particles. A Corpus-based Approach. Bern: Lang. Mulder, Jean, Sandra A. Thompson and Cara Penry Williams 2009 Final but in Australian English conversation. In: Pam Peters et al. (eds.), 337–358. Müller, Simone 2005 Discourse Markers in Native and Non-native English Discourse. Amsterdam: Benjamins. Nevalainen, Terttu 2006 Synchronic and diachronic variation. In: Keith Brown (ed.), 356–363. Niemegeers, Sofie 2010 The Dutch Modal Particle ‘wel’ and its English Counterparts. A Corpus-based Contrastive and Translation Study. PhD, University of Gent. Nølke, Henning 2007 Connectors in a cross-linguistic perspective. Languages in Contrast 7(2): 167–183.
Pragmatic markers in a sociopragmatic perspective
239
Norrick, Neal R. 2009 Pragmatic markers: Introduction. Journal of Pragmatics 41: 863–865. O’Keeffe, Anne and Svenja Adolphs 2008 Response tokens in British and Irish discourse: Corpus, context and variational pragmatics. In: Klaus P. Schneider and Anne Barron (eds.), 69–98. Onodera, Noriko O. 2004 Japanese Discourse Markers. Synchronic and Diachronic Discourse Analysis. Amsterdam: Benjamins. Östman, Jan-Ola 2006 Constructions in cross-language research: Verbs as pragmatic particles in Solv. In: Karin Aijmer and Anne-Marie Simon-Vandenbergen (eds.), Pragmatic Markers in Contrast, 237–257. (Studies in Pragmatics 2). Oxford: Elsevier. Ouafeu, Yves Talla Sando 2006 Politeness strategies in colloquial Cameroon English: Focus on three pragmatic particles: na, ya and eithn. Nordic Journal of African Studies 15(4): 536–544. Peters, Pam, Peter Collins and Adam Smith (eds.) 2009 Comparative Studies in Australian and New Zealand English. Amsterdam: Benjamins. Peterson, Elizabeth 2008 The “invasion” of English? Variation of discourse markers in Finnish. Paper, presented at the Sociolinguistics Symposium 17, Amsterdam. Pichler, Heike 2009 The functional and social reality of discourse variants in a northern English dialect: I don’t know and I don’t think compared. Intercultural Pragmatics 6(4): 561–596. Pons Bordería, Salvador 2008 Do discourse markers exist? On the treatment of discourse markers in Relevance Theory. Journal of Pragmatics 40: 1411–1434. Pusch, Claus 2002 Preverbal modal particles in Gascony Occitan. In: Ton van der Wouden et al. (eds.), 105–118. Romero-Trillo, Jesús 2002 The pragmatic fossilization of discourse markers in non-native speakers of English. Journal of Pragmatics 34: 769–784. Romero-Trillo, Jesús 2006 Discourse markers. In: Keith Brown (ed.), 639–641. Romero-Trillo, Jesús 2008 Introduction: Pragmatics and corpus linguistics – a mutualistic entente. In: Jesús Romero-Trillo (ed.), 1–10. Romero-Trillo, Jesús (ed.) 2008 Pragmatics and Corpus Linguistics: A Mutualistic Entente. (Mouton Series in Pragmatics 2). Berlin and New York: Mouton de Gruyter. Roulet, Eddy 2006 The description of text relation markers in the Geneva model of discourse organization. In: Kerstin Fischer (ed.), 115–131.
240
Ad Foolen
Sakel, Jeanette 2007 Language contact between Spanish and Mosetén: A study of grammatical integration. The International Journal of Bilingualism 11(1): 25–53. Schiffrin, Deborah 1987 Discourse Markers. Cambridge: Cambridge University Press. Schiffrin, Deborah 2001 Discourse markers: Language, meaning, and context. In: Deborah Schiffrin et al. (eds.), 54–75. Schiffrin, Deborah, Deborah Tannen and Heidi E. Hamilton (eds.) 2001 The Handbook of Discourse Analysis. Oxford: Blackwell. Schlieben-Lange, Brigitte 1979 Bairisch eh-halt-f˜ei. In: Harald Weydt (ed.), Die Partikeln der deutschen Sprache, 307–317. Berlin/New York: Walter de Gruyter. Schneider, Klaus P. and Anne Barron (eds.) 2008 Variational Pragmatics. A Focus on Regional Varieties in Pluricentric Languages. Amsterdam: Benjamins. Schourup, Lawrence 1985 Common Discourse Particles in English Conversation: ‘like’, ‘well’, ‘y’know’. New York: Garland. Sewell, Alyson 2009 Qualitative Study of the German Modal Particle ja in Internet Messaging. BA thesis, Ohio State University. https://kb.osu.edu/dspace/handle/1811/37041. Simon-Vandenbergen, Anne-Marie and Karin Aijmer 2007 The Semantic Field of Modal Certainty. A Corpus-based Study of English Adverbs. (Topics in English Linguistics 56). Berlin/New York: Mouton de Gruyter. Slembrouck, Stef, Miriam Taverniers and Mieke Van Herreweghe (eds.) 2009 From will to well. Studies in Linguistics, Offered to Anne-Marie Simon-Vandenbergen. Gent: Academia Press. Stenström, Anna-Brita 2009 Pragmatic markers in contrast: Spanish pues nada and English anyway. In: Anna-Brita Stenström and Annette Myre Jørgensen (eds.), 137–159. Stenström, Anna-Brita and Annette Myre Jørgensen (eds.) 2009 Youngspeak in a Multilingual Perspective. Amsterdam: Benjamins. Strauss, Susan and Xuechua Xiang 2009 Discourse particles: Where cognition and interaction intersect – The case of final particle ey in Shishan dialect (Hainan Island, P.R. China). Journal of Pragmatics 41, 1287–1312. Tagliamonte, Sali 2005 So who? Like how? Just what? Discourse markers in the conversations of Young Canadians. Journal of Pragmatics 37: 1896–1915. Thüne, Eva-Maria and Franca Ortu (eds.) 2007 Gesprochene Sprache – Partikeln. Frankfurt am Main: Peter Lang. Torres, Lourdes and Kim Potowski 2008 A comparative study of bilingual discourse markers in Chicago Mexican, Puerto Rican, and MexiRican Spanish. International Journal of Bilingualism 12(4): 263–279.
Pragmatic markers in a sociopragmatic perspective
241
Traugott, Elizabeth Closs 2007 Discussion article: Discourse markers, modal particles, and contrastive analysis, synchronic and diachronic. Catalan Journal of Linguistics 6: 139–157. Trester, Anna Marie 2003 Bienvenidos a Costa Rica, la tierra de la pura vida: A study of the expression “pura vida” in the Spanish of Costa Rica. In: Lotfi Sayahi (ed.), Selected Proceedings of the First Workshop on Spanish Sociolinguistics, 61–69. Somerville, MA: Cascadilla Proceedings Project. www.lingref.com, document #1008. Urgelles-Coll, Miriam 2010 The Syntax and Semantics of Discourse Markers. London and New York: Continuum. van Balen, Erna, Johanneke Caspers and Ton van der Wouden 2010 Modale partikels in het Nederlands als tweede taal: aanwijzingen voor een vaste verwervingsvolgorde. [Modal particles in Dutch as a second language: indications for a fixed order of acquisition]. Internationale Neerlandistiek 48(2): 31–40. http://www.internationaleneerlandistiek.nl/. van der Wouden, Ton 2002 Particle research meets corpus linguistics: On the collocational behavior of particles. In: Ton van der Wouden et al. (eds.), 151–174. van der Wouden, Ton and Ad Foolen 2010 Adverbs after verb-final in Dutch. Paper presented at the conference on final particles, May 27–28, University of Rouen. van der Wouden, Ton, Ad Foolen and Piet van de Craen (eds.) 2002 Particles. (Belgian Journal of Linguistics 16). Amsterdam: Benjamins. Vandeweghe, Willy 2010 Modale partikels en vertaling. [Modal particles and translation]. Internationale Neerlandistiek 48(2): 19–30. http://www.internationaleneerlandistiek.nl/. Verikaite˙, Daiva 2005 Variation of conjunctive discourse markers across different genres. Man and the Word 7(3): 68–75. Vincent, Diane 1993 Les Ponctuants de la Langue et Autres Mots du Discours. Québec: Nuit Blanche Éditeur. Vincent, Diane 2005 The journey of non-standard discourse markers in Québec French: Networks based on exemplification. Journal of Historical Pragmatics 6(2): 188–210. Vismans, Roel 1994 Modal Particles in Dutch Directives: A Study in Functional Grammar. (Studies in Language and Language Use 11). Amsterdam: IFOTT. Waltereit, Richard and Ulrich Detges 2007 Different functions, different histories. Modal particles and discourse markers from a diachronic point of view. Catalan Journal of Linguistics 6: 61–80. Wang, Yu-Fang, Pi-Hua Tsai and Ya-Ting Yang 2010 Objectivity, subjectivity and intersubjectivity: Evidence from qishi (‘actually’) and shishishang (‘in fact’) in spoken Chinese. Journal of Pragmatics 42: 705–727.
242
Ad Foolen
Wenzel, Veronika 2002 Relationelle Strategien in der Fremdsprache. Pragmatische und interkulturelle Aspekte der niederländischen Lernersprache von Deutschen. Münster: agenda Verlag. Weydt, Harald 1969 Abtönunspartikel: Die deutschen Modalwörter und ihre französischen Entsprechungen. Bad Homburg v.d.H.: Gehlen. Weydt, Harald (ed.) 1989 Sprechen mit Partikeln. Berlin/New York: Walter de Gruyter. Wong, Yin-Ting 2009 The Linguistic Function of Cantonese Discourse Particles in the English Medium Online Chat of Cantonese Speakers. MA Thesis, University of Wollongong (Australia).
9.
Interjections Neal R. Norrick
1.
Introduction
Interjections are a natural pragmatic topic for several reasons. First, they are not intergrated into the grammatical structure of utterances/sentences, so that they cannot be described in purely syntactic terms. Second, interjections are associated with emotions or at least cognitive states of various kinds. Third, interjections act as discourse markers signaling contrast, elaboration and transition. Fourth, they often initiate utterances and relate them to the foregoing interaction. Fifth, interjections frequently occur with exclamative constructions. Sixth, the class of interjections is open-ended; it apparently accepts an unlimited number of new items, but, despite their variability, the pragmatic functions of ever new interjections seem always to be clear to participants in the concrete context. Interjections represent a large, potentially infinitely extendable class of items, unlike the relatively circumscribed, closed classes of other uninflected words, and their functions seem to follow from their general status as expressions of shifts in cognitive states of various kinds. By way of an initial definition, interjections are words or short phrases “which can constitute an utterance by themselves and do not normally enter into constructions with other word classes” Ameka (1992:105). Interjections as traditionally classified include items familiar in their functions as discourse markers, continuers, attention signals, hesitators, expletives etc. (see e.g. Quirk et al. 1985). They are roughly the same as the inserts of Biber et al. (1999), though traditional accounts often exclude greetings, farewells and politeness formulas. Kockelman (2003) includes the whole range of interjections in his purview, not just or even primarily expressions of emotion. He identifies interjections functioning as listener response tokens, discourse markers, attention getters, floorholders, transitions, and markers of self-repair, as well as to signal disapproval and to point out objects or social relations. Wharton (2003) includes for English: wow, yuk, aha, ouch, oops, ah, oh, er, huh, eh, tut-tut (tsk-tsk), brrr, shh, ahem, psst, bother, damn, (bloody) hell, shit (etc.), goodbye, yes, no, thanks, and well. Ward (2006) treats several traditional interjections like oh, mm and yeah in his research on non-lexical conversational sounds. The remainder of this essay is organized as follows. In the next, state of the art, section, I first illustrate the basic types of interjections, then I offer an overview of the literature on interjections. I turn then to the matter of classifying interjections structurally and functionally, primarily following Ameka (1992a). After that, I
244
Neal R. Norrick
present corpus-based statistics on interjections in spoken English, commenting on the most frequent interjections. The next section focuses on yeah to illustrate the range of functions of a typical example. I then move on to a discussion of interjections as pragmatic markers in the sense of Fraser (1996). I show how interjections function as pragmatic markers of various types, in particular parallel markers, that is as words or phrases separate and distinct from the propositional content of the surrounding utterance and signaling a message in addition to the basic utterance message, namely the speaker’s emotional involvement with the foregoing turn or the interaction as a whole. After a brief treatment of interjections as intensifiers in phrases like hell no, which initially seem like parallel markers, but are not, I go on to describe functions of interjections which go beyond those of parallel pragmatic markers. In the next two sections, I treat combinations of interjections and the phrase I tell you as illustrative of the class of idiomatic phrasal interjections. The traditional relation of interjections with exclamative clauses is taken up in the following section. In the penultimate section, I show that interjections must be treated as constituting an open class of items. A final conclusion pulls together the main points of this essay.
2.
State of the art
Although interjections are often characterized as the neglected part of speech, there is actually a fair amount of literature on them (see Ameka 1992; and Nübling 2004 for overviews of research on interjections). There is also fairly general agreement about basic formal and functional characteristics. The disagreement and confusion are due primarily to unsubstantiated claims based on intuitions without real evidence and a failure to recognize the differences between spoken and written discourse. Now that we have reliable, large corpora of spoken language, the problem of evidence versus intuitions seems to be taking care of itself. We finally need no longer guess about distributions and frequencies of linguistic items. Corpus investigations will also show that interjections are characteristic of and ubiquitous in everyday talk while they are usually oddities in written texts which do not seek to imitate speech. Claims that interjections are peripheral elements or non-words emanate from the perspective of written texts with their carefully marshalled grammatical sentences; in the intonation groups of everyday talk, interjections are anything but peripheral. Interjections are constituents of the “grammar of speech”, in the sense of Brazil (1995), as opposed to constituents of sentence-oriented written discourse. In any large corpus of conversation, it is easy to find sequences where every single turn contains an interjection of some kind, as in the passage below from the SCoSE (see the note on data sources at the conclusion of this chapter).
Interjections
(1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Brianne: Addie: Brianne: Addie: Brianne:
Addie:
Brianne: Addie: Brianne: Addie:
245
… they have to take it off from the shoulders. uh-huh. you have to PAY for [all this?] [oh yeah.] you have to pay for alterations. oh my god. it’s all like, shit I haven’t even gotten my shoes yet. that’s another- what thirty forty bucks. o::h. and then you try to go, and get them dyed that [color,] and get them dyed that [yeah.] right? mhm-mhm. oh jee:z.
In line 2, uh-huh functions as a response token; oh yeah in line 4 responds to the foregoing question; the phrasal interjection oh my god in line 6 provides an assessment of the previous statement; in line 8, shit functions as a kind of discourse maker to introduce an assertion; the elongated oh in line 10 is again a discourse marker; yeah in line 13 acts as a response token, while mhm-mhm in line 15 responds to the question tag right pertaining to Addie’s turn in lines 10–12; and oh jeez finally in line 16 registers Addie’s negative assessment of Brianne’s response. Missing from sample passage (1) above are hesitation markers like um, tsk and uh, as in: (2) BRAD:
and u:m, ((tsk)) they’re pretty conservative in their, u:h, specifications,
SBCSAE 016 All three interjections signal that the speaker is engaged in planning his utterance or searching for a particular word. Further, interjections like huh and eh function as questioning elements, either free-standing to indicate that one has failed to understand something about the foregoing turn, as in example (3) below, or in tag position to signal a query, as in example (4).
246
Neal R. Norrick
(3) Margaret: what part? Joe: huh? Margaret: what part of California? LSWEC-AC 111501 (4) Barry: Andrew:
it’s frustrating when there’s too much though eh? yeah so, wh- what’s it like down there at the moment.
WSC DPF026 All these sorts of interjections are common in and familiar from natural conversational contexts. Various other interjections do not generally appear in ongoing dialogue, for instance oops and uh-oh as signals that something has gone wrong, as in example (4), and ssh as a signal to be quiet, as in example (5). (4) Josh oh we forgot to say Grace Steve oops. Becky: uh-oh. Josh: well. LSWEC-AC 120601 (5) Nadia that’s not funny at all. Ayesha no, no, no. Nadia ssh. Marcus Ayesha would do the lizard thing to her, LSWEC-AC 122701 These interjections may actually deserve to be called peripheral to the language system and to language in interaction, in that they really do form separate utterances only paratactically connected to surrounding talk, if, indeed, at all. Due to this separateness from their local discourse environment, it is difficult to determine their distributions and functions, even with the aid of a large corpus, and I will not say anything of substance about them in the following. Before the advent of large corpora, linguistic theory generally considered interjections at the periphery of language and primordially related to emotion. For example, Sapir (1921:6–7) said interjections were “the nearest of all language sounds to instinctive utterance.” Jespersen (1924: 90) writes: “As a last part of speech the usual lists give interjections, under which name are comprised both words which are never used otherwise (some containing sounds not found in ordi-
Interjections
247
nary words e.g. an inhaled f produced by sudden pain, or the suction stop inadequately written tut and others formed by means of ordinary sounds e.g. hullo, oh), and on the other hand words from the ordinary language e.g. Well! Why? Nonsense! The only thing that these elements have in common is their ability to stand alone as a complete utterance; otherwise they may be assigned to various word classes. They should not therefore be isolated from their ordinary uses. Those interjections which cannot be used except as interjections may most conveniently be classed with other particles.” Bloomfield includes among the interjections forms that “occur predominantly as minor sentences, entering into few or no constructions other than parataxis” (1933: 176 ff.). They may be “either special words such as ouch, oh, sh, gosh, hello, sir, ma’am, yes, or else phrases … such as dear me, goodness me, goodness gracious, goodness sakes alive, oh dear, by golly, you angel, please, thank you, goodbye”. Bloomfield classifies minor sentences as ‘completive’ and ‘exclamatory’ and interjections also fall into one or the other category along with minor sentences. For instance, yes and no fall together with other answer phrases into the completive type. Jakobson (1960: 354), too, considered interjections exemplars of the “purely emotive stratum of language.” Even when interjections are no longer characterized purely in terms of emotion, they are still characterized in terms of “mental states.” Like Bloomfield, descriptive grammars often lump formulas like hello and goodbye together with interjections. For instance, Leech et al. write: “Interjections are rather peripheral to language: words like ugh, phew, oh, ah and ouch are linguistically somewhat primitive expressions of feeling, only loosely integrated into the linguistic system. We can include here too swear words (damn etc.), greetings (hello) and other signalling words like goodbye, yes, no, okay etc.” (1982: 53). For example, Ameka (1992a:107) says that “interjections may be defined as a subset of items that encode speaker attitudes and communicative intentions and are context bound,” and Montes (1999:1289) notes that many interjections “[focus] on the internal reaction of affectedness of the speaker with respect to the referent.” In his insightful exploration of self-talk, Goffman (1978) describes a group of “response cries, namely exclamatory interjections which are not full-fledged words. Oops is an example” (99). Response cries like oops, ouch and damn characteristically mark internal states for random unratified listeners or for no one at all, but they extend to include interactional functions such as using ouch to signal recognition of some interlocutor’s pain. Even disfluency markers like uh and um count as response cries, in that they “facilitate tracking” of the speaker’s inner word search, despite the fact that they are more like murmurs than cries. “In effect, speakers make it evident that although they do not now have the word or phrase they want, they are giving their attention to the matter and have not cut themselves adrift from the matter at hand” (109). Goffman’s characterization of certain interjections as cries does not deny that frequent examples are fully lexicalized and differ substantially from one language community to the next. Nor should identify-
248
Neal R. Norrick
ing interjections with cries lose sight of their functional role in a language system (Ehlich 1986: xiv). Wharton (2003) points to the heterogeneous character of interjections and to the disparate treatment they have received. He proposes to amalgamate features of these diverse analyses in placing interjections along a continuum between showing and saying, where showing is relatively natural behaviour, and saying is properly linguistic.
Figure 1. The showing – saying continuum (Wharton 2003: 210)
Still, in the face of such great diversity among interjections, describing the distinctions between subclasses is more to the point than attempts to place the diverse items on a single scale (which itself reflects more than a single parameter). Wierzbicka (1991, 1992) and Wilkins (1992) propose definitions for various interjections using a natural semantic metalanguage. For the English interjection wow, Wierbicka (1992: 164) proposes for explication below. (6) Wow! I now know something I wouldn’t have thought I would know it I think: it is very good (I wouldn’t have thought it would be like that) I feel something because of that But such semantic analysis of interjections misses the pragmatic point: who uses wow, in what context, to what effect? For pragmatic units like interjections we need a description of functions in context, as one finds in Kockelman (2003). For a grab-
Interjections
249
bag category like interjections we need to recognize sub-groups on functional grounds rather than to attempt an analysis which glosses over pragmatic distinctions. Cuenca (2000) adopts a prototype approach to categories, in which interjections are considered a context-sensitive peripheral class of the category “sentence” that typically encode pragmatic meanings. However, since interjections are characteristic of spoken language (with its utterances or intonation units) rather than written language (with its proper sentences), it is more accurate to say they instantiate a (peripheral) type of speech act or utterance type. In particular, interjections are often grouped with exclamatives as items which signal both surprise and either positive or negative emotional involvement. Investigating interjections in the role of response tokens has been a particularly fruitful area of research into variation in the use of interjections. Tottie (1991) compares British English and American English minimal responses based on a very limited corpus: just one single conversation between two speakers for American English and two conversations involving two speakers each for British English, yet she discovers some significant differences in the usage of listener response tokens, including both free-standing responses and those initial in utterance. In particular, Tottie finds yeah to be by far the most frequent response token in American English, whereas yes was a frequent listener response in British as opposed to American English. British speakers use a short monosyllabic m with falling intonation, while Americans use a disyllabic, often lengthened mhm with aspiration between the syllables. These findings hold up very well in later work on listener responses. Heinz (2003) determined that German speakers produce significantly fewer overlapping listener responses and fewer listener responses overall than American English speakers. O’Keeffe and Adolph (2008) found some striking differences between British and Irish English in the frequency of listener response token activity, namely the British speakers used more overall. Tellingly, the Irish used more responses involving religious references and swear words, namely God, oh God, oh my God, Jesus and Jesus Christ, while British speakers used only God and oh God. It has been variously demonstrated that Japanese talk-in-interaction contains more listener responses and more repetition of each other’s words, reflecting degrees of deference and solidarity, than American English (see Maynard 1986, 1990; Tanaka 2004; Kita and Ide 2007). Ide (2001) proposes a scale of contrast based on involvement, with Japanese talk illustrating high involvement versus U.S. English low involvement, in large part because Japanese talk-in-interaction contains more listener activities subsumed under the heading of aizuchi. In addition, overlapping listener tokens are common in Japanese aizuchi, even though they seemingly violate the one-speaker-at-a-time rule and potentially seem impolite, the logic being that overlapping attention signals count as more polite because they cannot be heard as attempts to take over the turn in the way those at transition
250
Neal R. Norrick
relevant places can. A whole range of cultural assumptions affect the forms and frequencies of interjections used as response tokens, so that one could argue that mhm shows more considerateness than uh-huh, simply because it is nasal rather than oral and thus less like a normal turn and more clearly a pure continuer. In parallel fashion, one might maintain that assessments like wow and weird also mark (positive) politeness in signaling involvement, so long as they are heard as expressing appreciation rather than skepticism. Thus, we could look for cultural assumptions and biases throughout the entire system of forms and relative frequencies of interjections used as listener responses.
3.
Classifying interjections
Perhaps the most fundamental classification of interjections is into primary versus secondary interjections. According to Ameka (1992:105), interjections divide into primary interjections like oh and uh and secondary interjections like boy and damn. Primary interjections are called primary because “they are not used otherwise,” while secondary interjections are so called because they are “forms that belong to other word classes based on their semantics and are interjections only because they can occur by themselves non-elliptically as one-word utterances” (Ameka 1992:105). The implication of this view is that secondary interjections are forms that belong to other word classes based on their meanings and function as interjections when they occur by themselves non-elliptically as one-word utterances and in this usage refer to mental acts. Primary interjections are little words or non-words which in terms of their distribution can constitute an utterance by themselves and do not normally enter into construction with other word classes, for example, ouch, wow, gee, oho, oops etc. They could be used as co-utterances with other units. This definition and distinction of primary and secondary interjections captures the scope and characteristics of the class as traditionally understood. Primary interjections often present anomalous phonetic patterns; in many cases, sound sequences functioning as interjections receive spellings that have no regular relationship with their phonetic form, e.g. whew, representing an exhalation of breath, often with a whistled component, and tsk or tut, both of which are supposed to represent a dental or post-dental suction click. Conversationalists use a wide range of sounds (perhaps most frequently inhalations, exhalations and clicks) along with head movements and gestures to signal the beginnings of turns, especially to signal surprise, uncertainty, relief or disagreement. Any of these sounds may function as an interjection of sorts, the contextual significance dependent on the slot in which they appear and the intonation which accompanies them, so that primary interjections constitute an open, in principle unlimited class. Secondary interjections are words or phrases from various lexical classes. The most frequent are yeah, well, okay and hey. In addition, we find interjections from
Interjections
251
nouns such as boy and shit, apparent verbs such as damn and fuck, and phrases like goddamn from goddammit, often in reduced forms such as blimey from God blind me. Of course, secondary interjections from taboo expressions like shit and damn must be distinguished from expletives serving as elements within regular grammatical constructions such as what a load of shit or this damn car, inasmuch as interjections can stand on their own as independent turns at talk. Besides shortening, as we might expect from lexical items frequently used as interjections, various processes of alteration are found, as in goshdarn, gosh, golly etc. from goddamn or gee, jeez, jeez Louise and even jiminy christmas from Jesus (Christ). Due to such shortening processes and mixing of items from different categories in phrases like hot damn, fucking A, gee whiz, holy shit and jesus fucking christ, secondary interjections are often semantically anomalous as well. Like primary interjections, secondary interjections constitute an open class of items. Ameka (1992) further distinguishes routines formulas like hello and goodbye from interjections proper. In doing so, Ameka follows Coulmas (1982: 2–3) in seeing routines as “highly conventionalised pre-patterned expressions whose occurrence is tied to more or less standard communication situations.” Routine formulas such as hi, bye, thank you and you’re welcome are produced more or less automatically in specific situations with specific addressees (note the inclusion of the personal pronoun you in typical formulas of thanking and acknowledging thanks), while interjections (as response cries in Goffman’s sense) mark internal states for random unratified listeners or for no one at all. In the following, I shall exclude such routine formulas from consideration. Similarly, yes and no, in their characteristic function of responding to polar questions, seem too distinct from interjections proper to be sensibly classified with them. Both yes and no are formulaic responses to various speech acts in question form, for instance requests (for information or action), offers, and invitations, but they can also serve as true interjections as in the simulated examples below. (7) A: You just won a car. B: Yes! (8) A: Your cat died. B: No! To avoid confusion, Ameka puts examples like I tell you, goodness gracious and lord have mercy in a special class of interjectional phrases. Following Ameka (1992), we can further distinguish three functional categories of interjections: (1) those that focus on the speaker’s mental state, either on the emotive level like ouch or on the cognitive level like oh; (2) those that focus on the interaction and require some response from the interlocutor like hey; and
252
Neal R. Norrick
(3) those with primarily phatic function like um. Ameka also includes in this class interjections used in interactional routines such as greetings, farewells and welcoming. Such a classification was also used by Taavitsainen (1995) in describing interjections in Early Modern English. O’Connell and Kowal (2005) distinguish disfluency markers like uh and um from interjections proper. They maintain that disfluency markers occur typically in initial position, whereas interjections occur in medial positions; and, further, that disfluency markers do not constitute an integral turn by themselves, whereas interjections do. But research on any large enough corpus of conversation shows, first, that interjections like damn and hell routinely occur in initial and not just intermediate position, and, second, that free-standing uh and um can indeed constitute turns, and so the distinction is at best muddy. Consider as examples the passages below, where damn introduces a turn and where um constitutes a complete turn. (9) 1 Mark: that’s poison oak. 2 Christa: damn I don’t want that on my body. LSWEC-AC (122401) In this next excerpt, um forms a complete turn apparently signaling dissent on the part of Earl. This interpretation is reflected in Doug saying explicitly that the candy will be “for after dinner” in line 4. (10) 1 Madonna: oh, well you can move it in there, put it on the floor. 2 I brought some candy. 3 Earl: um. 4 Doug: for after dinner. 5 Madonna: no need for me to start on things, 6 I’ll get all in a hurry and everything will be ready [before] 7 Earl: [before the] turkey. LSWEC-AC (144801) O’Connell, Kowal and Ageneau (2005) seek to operationalize the category of primary interjections by excluding “all the standard discourse markers”, naming well, and, but, or, so, because, now, then, I mean, y’know, and citing Schiffrin (1982 [sic]: 31). They certainly mean Schiffrin (1987), which appears in their references, but Schiffrin’s list (1987:31) is headed by oh, which O’Connell, Kowal and Ageneau include as a “noncontent form” (2005: 158) among the primary interjections. Clearly, any distinction is difficult to motivate, and I have opted for a more inclusive definition of the class of interjections, following writers from Bloomfield (1933) to Nübling (2004) among many others. Of course, these items may have multiple functions. Moreover, as Ameka points out, many of these functions fall under the heading of pragmatic markers
Interjections
253
(Fraser 1996, Norrick 2009). I will generally follow Ameka’s definitions and classifications in the following. After assigning yes/no to a special class and recognizing that greetings, thankyous etc. belong with the routine formulas, we are left with the following sub-categorization:1 primary interjections: expressive interjections emotive: ouch, wow cognitive: oh, eh? conative interjections sh, pst phatic interjections uh, uh-huh I tend to agree with Jespersen (1924: 90) that “the only thing that these elements have in common is their ability to stand alone as a complete utterance; otherwise they may be assigned to various word classes”. For lexical purposes, the ‘swear words’ can seemingly remain wherever they belong otherwise. Functionally, however, from a pragmatic perspective, the class of interjections draws from various lexical classes and is not just large but always open to new entries. As we shall see, however, there are not just a few swear words involved; any sort of expression even mildly offensive or taboo can serve the purpose of an interjection in the proper slot with appropriate intonation. Indeed, the functional class of interjections is virtually open, as I will demonstrate below. Surprisingly, however, some of the commonest interjections are innocuous words with no offensive connotations like boy, man and nuts. Moreover, interjections are not really independent of other constructions when they act as discourse markers integrated into a single intonation unit with another word or clause, as in: (11) hell yes, shit no damn is that hot, boy are they ever so they’re gone, huh; not bad, eh? At least interjections are bound into discourse grammar, if not into the sentential grammar of written language. As Wierzbicka (1991, 1992) stresses, interjections are not universal. Different languages feature distinct sets of sometimes idiosyncratic interjections for various purposes. She illustrates with interjections from Polish and other languages, dividing them into volitive, directed at animals (e.g. Polish wio to get a horse going) or
254
Neal R. Norrick
humans (e.g. Polish sza for silence), emotives, expressing strong feelings like disgust (e.g. Polish fe for moral disgust), and cognitive, expressing mental states like recalling (e.g. Polish aha meaning ‘now I know it’). She demonstrates that it is possible to produce semantic formulas capturing the “meanings” of all these kinds of interjections. Polish fu, English yuck (yuk) and Russian fu all share the feature ‘I feel something bad’ and ‘I don’t want’, but they differ in that Polish fu expresses ‘I don’t want this to come to be in my nose or in my mouth’, while English yuck expresses ‘I don’t want to be in the same place as this’, and Russian fu expresses more generally ‘I don’t want this’2. Wierzbicka argues that there are even cross-cultural differences in the frequency of emotive interjections, because some language communities (English) circumscribe the uninhibited show of emotions more than others (Russian). Kockelman (2003) offers an account of interjections in Q’eqchi’ Maya that illuminates their social and discursive functions. He discusses the grammatical form of interjections, both in Q’eqchi’ and across languages, and characterizes the indexical objects and pragmatic functions of interjections in Q’eqchi’ in terms of a semiotic framework that may be generalized for other languages. Kockelman argues against interpretations of interjections that focus on internal states, as proposed by Wierzbicka, instead providing an account of their meanings in terms of situational, discursive, and social context. He exemplifies and orders the situational indexical objects of interjections based on the distinctions: object/event, sign-based transposition, and addressee-based transposition, and he exemplifies and orders the discursive indexical objects of interjections based on the distinctions: response versus nonresponse; within responses, solicited versus unsolicited; within solicited responses, preferred and nonpreferred; and within nonresponses, addressed and unaddressed. For instance, the Q’eqchi’ Mayan interjection ih is used to index the speaker’s registering of the addressee’s previous comment, roughly similar to English oh. It functions as a preferred solicited response to another’s utterance in the discursive context. Chaiqin Yang (2004) contrasts German and Chinese interjections and onomatopoeia with regard to phonological, prosodic, graphic, morphological, and syntactic differences. In both languages interjections include emotion words (ah and pfui) onomatopoeic words (peng and plumps), and words directed at animals (like putput for calling chickens) along with secondary interjectionen like oh Gott. Interestingly, in the written form, Chinese interjections use a type of graphic interjection indicator. As a typologically isolating language, in which none of the words are inflected, Chinese can fully integrate its interjections and onomatopoeia syntactically, while German with its rich system of inflections cannot, but these structural differences do not seem to call for differential pragmatic treatment of interjections in the two languages (see also Chaiqin Yang 2001).
Interjections
4.
255
Corpus statistics
Turning now to corpus statistics, based on the American English conversation portion (329 texts, 2,480,800 words) of the Longman Spoken and Written English Corpus (LSWEC), the most frequent interjections are, in descending order: yeah, oh, well, uh-huh, mhm, mm, um, uh, huh, hey, hm, wow, ah, ooh. The absolute numbers are displayed in Table 1. These statistics include only free-standing and utterance-initial interjections, since their occurrence in other positions is too varied and/or infrequent to count significantly. The statistics were determined by means of word searches with reference to markers separating utterances. Table 1.
yeah oh well3 uh-huh mhm mm um uh huh hey hm wow ah ooh
Most frequent initial and free-standing interjections in the Longman Spoken and Written English Corpus – American Conversation total 40652 28380 17789 5730 5325 4000 3803 3608 2222 1767 1520 1261 846 537
free-standing 16617 5009 934 4468 4339 1169 845 536 1701 717 1096 802 283 248
initial 24035 23371 16855 1262 986 2831 2958 3072 521 1050 424 459 563 289
The free-standing interjections generally count as “activities in the back-channel” – ‘by contrast with the primary channel for talk produced by the primary speaker (Yngve 1970); see Tottie (1991) for a corpus-based description of these verbal back-channel activities in English. Related notions like “continuer”, “response token” and “assessment” have been discussed by Schegloff (1982), Goodwin (1986), Gardner (1998, 2001) and McCarthy (2003) among many others. Items which typically occupy the utterance-initial slot have received less systematic attention, but see Tao (2003). Despite differences in how we classify the units in question, my statistics agree generally with other counts of interjections and related items in US-American English conversation (compare the results for all “inserts” based on the whole LSWEC in Biber et al. 1999; and Ward 2006). Like Ward (2006: 132), who also limits his data base to American English, I find yeah to be by far the most frequent item in this category, most usually as a free-standing response token, followed by
256
Neal R. Norrick
oh, which can occur as either a response token or a disfluency marker, while um is primarily a disfluency marker. The high frequency of uh-huh in the LSWEC visà-vis its very low frequency in Ward’s statistics is to be explained by his careful distinction in transcribing of a range of items such as uh-huh, uh-hn, nn-hn and so on. Regarding socio-pragmatic differences resulting from differences in the corpora investigated, yeah is certainly commoner in American English than in British English, and commoner for younger speakers than older ones in both communities.The overall frequency of yeah as a listener response agrees with the results reported by Tottie (1991). Tottie also notes that mhm and uh-huh represent forms typically used as listener responses in American English by contrast with British. Among British speakers of all ages, right is more common in this function than in American English.The high score for wow in American English differentiates it from British usage as well. These findings correspond to what McCarthy (2002) found for these forms. Huh is likewise more frequent in American English than in British English: the free-standing questioning function of huh is filled to a great extent by other items, especially eh and what/wot in British, Australian and New Zealand varieties Interjections can also function as invariable tags, as demonstrated by right in example (1) above. Though not included in my statistics, the tag function of interjections such as huh, yeah, okay and right is very significant in conversational interaction. Stenström et al. (2002) demonstrate the importance of invariable tags in English teenage talk, in particular yeah, innit, right, okay and eh, in decreasing order of frequency. As Andersen (2001) has shown, innit, deriving from isn’t it, has undergone “invariabilization” to become a tag attachable to an utterance, no matter what verb form appears in the main clause. Innit began as a London (Cockney) feature, but has greatly expanded its range among younger speakers in England. Stenström et al. (2002) found that middle adolescents (14–16 years of age) use more tags overall than any other age group, and this is particularly true for the most frequent tag yeah. Taken together with its very high frequency as a free-standing response token and an utterance initiator, as reported below, this makes yeah the most common interjection overall by far. By contrast with the primary interjections in Table l above, secondary interjections occur much less frequently either free-standing or in initial position, including combinations like oh boy, boy oh boy; oh god, oh my god; oh man, man alive etc. For instance, I counted in these contexts:
Interjections Table 2. boy god man shit damn whoa fuck gosh gee jesus hell jeez yuck holy shit golly dammit fucking A
257
Secondary interjections in the Longman Spoken and Written English Corpus – American Conversation 271 271 234 171 160 117 116 113 91 57 43 38 25 18 11 10 4
I will have some general remarks on the most frequent interjections, but my focus will be on these less studied secondary interjections. Of course, the use of items like god, gosh, golly, gee, jeez, jesus, shit, holy shit fuck, fucking A and so on reflects different speaker identities and is sensitive to variables such as age, sex, regional background and social group. Whoa, holy shit and fucking A are only frequent in and perhaps restricted to younger speakers from the United States: whoa as an interjection indicating surprise appears to be a fairly recent development, while holy shit and fucking A are at once rude and jocular. Gosh, gee, jeez and particularly golly strike me as more typical of older speakers. All the rest of the items in the above list except gosh seem to be much more frequent in my data than in the CANCODE results reported by McCarthy (2002), but he presumably excluded very similar items like boy, god and man. A few of the commonest primary interjections function as pragmatic markers, namely oh, mhm, mm, um, uh, huh, wow, hm, uh huh, ah, and ooh; a few secondary interjections routinely function as pragmatic markers as well, namely yeah, well, okay and hey. The more frequent primary interjections have generally taken on specific functions as pragmatic markers, for instance oh and, to a lesser degree, ooh and ah, which have assumed functions as discourse markers signaling a change in cognitive state. It is important to note, following Schiffrin (1987), that much of the interactional significance of these primary interjections derives from their characteristic position as turn initiators, and, following Fraser (1996), that much of their meaning in any particular case depends on their intonation contour. Thus, for instance, huh with a rising contour signals a desire for a repeat of the previous utter-
258
Neal R. Norrick
ance, while huh with a level contour signals some difficulty processing the previous utterance, often perplexity or disagreement. Heritage (1984:299) characterizes the primary interjection oh as a particle “used to propose that its producer has undergone some kind of a change in his or her locally current state of knowledge, information, orientation or awareness”. He says ohs “provide a fugitive commentary on the speaker’s mind” (300) and cites Goffman (1978) as saying they “are taken to index directly the speaker’s state of mind.” This places oh among the standard means of expressing evidentiality rather than emotional involvement. Aijmer’s (1987) characterization of oh as a marker of relevance also underlines its role in signaling the speaker’s response to presuppositions and predictions inherent in the foregoing utterance. In the same vein, Schiffrin (1987) describes oh as a discourse marker within the participation framework of information state, again placing it within the domain of information as opposed to emotion. Based on my investigations so far, the other common primary interjections, mhm, mm, um, uh, huh, hm, uh huh, ah and ooh, adhere to this pattern as well (compare Ehlich 1986 on hm in German); they differ from primary interjections like ouch, which typically express emotion (see Ameka 1992). That is to say, many primary interjections do not express emotions, as is often maintained of interjections generally, but rather information states, as in example (12) below. It is usually up to the secondary interjections like damn, fuck and shit, often with their roots in religion, sex and scatology, to express strong emotions. (12) 1 Brianne: I came on Wednesday night. 2 Addie: mhm-mhm. 3 Brianne: oh, 4 I tried something different this time SCoSE (Addie and Brianne) My statistics only include interjections functioning as pragmatic markers when they head a new turn, as in examples (13)-(15) below. (13) 1 Addie: my mom was saying that Mary had said, 2 that you weren’t there. 3 Brianne: uh-huh y’know uhm4 a friend will marry in Rockford tomorrow. SCoSE (Addie and Brianne)
Interjections
(14) 1 LENORE: 2 3 KEN: 4 5 SBCSAE 015
is it the season, for your turtles to be romantically involved? u:h, I guess so. they’ve been doing it for a while now.
(15) 1 2 3 4 5 LLC 1–1
now what was the other thing I wanted to ask you. i- is is it this year that uh Nightingale goes. uh no next year. um sixty f- four sixty five. sixty five yeah.
259
As reflected in the statistics in Table 1 above, I counted free-standing interjections functioning as listener response tokens separately. As illustrations, consider freestanding uh and uh-huh in the first two passages: (16) 1 SETH: 2 LARRY: 3 SETH: 4 SBCSAE 029
I’m wondering if I should just sketch it out. u:h, it’s two-story though, right?
(17) 1 Addie: and he rents out rooms to kids. 2 Brianne: uh-huh. 3 Addie: I don’t know it’s like eight hundred dollars for the semester. SCoSE (Addie and Brianne) Then consider mm twice in a row as a response token from speaker A in: (18) 1 2 3 4 5 LLC 1–1
whatever time your council meeting is. mm. again I can spend the whole time on them. mm. you see uh I shall get uh scripts from ten assistant examiners.
Based on the American conversation section of the LSWEC, the most frequent turn initiators are, in descending order: yeah, oh, and, well, okay, so, but, mhm, y’know, mm, um, uh, (be)cause, I mean, like, huh, or, hey, hm, uh huh, wow, ah, ooh, anyway.
260 5.
Neal R. Norrick
Yeah as an interjection
Yeah is by far the most frequent interjection functioning as a turn initiator in my sample of spoken American English. Jucker and Smith (1998) also find yeah to be the most frequent discourse marker overall in their corpus of American talk. They write: “The most frequent use of yeah is to acknowledge the receipt of information that is new to the discourse but consistent with current active information” (Jucker and Smith 1998:179). Let us briefly consider its characteristic discourse functions. Like all interjections, yeah can stand alone as a response token, as in the excerpt below. (19) 1 BRAD: 2 TAMMY: 3 BRAD: SBCSAE 016
our blue book usually shows the U:hers. yeah. our older Uhers.
In this passage, yeah appears to function just as Jucker and Smith (1998) describe in acknowledging information new to the discourse but consistent with current active information. In introducing a new turn, yeah may also function as a direct positive response to a question, as in: (20) 1 Christa: you’re at U C S B? 2 Mark: yeah I’m a, uh, graduate student in Anthropology? 3 Christa: uh-huh. LSWEC-AC (122401) And yeah may signal agreement with a statement in the foregoing turn, as seen in the sequence below. (21) 1 Ida: someone’s playing the trumpet over there. 2 Ray: yeah he plays that. 3 he plays tuba and uh all the brass. LSWEC-AC (131102) But yeah also occurs as an initial transition word with no obvious positive response or agreeing function as such, but to be “expressing a general acknowledgment of the previous interactive unit” (Jucker and Smith 1998:181), as in: (22) 1 Nadia: do you have any more complaints for the evening. 2 Marcus: I just feel sick. 3 Ayesha: yeah it’s something you ate ((laugh)) 4 does anyone want any dessert? LSWEC-AC (122701)
Interjections
261
(23) 1 Margaret: well, I think that’s so wonderful she’s interested in that. 2 and y’know, like I said, I’d like to foster it some way, y’know. 3 Nancy: uh-huh. 4 Margaret: that’s why I sent that picture of mother. 5 Nancy: yeah she’s just fascinated by that. 6 Margaret: yeah uh-huh. LSWEC-AC (111401) Further, followed by but, yeah regularly introduces turns denying a presupposition or raising an objection to something, as in the excerpts below, where Marcus denies that all the cake was given away and where Nadia initiates a protest against equally splitting the check in a restaurant. (24) 1 Nadia: how come, how come, 2 I thought, I thought you all gave it to her, gave her the piece. 3 Marcus: yeah but there’s still a little piece left. LSWEC-AC (122702) (25) 1 Lise: but everyone’s was almost the same price. 2 Ayesha: well how much, what price is that? 3 Nadia: yeah but, you guys I’m really poor. 4 Lise: I mean almost the same price, I mean LSWEC-AC (122702) Finally, in the construction yeah right, yeah often ironically signals disagreement with the foregoing turn, as in: (26) 1 Steven:
they’re about between seventy and ninety percent fat … calories from fat. 2 pretty high. 3 Christa: yeah 4 Mark: people fool themselves into thinking the dry roasted are any different 5 Christa: yeah right. 6 Mark: ((laugh)) LSWEC-AC (122401) Notice that Christa at line 5 is presumably disagreeing with people’s foolish assumption that the dry roasted variety are different (namely lower in fat and calories) and not with the assessment by Mark that such people are fooling themselves, hence the laugh from Mark, though the disagreement could aim at either
262
Neal R. Norrick
from a purely structural point of view. Thus, we have seen that yeah fulfills a number of different functions even as it initiates a new turn – a pattern we will observe in many of the other interjections as well.
6.
Interjections as pragmatic markers
Pragmatic markers are words or phrases, separate and distinct from the propositional content of the utterance containing them. They signal different types of potential messages the utterance may convey. Pragmatic markers are linguistically encoded clues which signal the potential communicative force of an utterance. According to Fraser (1996), pragmatic markers fall into four fundamental types: basic markers signal the force of the basic message; commentary markers signal a message which comments on the basic message; discourse markers signal the relationship of the basic message to the foregoing discourse; and parallel markers signal a message in addition to the basic message, specifically, they index the speaker’s emotional involvement with the foregoing turn or with the interaction as a whole. Interjections often function as discourse markers like oh and well or parallel markers like wow and gosh, including markers of agreement like yes and disagreement like uh-uh. According to traditional classifications, the interjections in the most frequent group by my count are: yeah, oh, okay, mhm, mm, um, uh, huh, hey, hm, uh huh, wow, ah, and ooh; from this list one might separate out: yeah and okay, when they function as markers of agreement like yes (also included in this group are some occurrences of mhm and uh huh along with the marker of disagreement uh-uh parallel to no). This would leave just oh, mhm, mm, um, uh, huh, hey, hm, uh huh, ah and ooh as primary interjections among the most frequent turn initiators. The interjection hey is basically a summonsing or attention getting device. Oh is fairly well described already as a marker within the information framework of discourse (see Heritage 1984 and Schiffrin 1987), and several others, namely huh, hm, ah and ooh, also pertain primarily to the information state, signaling some change in the speaker’s cognitive state, while the others, namely mhm, mm, um and uh function within the participation framework of discourse, serving mainly to pass or hold the turn and to fill pauses. Consider a few examples: (27) 1 2 3 LLC 1–1
he can’t feel very comfortable in there. I should think with all that crowd. mm it makes it pretty awkward doesn’t it.
Interjections
(28) 1 REED: 2 DARREN: SBCSAE 046
tell me where it hurts. uh, … still a little bit back here.
(29) 1 DAN: 2 JUDY: SBCSAE 048
what do you do with it. um, we put em in that bag.
263
I should, however, point to a few examples where these pragmatic markers take on a function like the discourse marker well, signaling some sort of difficulty in responding to the foregoing turn. Thus, in the excerpt below, uh does more than simply fill a pause and hold the turn; it rejects an assumption of the foregoing question, namely that ‘from Diana’ is the only applicable category, since some of the packages are for Diana. (30) 1 JUDY: 2 LEA: 3 4 5 SBCSAE 048
are all the rest of these from Diana? uuh, to Diana, and fromfrom Diana.
In a different sort of example, um introduces a response, which goes on with well to signal that the receiver cannot offer a simple answer to the foregoing question. (31) 1 LEA: 2 JUDY: 3 4 5 SBCSAE 048
how’d you open it. um it’s, let’s see it, well, it’s supposed to … open on the side.
Perhaps these more complicated uses of markers like uh and um occur only in direct responses to questions. Finally, the interjection huh functions to form a tag question, as in the passage below.
264
Neal R. Norrick
(32) 1 Joyce: 2 Joan: 3 Joyce:
and so our boss Jane, talked and got her switched up there. but basically everybody else quit though, huh? yeah, yeah. everybody else. LSWEC-AC (151101)
On this function, huh works like the canonical tag didn’t they or like other invariant tags such as yeah, innit, right, okay and eh, as described by Stenström et al. (2002: 165–191). One also routinely finds huh and eh as tags attached to ironic statements, as in example (33) below. (33) George:
Andrew:
you could spend all day doing nothing. it’s a- appalling. it’s a charming attitude, eh? really. you can’t do that at my work.
WSC DPC218 Both primary and secondary interjections can stand alone as complete utterances, generally meant to index an internal state of the speaker. Free-standing interjections are also common as response tokens, for instance the primary interjection forms oh, uh-huh, mhm and wow in (34)-(36) just below, and the secondary interjections man, jeez and holy shit in the passages (37)-(39) further down. (34) 1 FRANK: 2 3 RON: 4 FRANK: 5 RON: 6 FRANK: 7 SBCSAE 019
it was a softball. he had a softball sitting there. [oh]. [if the] sun were this softball, uh-huh. then the size of these other ones would be, something like that.
(35) 1 FRANK: 2 RON: 3 MELISSA: 4 FRANK: 6 SBCSAE 019
no bigger than that. mhm. [wow]. [weand we got] out to … Earth,
Interjections
(36) 1 ALINA: 2 3 LENORE: 4 ALINA: 5 SBCSAE 0006
265
and he’s going to CSUN, because that’s the only place you can go at night. wow. brilliant. he’s been in there since nineteen eighty,
The passages (37)-(39) below illustrate secondary interjections in their role as response tokens functioning as complete turns at talk. (37) 1 Steve: oh my grits. 2 Kalani: oh yeah we need grits. 3 Becky: man. 4 Steve: I forgot about my grits. LSWEC-AC (120602) (38) 1 Mack: I, I don’t know I mean I almost make six figures, 2 I guess that’s upper class. 3 Christabel: jeez 4 Mack: and that’s, that’s like, 5 I make less than I used to make last year. LSWEC-AC (123201) (39) 1 Greg: but anyway, then what, what did it register? 2 Chris: about three. 3 Greg: about three. 4 anyway. 5 Shelley: holy shit. 6 Chris: and when you shower, it’s fifty to sixty times higher than that. LSWEC-AC (157602) Further, in example (40), free-standing hell forms a complete turn, followed by shit as a proper pragmatic marker in initial position.
266
Neal R. Norrick
(40) 1 DN> 2 3 CH> 4 DN> 5 CH> WSC DPC066
so I made this other guy who had got him to come back and do it and um paid him for the parts. hell. shit it was terrible. well do you know those guys have never been back for the spouting.
As in this last example, in the cases we are concerned with here, interjections initiate a new turn, as defined above. In this role, they count as parallel pragmatic markers of cognitive states on Fraser’s (1996) definition. The interjection wow provides characteristic illustrations of this parallel marker function in the next two sample excerpts. (41) 1 WALT: 2 3 JILL: SBCSAE 028
y’know, get in tune with him- … with himself. wow did he want you to go with him?
(42) 1 Brandon: 2 Lydia: 3 Brandon: SCoSE (Jack)
with two bodyguards to protect him. wow to think [of it.] [to see] a person in that position.
Note also the pair oh wow, a frequent combination of interjections, in the excerpt below. (43) 1 AG> 2 BA> WSC DPC045
u::m four altogether including myself. oh wow and you’re the youngest.
Oh often occurs with – and always precedes – other interjections. In the example below, we see oh with man in initial turn position. (44) 1 Brianne: Sheila was so mad. 2 Addie: oh man not to trust your employees at all= 3 Brianne: =yeah. SCoSE (Addie and Brianne) Compare another example with man alone as a turn initiator:
Interjections
(45) 1 ALICE: 2 3 MARY: 4 ALICE: SBCSAE 0007
267
and they ended up having to take um … Peggy White … by helicopter to Billings. man that’s pretty bad. I know.
The interjections jesus and god work in the same way, as emotionally loaded turn initiators. (46) 1 2 3 LLC 5–11
not a proper one something like that. jesus that must have come as a surprise. that must be not a proper one.
(47) 1 2 3 LLC 1–10
because they go at six fifty a pop. ((laughs)) god I know.
Of course, we find classic shit and oh shit as emotional outbursts and pragmatic markers. Note that in the first example, shit has positive emotional force with a following positive assessment as in (48), while it conveys negative emotional force in example (49) when coupled with a negative response. (48) 1 BH> 2 AG> WSC DPC331
I was just going oh wow congratulations and= =SHIT that’s great.
(49) 1 DN> 2 3 CS> 4 DN> WSC DPC273
yeah, cos um mine um are still going mouldy now that, it’s in the new place. shit you’re kidding. so I’m going to get some of that Damprid see if that helps.
Similar examples occur with fuck: again the interjection conveys positive emotion when coupled with a positive statement, as in:
268
Neal R. Norrick
(50) 1 Matt: you know what makes a really good dip? 2 you melt like a block of this. 3 and a can of chilli. 4 and put it in a bowl. 5 and serve chips a with it. 6 David: [oh yeah, people go crazy over that.] 7 Matt: [really g … peop- people think ((unclear))] 8 David: fuck it’s Velveeta and chili ((laugh)). LSWEC-AC (129102) But the clause-initial interjection fuck conveys negative emotion when coupled with a negative statement, as in (51): (51) 1 ME> 2 3 4 PQ> 5 ME> WSC DPC251
forty six prone. and forty seven … sitting. and forty six kneeling.= =oh yeah.= =fuck I only got forty three kneeling.
Here the interjections act as parallel pragmatic markers in the sense that they signal a message in addition to the basic message of the turn they initiate. Specifically, they index the speaker’s emotional involvement with the foregoing turn or the interaction as a whole. Moreover, interjections serving as pragmatic markers have identifiable meanings in context, and these meanings may be attended to by conversationalists, as can be seen in agreeing responses like man is right in excerpt (52) below. (52) 1 Roberto: see that’s, that’s why I asked how much he owed, 2 because if he owes like two thousand, man. 3 Robert: man is right. 4 Roberto: see but, shoot … LSWEC-AC (139101) Here in line 3, Robert picks up and comments on the appropriateness of man as an utterance final pragmatic marker in the foregoing turn. Consider a second example (53), where a participant comments specifically on the form of an interjection functioning as a response token.
Interjections
269
(53) 1 Freda: was Chris married in ninety-three. 2 Bud: oh lord. 3 Carol: ((laugh)) 4 (( ? )): yes. 5 Carol: oh lord is right. 6 that was the year of all the weddings. LSWEC-AC (145201) Carol says oh lord is right in line 5, ratifying Bud’s comment oh lord from line 2. Clearly, participants in conversations may specifically orient to the form of an interjection in a previous turn, commenting on its appropriateness. And this underlines the fact that these interjections bear identifiable messages besides just signaling emotional involvement.
7.
Interjections as intensifiers
Before we go on to consider interjections in functions besides those of parallel pragmatic markers, we must have a brief look at some apparently similar constructions. When interjections in initial turn position precede a bare yeah and no, they apparently function as intensifiers of the following items rather than pragmatic markers. Consider: (54) 1 Lana: did you burn anything? 2 Kevin: hell yeah. 3 Lana: oh good. 4 ((laughs)) SCoSE (Football) Here hell acts not as a parallel marker with an independent meaning, but instead forms a unit with the following yeah. The two words are pronounced as a single intonation unit with the primary stress on hell. Shit in passage (55) below similarly intensifies the following no. (55) 1 DN> 2 CE> 3 DN> 4 CE> WSC DPC238
any, any er improvement on seeing client base. °shit no°. no?= =he’s all up and down.
270
Neal R. Norrick
Interjections in these combinations with the appropriate intonation contours clearly function as intensifiers, and I have not included them in my statistics as pragmatic markers. When hell yeah introduces a new complete turn, as in the next example, the two words again form a unit consisting of an intensifier and a head word and functioning together as a pragmatic marker. (56) 1 ML> 2 IB> WSC DPP009
six months, yeah.= =hell yeah I’d better be careful actually.
Fuck yeah works the same way, as the example below demonstrates. (57) 1 AW> 2 BH> 3 WSC DPC032
Flying Man was a good horse. fuck yeah it placed third in the derby in his year. and fourth in the Great Northern Derby.
When initial oh appears forming a three word unit, as in examples (58)-(59) below, the secondary interjection hell still receives the primary stress, and again functions as an intensifier, while oh is a proper pragmatic marker. (58) 1 PT> 2 DV> WSC DPC303
some rude looks I suppose.= =oh hell yeah um.
(59) 1 JO> 2 3 BV> WSC DPC255
I said like, that I could identify with it. but I don’t think the guys could as much. oh hell no.
Consequently, we can say that in combination with yeah and no secondary interjections act as intensifiers rather than pragmatic markers. In passing we should note that no shit represents a completely separate structure. No shit is not a phrase like shit no or hell yes, but rather an idiom within the information state of discourse. With level stress it functions as a kind of ironic response to a turn with information already very familiar to the listener, similar to the ironic use of yeah right mentioned above.
Interjections
271
(60) 1 Chris: the omelet cost the same exact thing. 2 and … I got more. 3 David: well, no shit. 4 you got a hell of a lot more. LSWEC-AC (129601) In this function, it also occurs in the extended form no shit Sherlock. With a rising intonation contour, no shit typically responds to surprising new information, as in excerpt (61) below, though it can also as an ironic commentary on information already familiar. (61) 1 Ted: I we have the pleasure of watching a um remote-control aircraft. 2 Jesse: no shit? 3 Ted: buzz around, yes. LSWEC-AC (164801) In these information receipt functions, the combination no shit is related to idiomatic items such as bullshit, don’t bullshit me, are you shitting me? you wouldn’t shit me, would you? and so on.
8.
On beyond parallel markers
Interjections often act as parallel markers, but they can also serve other functions as well. Thus, hey is generally described as an interjection used for summonsing or an attention-getting device, as in: (62) 1 Ann: oh are you going to drink that out of the bottle? 2 Catherine: I always drink them out of the bottle. 3 Anna: hey Jenn there’s a diet coke out if you want one. 4 Jennifer: um, yeah I’ll have a diet coke. LSWEC-AC (122001) Here Anna seeks to obtain Jenn’s attention with hey in line 3, and Jenn immediately responds to the offer of a diet coke. On this function, hey too counts as a parallel marker, because it realizes a meaning independent of the turn it initiates. But hey can also serve to switch and refocus the topic of conversation, as can be seen in the excerpt below.
272
Neal R. Norrick
(63) 1 Cooper: can I have a bite of that cookie? 2 Sara: hey they’re low calorie. 3 you can have the whole cookie. 4 Cooper: thank you. LSWEC-AC (115301) In this example the summonsing and attention-getting functions are absent: hey redirects the focus of the exchange from a request for a bite of cookie to an offer of a whole cookie, because it is low in calories. Again in this next example (64), hey in line 4 works not to summons or to gain attention, but to move the focus of the discussion from the age of the child in question back to the previous point above enforcing rules. (64) 1 Geoff: we haven’t enforced it. 2 she’s been watching Christmas shows. 3 Sherie: oh well, she’s a two year old. 4 Geoff: hey a rule’s a rule no matter what age you are. 5 Sherie: oh my god, I’m so sure. LSWEC-AC (143601) Similarly, the interjection hell can serve not only as a parallel marker of emotional involvement, but also as a marker of contrast or elaboration, as in sample passage (65) below, where natural paraphrases might be nevertheless or still, respectively discourse markers of contrast and elaboration in the terms of Fraser (1996): (65) 1 Jennifer: he is a terrible human being. 2 he was just like this old crotchety man. 3 Christopher: hell he knew a lot about [soccer though]. 4 Jennifer: [he knew a lot] about soccer. LSWEC-AC (130802) Consider also the interjection shit in line 3 of the next example as a marker of contrast and refocusing on a related topic. Here what’s more or moreover would be natural paraphrases. (66) 1 John: I mean, the, y’know, 2 the penis is uh not shown that often in American culture. 3 Marcia: shit you can’t even say it. 4 Bruce: well I mean the whole thing. LSWEC-AC (155502)
Interjections
273
Compare also fuck in example (67) below, where the most natural unflavored substitute would be the standard contrast marker but. There is an element of surprise and even topic switch here as well. (67) 1 PQ> 2 RT> 3 PQ> 4 RT> 5 PQ> WSC DPC313
where’s Foodtown. in Lower Hutt. oh. I mean, you can justfuck that’s miles away.
If interjections are initially expressions of a change of cognitive state, then their grammatical extension to markers of contrast or transition seems rather natural; a sudden shift in cognitive state can be taken to justify a switch of perspective or topic. In the next example, fuck even more clearly signals transition to a new topic. Here by the way would be a natural unemotional choice for substitution: (68) 1 JU> 2 3 AH> 4 JU> WSC DPC012
((tsk)) yeah Lambda Lager. I quite liked it. ((clears throat)) fuck I haven’t been up Kaukau for ages. no.
Finally, an example where the current speaker self-selects using god to mark a sudden topic switch, as if initiating a new turn. In passage (69) below, notice the addition of though at the end of the clause, perhaps to signal some sort of coherence with an earlier topic for the present audience. By the way, FB’s turn initial fuck in the previous turn seems to me to be just a typical parallel marker use of an interjection. (69) 1 FA> 2 FB> 3 4 FA> 5 6 WSC DPC201
yeah oh oh bit about just (( )) fuck you guys are getting into this going out to dinner quite a bit. aren’t you spending quite a bit of money. oh, I’m spending far too much money. that’s my problem. GOD I can’t believe the COPS were at the hog though ugh.
A particularly telling example of interjections serving to redirect topical conversation is their deployment to initiate turns announcing a narrative. In the excerpt below, Suzanne begins a story at line 3 suggested by the current topic, with the interjection wow functioning as a pragmatic marker. Associated as they are with high
274
Neal R. Norrick
involvement in the sense of Tannen (1984), interjections provide a natural justification for tellability and a claim to the floor for a storytelling performance. (70) 1 Jacques: yeah … well, it’s like if you see your kid bleeding, 2 it’s totally different than if you see someone else. 3 Suzanne: wow y’know, 4 one day Leah walked through the back door. 5 and she had a gash like this in her leg here. 6 and I thought, well come on, Leah, sit down, 7 so I cleaned it out with alcohol and she’s yelling, y’know, 8 and then I put some of these stretch tapes across. 9 that pulls the skin together so it wouldn’t leave much of a scar, 10 and once I got involved in the mechanics of that, 11 I forgot all about the, 12 Jacques: yeah. LSWEC-AC (117601) As markers of emotional involvement and signals of transition, interjections provide a natural way to announce a storytelling performance and to gain the floor. To sum up this section, we have seen that secondary interjections can do more than act as parallel pragmatic markers. They routinely signal contrast and transition as well, and as such they serve to segue into storytelling performances. 9.
Oh no, more combinations of interjections
Interjections occur in various more or less formulaic combinations, as we have seen in cases like hell yeah and no shit, and also early on with the pairs yeah but and yeah right. By far the most frequent first element in pairs of interjections is oh. As pointed out above, oh is always the initial element when it co-occurs in combinations with other interjections. Formulaic to varying degrees are the constructions oh yeah, oh no and oh well. I will investigate these pairs in turn; then, for purposes of comparison, I will explore combinations with the interjection-like phrasal pragmatic marker I tell you. The combination oh yeah with the main stress on oh and a falling intonation contour may just be an intensified variant of yeah to respond in the affirmative or to agree to a request, as in excerpts (71)-(72). (71) 1 Doug: you have a thermostat on it? 2 Earl: oh yeah. 3 all you have to do is turn it over all the way, 4 but it still goes off. LSWEC-AC (144801)
Interjections
275
(72) 1 Gail: could you do that? 2 Christa: [oh yeah, okay.] 3 Gail: [that would be great.] LSWEC-AC (119201) In responses, oh is a signal of surprise, according to Aijmer (1987), and it often retains this force in combination with other interjections. Thus, in the passages (73)-(74) below, the combination oh yeah indicates that the speaker has just remembered something or has been reminded by another speaker. (73) 1 Paul: I’ve seen Star Wars god knows how many times. 2 I still watch that. 3 Jeff: speaking of that, 4 you have my Indiana Jones tape. 5 Paul: oh yeah, 6 if I would have remembered that last week, 7 I would have got it. LSWEC-AC (111801) (74) 1 Todd: what did she have? 2 Karen: your hot bottle. 3 Todd: oh yeah I let her get away with that this morning because, 4 she wanted something 5 and she was content with that so … I didn’t complain. LSWEC-AC (132901) By contrast, with a rising intonation contour, oh yeah generally indicates that the speaker questions or challenges something in the foregoing turn, as in excerpts (75)-(76). (75) 1 Chuck: Steve’s middle name is Steve, 2 Rick: oh yeah? 3 Chuck: David Steve Watson. 4 Rick: hm, I didn’t know that. LSWEC-AC (139901)
276
Neal R. Norrick
(76) 1 Mary Ellen: he’s making a decent amount of money. 2 Jack: oh yeah? 3 it’s also a regular rise with all the faculty who makes what? 4 I mean. 5 Mary Ellen: no, you negotiate every time you get a new position. 6 the department’s head negotiates with the dean … LSWEC-AC (147801) This oh yeah structure occurs as a response token, especially during storytelling to demonstrate that the listener appreciates the tellability of some particular point in the narrative, as in: (77) 1 Joe: I grew up in Ventura. 2 Margaret: oh yeah? 3 Joe: yeah, up until I was about fourteen. LSWEC-AC (111501) The construction oh no roughly parallels oh yes in many ways. The combination oh no with a falling intonation contour is often just an intensified variant of no to strongly disagree or refuse, as in examples (78)-(79): (78) 1 Stu: I thought you were going to say conversations with clients, 2 I was gonna go oh no way, man. 3 Joyce: oh no can’t do any telephone stuff, they, they want [y’know.] 4 Stu: [okay, yeah, that’s great.] LSWEC-AC (117901) (79) 1 Mary Ellen: it couldn’t be that bad, you kidding? 2 Jack: oh no I don’t think so. 3 I mean it’s. 4 Mary Ellen: yeah, um. Jack: I think you have to use reflexivity in English for that idea though. LSWEC-AC (147803) Parallel to the oh yeah construction with the main stress on the yeah and a level contour to indicate remembering in consonance with something in the context, the construction oh no with a level contour indicates recognition of an error in remembering, as in:
Interjections
(80) 1 BERNARD: 2 3 FRAN: 4 5 BERNARD: 6 7 8 FRAN: 9 BERNARD: SBCSAE 051
277
and then, … the Lower East Side. West … Tenth. well, that’s not the Lower East Side. West Tenth is the [Village]. [oh no] wait. not West Tenth, it must’ve been, forget [my last address]. [East Tenth]? yeah. East Tenth.
This functions as a kind of repair, in the sense of Schegloff, Jefferson and Sacks (1977), in cancelling a foregoing proposition. As such, oh no with a level contour naturally occurs as an initiation to other sorts of repair sequences, as in line 3 of excerpt (81) below, where Alina corrects Lenore’s phrase bar generation from line 2, replacing it with her own phrase pub generation in lines 4–6. (81) 1 LENORE: 2 3 ALINA: 4 5 6 7 LENORE: 8 ALINA: SBCSAE 0006
the ba:r generation? in what sense bar generation. ((swallow)) oh no, pub. the pub. [the pub generation]. [((laughs))] there you go.
Parallel to the oh yeah construction with a rising contour to question a foregoing turn, the combination oh no with a rising contour serves to question or challenge a negative statement, as in the following sequence. (82) 1 Deborah: yes a child. 2 intelligent and fairly well behaved. 3 I’m not big on kids. 4 Michelle: oh no? 5 Deborah: no. 6 Jennifer: it made it so smoky though last night. 7 Deborah: I’ve been told treat them like a small adult. LSWEC-AC (123401)
278
Neal R. Norrick
More interesting is the combination oh no with level stress across both elements to signal empathy with the person(s) in the situation described in the foregoing turn, often as a response token by a person listening to a story, as in the next example. (83) 1 Brianne: I have to take off work [take a holiday.] 2 Addie: [oh no.] 3 Brianne: so I can come back home with Mike on Friday. 4 Addie: yeah. SCoSE (Addie and Brianne) In this function, oh no in line 5 of excerpt (84) below does not differ significantly from other two-interjection combinations like oh jeez in line 2 or oh god in line 7: (84) 1. Brianne: it took me an hour to get out to O’Hare. 2. Addie: oh [jeez.] 3. Brianne: [and then] I was standing on the aisle for an hour. 4. ’cause it’s packed. 5. Addie: oh no. 6. Brianne: and7. Addie: oh [god.] 8. Brianne: [then-] okay, we left on time. SCoSE (Addie and Brianne) Finally, the amalgamation oh well has acquired formulaic status with the function of signaling resignation, as in the next example passage. (85) 1 WESS: 2 JO: 3 WESS: 4 5 JO: 6 FRED: SBCSAE 059
it just starts running out. oh= he [is something]. [oh well. live and] learn. ((laughs)) ((laughs))
Here the proverb live and learn in line 4 serves to formulate the resignation expressed by oh well, and to elicit laughter from the other two participants, thereby closing this segment of the conversation.
Interjections
10.
279
I tell you as a phrasal interjection
The phrasal unit I tell you constitutes a recurrent formula, which acts as an interjection. I tell you would count as a secondary interjection for our purposes, and it often functions as a pragmatic marker in the ways we have been observing. It is a syntactically odd, performative-looking construction with a topic-shifting and/or focusing function. While I tell you can occur by itself in both initial and final position in a turn, as in the two examples below, it frequently also occurs following other secondary interjections, as we shall soon see. (86) 1 Cooper: well, it just means, 2 that you have to remember to turn the stupid thing on all the time, 3 constantly. 4 Sara: I tell you if we ran it right in the hallway there at the reception area, 5 we could [get some interesting] 6 Cooper: [well,] 7 Sara: chit chat. LSWEC-AC (115001) In initial position, in line 4 of this first example, I tell you serves to re-direct the topic of conversation from the recording apparatus itself to the potential results of recording in a particular location. At the end of a turn introduced by the interjection jeez, functioning as a parallel pragmatic marker, the phrasal interjection I tell you in line 4 serves to highlight the preceding statement. (87) 1 Betty Jo: so how many, how far has it got to now? 2 Jimmie: I don’t know. 3 Ronald: there’s seventeen hundred and seventy yards in a mile. 4 Jimmie: jeez, you figured that out well, I tell you. 5 Ronald: so it’s a little bit less. LSWEC-AC (152902) Frequently, the phrasal interjection I tell you appears with some other introductory interjection like boy, man or well, as, for example, initially in line 4 of example (88) and finally in line 2 of example (89) below. (88) 1 Dallas: so what’s with these power bars? 2 Bob: have you tried one of those? 3 Dallas: no I haven’t. 4 Bob: boy I tell you you only want to eat about a fourth or a half of them. 5 Dallas: why? LSWEC-AC (141801)
280
Neal R. Norrick
(89) 1 Sara: that was weird the other day hanging out with Stan all day. 2 Cooper: well, he was in the baddest mood, man I tell you. 3 Sara: he’s a changed mother fucker these days. LSWEC-AC (115001) Whether in initial or final position, the combination of I tell you with other interjections seemingly underlines the focusing effect and the implication of emotional involvement. Along with well in the final example (90), I tell you serves to redirect the topic of conversation as well as signaling emotional involvement. (90) 1 Jane: no, she says there’s no taste to reduced fat stuff. 2 Albert: oh. 3 Jane: ((laugh)) 4 Albert: well I tell you you buy some sometime and don’t tell her, 5 and she’ll eat it and say they’re good. LSWEC-AC (167101) In these passages, the combination of interjections redoubles the force of the individual items as pragmatic markers. 11.
Interjections and exclamatives
Due to their expression of emotion, interjections have often been related with exclamative clauses. Subject-Auxiliary Inversion (SAI) is one standard index of the exclamative clause type. Consider two examples of exclamative clauses preceded by an interjection, in both cases boy, which is the most frequent interjection preceding exclamatives with SAI so far in my corpus research. (91) 1 Amy: we just saw a police car using a siren. 2 Pat: boy did we ever. 3 somebody was going out on a call. SCoSE (Mary at home) (92) 1 DORIS: re[member]? 2 SAM: [o:h] ye:[s]. 3 DORIS: [oh boy] was I soaked. SBCSAE 0011 In these examples, first boy then oh boy heads an exclamative construction marked by SAI. Other secondary interjections like god occur with exclamatives with SAI as well, as the example below demonstrates.
Interjections
281
(93) 1 Judy: good, I’m going to get my hot glue. 2 Judith: (( )) 3 Julie: god is it cold outside. 4 Judy: is it? 5 it’s suppose to get colder isn’t it? 6 Julie: yes. LSWEC-AC (145402) SAI is, of course, standard in questions, but it also characteristically indexes exclamative constructions. This inversion of standard word order instantiates one type of exclamative sentence and is itself a marker of emotional involvement; for Fraser (1996) it would count as one kind of basic pragmatic marker. Since interjections are essentially expressions of sudden shifts in cognitive states, their occurrence with exclamative clause types might be expected. Interjections also occur with other sorts of exclamative sentence types, for instance constructions beginning with wh-words, as in (94)-(95) below. (94) 1 Susan: one year you’ve known me? Two years. 2 Janine: boy how time flies. 3 Mark: I feel like I’ve known you five, six. LSWEC-AC (112102) (95) 1 Steve: I’m going to show you where there is a pair of pitiful feet. 2 look at those feet. 3 Becky: whoa Joshua what dirty feet. 4 Steve: whew let me see that face I want to see pitiful face … aargh. 5 Becky: that is not a pitiful face. LSWEC-AC (120601) It is natural that parallel markers of emotional involvement should occur along with different exclamative clause types to underline the emotive force of the turn as a whole.
12.
Interjections as an open class
I have already mentioned the wide range of secondary interjections and the tendency to modify and combine them in phrasal forms like: holy smoke and fucking A. We should also note the tendency for certain interjections to generate a whole series as in holy jesus, holy christ, holy mother of god, holy moses, holy smoke, holy cow, holy mackerel, holy shit, holy moley and so on, but also good god, good lord,
282
Neal R. Norrick
good golly, good heavens, good gracious, good grief and hot dog, hot diggety, hot ziggety, hot diggety dog, hot damn, hot diggety damn, hot shit and so on. Such variation ensures the development of ever new interjections. In addition, there are also any number of formulaic phrasal interjections such as boy oh boy, son of a gun, I tell you, lord have mercy (lord-a-mercy), goodness gracious, land sakes alive, and even oh my goodness along with jiminy christmas in the excerpt below. (96) 1 Armrel: is this Debbie’s? 2 Debbie: yeah, that’s fine, you’ve got to hear Vicki’s too. 3 Vicky: okay. 4 Armrel: oh my goodness, you’re busy, jiminy christmas. LSWEC-AC (133003) Certainly part of the force of interjections as markers of emotion derives from their variability. Indeed, the class of interjections must always remain open to fresh entries and new combinations. By way of demonstrating the open-ended nature of the class of interjections, consider two final examples, first (97) with goddamnation then example (98) with oh bloody hell. The first appears in the LLC and the second in the WSC, and I daresay neither form would be found in North American English, although, of course, there is always the possibility of creating new interjections like fucking A or shit mongers – these last seem limited to younger generations in the U.S. (97) 1 2 3 4 LLC (1–1)
the the other the other the other the other man um, who (( )) I thought was going to get you wild was Potter. goddamnation I’ll crown that bastard, before I’m finished with him.
(98) 1 TM> 2 3 SU> WSC DPC013
yep Europe. ((television is turned up and a news item is on)) OH BLOODY HELL another oil slick.
A friend of mine routinely uses boogers as an interjection in all the standard contexts, and though it strikes listeners as idiosyncratic, they never seem to have any difficulty understanding it. Despite their infinite variability, the pragmatic functions interjections realize seem always to be clear to participants in the concrete context. Interjections thus represent an infinitely extendable class of items – or even two open classes of both primary and secondary interjections – unlike the
Interjections
283
relatively circumscribed, closed classes of other pragmatic markers, and their pragmatic marker functions follow from their general status as expressions of shifts in cognitive states.
13.
Conclusions
Interjections, even primary interjections, are not “linguistically somewhat primitive expressions of feeling, only loosely integrated into the linguistic system” (Leech et al. 1982: 53). First, many primary interjections function in the participation and information frameworks of discourse, rather than marking emotional involvement. Second, interjections are fully integrated into the system of everyday spoken language. Third, interjections are thoroughly lexicalized, often with a variety of functions. Particularly secondary interjections like boy and shit exhibit complex patterns of distribution and function; thus, boy occurs as a response token, an initial parallel marker of emotional involvement, an initial pragmatic marker signalling topic shift, in exclamative constructions, especially those exhibiting SAI, and along with the focusing phrase I tell you; while shit appears as a response token, an initial parallel marker of emotional involvement, an initial pragmatic marker signalling topic shift, an intensifier in shit no, shit yeah, and as a response marker in the information framework in no shit. They have identifiable meanings in context, as can be seen in agreeing responses like man is right. We reviewed relevant literature concerning initial position in turns, and we found that turn-initial position has special discourse significance for various reasons from a wide range of different linguistic perspectives. Corpus data confirm the importance of initial position in turns and demonstrate the frequency of interjections serving as pragmatic markers in this position as well as their functions as response tokens. Any item in the initial position in a turn tends to take on pragmatic marker functions. This holds not only for traditional primary and secondary interjections like ooh and boy and their variants and extensions like goddamnation and oh bloody hell, but also for imaginative combinations like land sakes alive and heavens to Betsy. Further, any item which routinely appears in initial turn position tends to expand its range of functions, so that interjections as pragmatic markers tend to be multifunctional. We saw that these multiple functions are often differentiated by separate intonation contours. Interjections often function as parallel pragmatic markers in turn-initial position, but we also observed various functions for interjections beyond those of parallel markers, namely typical discourse marker functions of signalling contrast, elaboration, transition and so on. We inspected a wide range of functions for frequent interjections. Moreover, a few secondary interjections in initial turn position function as intensifiers rather than pragmatic markers in combinations like hell yeah and shit no. We saw that interjections as parallel pragmatic markers fre-
284
Neal R. Norrick
quently appear with exclamative constructions, doubly marking the emotive force of the turn. Finally, we inspected the distribution and functions of the formula I tell you as an example of a phrasal unit functioning as a pragmatic marker. Interjections are too complex and multifunctional to be sensibly listed among the specific classes of pragmatic markers. The open-ended nature of the classes of primary and secondary interjections makes it impossible to list them in any case. It is more expedient to treat the interjections as sui generis classes with various functions generally emanating from their status as (certain kinds of) interjections along with various more or less formulaic meaning/functions.
14.
Data Sources
The data cited in this investigation derive from five different corpora of transcribed conversation. First, there is the Saarbrücken Corpus of Spoken English (SCoSE), an extensive collection of audio and video recordings of free conversation and conversational interviews, involving a wide range of speakers from the United States and Britain. Notes on our transcription conventions and on participants in the recordings, along with steadily increasing numbers of transcribed excerpts from the SCoSE are available online at: http://www.uni-saarland.de/fak4/norrick/sbccn.htm. Other corpora used are: Santa Barbara Corpus of Spoken American English (SBCSAE) http://www.talkbank.org/media/conversation/SBCSAE/ The London-Lund Corpus of Spoken English (LLC) http://khnt.hit.uib.no/icame/manuals/LONDLUND/INDEX.HTM The Wellington Corpus of Spoken New Zealand English (WSC) http://khnt.hit.uib.no/icame/manuals/wsc/INDEX.HTM The Longman Spoken and Written English Corpus (LSWEC) developed for the Longman grammar of spoken and written English, by Douglas Biber, Stig Johansson, Geoffrey Leech, Susan Conrad, and Edward Finegan (Pearson Education, Harlow, 1999), and the Longman student grammar of spoken and written English, by Douglas Biber, Susan Conrad, Geoffrey Leech (Pearson Education, Harlow, 2002). Statistics and examples derive in particular from the section containing American English conversation (329 texts, 2,480,800 words), signified by LSWEC-AC; the section containing British English conversation (3,929,500 words) was used only for comparison. I accessed the LSWEC at the Corpus Linguistics Research Program administered by Doug Biber at Northern Arizona Uni-
Interjections
285
versity in Flagstaff in the spring of 2007 and again in 2008. I gratefully thank Doug for his help and the opportunity to take advantage of this rich data source. Excerpts from these other corpora have in some cases been partially adapted to the transcriptions conventions of the SCoSE.
15.
Transcription Conventions
Each line of transcription represents spoken language as segmented into intonation units. In English, an intonation unit typically consists of about four to five words and expresses one new idea unit. Intonation units are likely to begin with a brief pause and to end in a clause-final intonation contour; they often match grammatical clauses. Each idea unit typically contains a subject, or given information, and a predicate, or new information; this flow from given to new information is characteristic of spoken language (Chafe 1994). Arranging each intonation unit on a separate line displays the greater fragmentation inherent in spoken language (Chafe 1982). Capitalization is used for the pronoun I and proper names. Otherwise, capitalization; punctuation and diacritics mark features of prosody rather than grammatical units. Non-lexical items, for example disfluency markers like eh and um, affirmative particles like aha or surprise markers like oh are included in transcripts. The specific transcription conventions are as follows. she’s out. oh yeah? nine, ten DAMN °dearest° says “oh” (2.0) · … ha:rd
>watch out
}: Discourse intonation and vague language. In: Joan Cutting (ed.), Vague Language Explored, 182–197. Basingstoke, UK: Palgrave MacMillan. Weinreich, Uriel 1966 On the semantic structure of language. In: Joseph Greenberg (ed.), Universals of Language, 114–171. (2nd edition) Cambridge, Mass.: MIT Press. Youssef, Valerie 1993 Marking solidarity across the Trinidad speech community: The use of an ting in medical counseling to break down power differentials. Discourse and Society 4: 291–306. Yule, George 1996 Pragmatics. Oxford: Oxford University Press. Yule, George 2010 The Study of Language. (4th edition). Cambridge University Press. Zadeh, Lofti 1965 Fuzzy sets. Information and Control 8: 338–353.
IV. Different interpretational levels – speech acts, politeness and beyond
11.
Requests and orders: a cross-linguistic study of their linguistic construction and interactional organization Carmen Taleghani-Nikazm
Requests can be linguistically realized by means of interrogatives, imperatives, and declaratives. This conversation analytical study provides an empirical examination of instances of requests implemented as interrogatives and imperatives in two different cultures, namely Persian and German. The paper begins with a discussion of some influential and prolific cross-linguistic studies on requests (section 1). While presenting their findings, the paper focuses on methodological approaches, such as Brown and Levinson’s (1987) instrumental politeness theory, that these studies have in developing their examination of requests and tackles some general concerns regarding the fitting approach to the study of requests and orders (1.2). The paper continues with a section that provides an overview of the most recent studies on requests and orders within a conversation analytical framework (1.2). Following a description of the corpus (section 2), the paper focuses on requests constructed as interrogative and imperative in everyday Persian and German conversations and analyzes the specifics of their socio-contextual and sequential organization (section 3). While comparing occurrences of requests formulated as interrogative and imperative in Persian and German, the paper demonstrates how the distributional patterns of the two formats may not only be related to social/cultural factors but rather to speakers’ orientation and understanding of the contextual particularities associated with their request. The paper concludes with a discussion of how cross-linguistic comparison of interactional organization of requests as interrogative and imperative in Persian and German exhibit commonalities in how grammatical composition of social activities is shaped by the interactional environment in which they occur (section 4).
1.
Previous studies on requests
1.1.
Studying requests in the context of politeness theory
Requesting is the most basic activity in human interaction. By making a request speakers get their recipients to perform an action that is for the benefit of themselves or a third party (Searle 1979; Trosborg 1995). Particularly in virtue of their inherent sensitive features, requests have attracted many researchers in the field of linguistics, second language acquisition, and sociology. The core motivation for
322
Carmen Taleghani-Nikazm
studying requests and their variations has been the assumption that the activity is “intrinsically impolite” and carries heavy social implications, and is therefore governed by putative principles of politeness. Thus, when making a request, speakers will use politeness strategies to minimize the threat that is intrinsically present within them. Using different approaches, the majority of these studies have investigated the activities of requesting with the goal of exploring and contrasting politeness strategies in various cultures1. One of the most influential and productive studies on request and its variations is Brown and Levinson’s politeness theory (1978; 1987). They conducted a sociolinguistic, cross-cultural study to investigate the principles of polite behavior in three unrelated languages/cultures, namely, English, Tzeltal (a Mayan language, Mexico) and Tamil (a south Indian language), with the goal of discovering universal principles of politeness. Utilizing the notion of face (Goffman 1955), Brown and Levinson (1987) propose a dual nature for politeness: positive politeness and negative politeness. Positive politeness is expressed by satisfying positive face by indicating similarities amongst interactants or by expressing an appreciation of the interlocutor’s self-image. Negative politeness, on the other hand, involves saving the interlocutor’s face by mitigating facethreatening acts (FTAs), or by satisfying negative face, which indicates respect for the addressee’s right not to be imposed. In their view, participants in an interaction select from a set of politeness strategies that enable them to either avoid or minimize face-threatening activities. Brown and Levinson (1987) suggest that the more threatening an act is the more polite and indirect are the means used to accomplish it. Their goal was to ultimately postulate a universal assumption of cooperation with minimal friction as a basis for polite behavior among members of a speech community. Although Brown and Levinson’s (1987) politeness model is widely acknowledged in the literature, it has generated quite a degree of controversy (Arundale 2005; Eelen 2001; Watts 2003; see also Culpeper, this volume). Overall, some of the limitations of their politeness theory that have been highlighted in the literature include that the claims emerging from their politeness theory are based on fabricated and abstracted examples and do not consider utterances that are produced in interactional environments. Furthermore, their claim that certain lexico-syntactic forms are intrinsically more polite or impolite than others, as shown by Watts (2003), Golato (2005) and Taleghani-Nikazm (2006) may not be the case. Watts (1992) describes politeness as a “socio-culturally determined behavior” that is directed at establishing and maintaining the relationship between members of a social group during an interaction. Watts (2003), particularly, has shown that linguistic structures that are used for politeness purposes in one interactional circumstance may be used for a completely different purpose in another. Taleghani-Nikazm (1999, 2005, 2006) has shown that not only the composition of utterances but also their interactional placement within talk are closely related to politeness issues.
Requests and orders
323
Brown and Levinson’s positive and negative politeness theory (1987) has also inspired many comparative studies on requests in cross-cultural pragmatics (House and Kasper 1981; Blum-Kulka, House, and Kasper 1989; Elsamirasekh 1993; Garcia 1993; Sifianou 1993; Fukushima 1996; Le Pair 1996; Van Mulken 1996; Rue and Zhang 2008). In general, these studies adopted the “Cross-Cultural Speech Act Realization Project”, or CCSARP, analytical method, developed by Blum-Kulka, House, and Kasper (1989) using discourse completion tasks and production questionnaires as the method of data collection. The aim of these cross-cultural studies was to examine similarities and differences between politeness strategies employed by speakers when making requests and apologies in a number of languages (English, Canadian French, Hebrew, Argentinian Spanish, Russian, German, Thai, Chinese and Korean). In the case of request, the core motivation for these studies was the assumption that the activity of request is intrinsically impolite and carries heavy social implications, and is therefore governed by putative principles of politeness. Thus, when making a request, speakers will use politeness strategies to minimize the threat that is intrinsically present within them. Focusing on conversational strategy types such as directness, indirectness, and “internal and external” modifications,2 these studies of requests have illustrated that the relative importance played by social distance, situational settings, and degree of imposition may differ from one culture to another, and that the availability and use of more direct versus more indirect strategies will be culture-specific. Taken together, the results of these studies suggest that indirectness and politeness are not necessarily correlates of each other, either universally or within a given culture. Similarly, other cross-linguistic studies of speech acts have indicated that speech acts tend to vary in terms of their conceptualization and verbalization across cultures and languages and that indirectness may have very little to do with politeness. For example, Wierzbicka (1985: 154) proposes that directness in Polish requests formulated as imperatives is considered appropriate in normal contexts, and that direct requests are avoided in marked situations (such as when the speaker is angry with the hearer). In contrast, in a similar context in English-speaking society the same imperative construction may be considered inappropriate or even impolite. German Requests in German have also been studied within the aforementioned context of politeness strategy research (House and Kasper 1981; House 1989). For instance, using elicited role plays as their data corpus3, House and Kasper (1981) compared politeness strategies employed by speakers of German and English when producing requests in given social situations.4 In their study House and Kasper used a directness level schema as an instrument to measure the degree of politeness. With
324
Carmen Taleghani-Nikazm
reference to this schema, the researchers noted that Germans selected higher levels of directness in requests compared to English speakers and that they employed different strategies from English speakers in order to mitigate request actions. Overall, House and Kasper’s analysis suggested that for mitigating requests, English speakers have a preference for using syntactic means, e.g. “well can you er can you prescribe anything for the allergy” while German speakers tend to use lexical means such as modality markers, e.g. “ja können Sie mir dann vielleicht was gegen die Allergie verschreiben”. House and Kasper proposed that their German subjects used modal particles such as ja ‘yes’, mal ‘just’, wohl ‘well’, einfach ‘simply’, dann ‘then’, and vielleicht ‘maybe’ in their request sentences in order to soften the impact of their utterance on its recipient (House and Kasper 1981: 177–180). In another cross-cultural study, House (1989) examined how German native speakers, German learners of English and British native speakers use the English politeness marker please and the German politeness marker bitte ‘please’ in the context of requests. The motivation for her examination came from House and Kasper’s (1981) study, in which they observed that German bitte ‘please’ was used more frequently and differently than the equivalent English please. House (1989) noted that the choice of certain request strategies is more closely connected to the use of the markers bitte and ‘please’. The analysis of her data suggested two request strategies in which the German marker bitte occurred: the most direct request strategy, i.e. the imperative, and the conventionally indirect strategy (for example, can you/could you …). Furthermore, in her analysis of how German learners of English produce elicited request utterances, House observed that in particular situations, German learners of English used the English marker please more frequently than did English speakers. House attributed this distribution of the usage of ‘please’ to possible pragmatic transfer from the subjects’ native language (German). Persian The social act of requesting has also been of interest to Persian linguists and sociolinguists. Overall the objective of these studies has been to investigate politeness and its realization in the Persian language by examining patterns of request speech acts produced by Persian speakers under the influence of various social factors. Inspired by Brown and Levinson (1987), Grice (1975), and Searle (1979), the studies were principally concerned with politeness as an underlying motivation for indirectness. Using the coding system of Blum-Kulka, House, and Kasper’s (1989) CCSARP project,5 Eslamirasekh (1993) conducted a cross-cultural examination of similarities and differences in the realization patterns of requests between Persian speakers and American English speakers. Overall, Eslamirasekh’s analysis sug-
Requests and orders
325
gests that Persian speakers are more direct when making requests compared to American speakers. In comparison to the American speakers, Persian speakers, used considerably more alerters (e.g. dear friend, excuse me, for me)6, supportive moves, and internal modifiers to soften the impact of a direct request. She explains the results by claiming that Persian speakers may compensate for the level of directness in their requestive speech acts by using more supportive moves, alerters, and internal modifiers in positive politeness strategies. Eslamirasekh relates the higher trend of using various internal and external modifiers in direct requests to the value of group and group harmony tendency in the Iranian culture. She further explains that Persian society is less individualistic and more a group-oriented society, in which members tend to depend on their relationships with others, particularly, within the family. Thus, in a society that stresses group solidarity, directness and assumption of compliance may be the norm. Also inspired by Blum-Kulka et al. (1989) and Brown and Levinson’s (1987) politeness theory, Salmani-Nodoushan (2008) investigates the notion of indirectness in the speech act of requests among speakers of Persian. Using the Discourse Completion Test (DCT) with six scenarios ranging from formal to informal degrees of Perceived Situational Seriousness, Salmani-Nodoushan reports that native speakers of Persian, in general, prefer conventionally indirect strategies when making requests. Additionally, his data analysis suggests that social distance was the trigger for indirectness in requestive speech acts and that solidarity seemed to enhance addressors’ inclination towards directness when issuing requests. For example, in situations where there is social distance between interlocutors, direct requests are very rare. However, in situations where there is no social distance, Persian native speakers frequently use direct requests as if they have a potential for expressing camaraderie and friendship. The results from his study coincide with findings from Eslamirasekh (1993), and studies on other languages suggest that conventional indirectness (for example, can/could you …) is the most preferred strategy when making requests, whereas directness is a sign of closeness and affiliation rather than impoliteness (Blum-Kulka et al. 1989; Felix-Brasdefer 2005; Le Pair 1996; Marquez Reiter 2000; Pavlidou 2000). One possible concern about the aforementioned studies is that their findings are based upon examinations of the compositional features of isolated request sentences, focusing on the semantics of individual elements in request sentences and the illocutionary force they may carry. In keeping with their methodological framework, these studies only examined the actual requests themselves, which were produced in unnatural, fabricated situations; they did not examine the entire natural discourse (i.e. interactional environment) in which requests are produced, including conversational strategies which may occur before and after a given request sentence. Answering questions in (written) questionnaires (such as in a DCT) gives the participants time to think, reflect, and possibly choose what they perceive as an appropriate response as opposed to what may be a more naturalistic response7. This
326
Carmen Taleghani-Nikazm
concern is, of course, not new and has been raised by scholars who emphasize the importance of studying the link between form and function in spoken discourse. In other words, exploring speech acts, their lexico-grammatical compositions and the context in which they occur (Adolphs 2008; Ford, Fox and Thompson 2002; Golato 2005; Taleghani-Nikazm 2006). In the case of requests, it has been argued that if our goal is to better understand the speaker’s choice of a specific form of request, for example, conventionalized indirect or direct forms, and the social context within which particular formats of request are produced, we ought to examine the close relationship between linguistic constructions and contextual variables as evidenced in language use (Adolphs 2008; Curl and Drew 2008; Heinemann 2006; Taleghani-Nikazm 2006). For this reason, a number of researchers have used conversation analysis as the methodology to examine request linguistic structure and composition in relation to its social context, which is the sequential environment within which the request is achieved. 1.2.
The conversation analytical approach to the study of requests and orders
With its focus on real-time interactional language use, CA has been used to examine the sequences that are initiated by requests and the turn design features, that is, how a turn is linguistically constructed, associated with “preference organization” (Levinson 1983; Schegloff 2007; Taleghani-Nikazm 2006; Wootton 1997). The underlying idea in CA is that turns at talk are sequentially ordered; that is, they are sequentially linked to one another in the sense that each turn at talk is shaped by the context of prior talk and that each turn establishes a context to which the next turn will be oriented. Social activities such as requests, offers, invitations, etc. occur in pairs (minimally) where one speaker produces the first part (e.g. an offer) and the second speaker the second part (e.g. an acceptance or a rejection). This basic unit of sequences, i.e. the first pair part and the second pair part is called “adjacency pairs” (Schegloff and Sacks 1973; Schegloff 2007). An important feature of the relationship between the first and second pair part is the notion of preference structure. According to this notion, there is a set of possible responses to a first part; however, elements of this set are frequently not treated by interactants as equivalent choices, in that one second part will be preferred over the others. For example, in English, a preferred response to an offer is an acceptance, while a dispreferred response is a rejection (Davidson 1984; Schegloff 2007). Generally, preferred responses are produced immediately and briefly. In contrast, dispreferred responses exhibit a more elaborate turn structure: they are frequently produced with some sort of hesitation, hedges (e.g. uhm, well), mitigation, and accounts (Davidson 1984; Heritage 1984a; Levinson 1983; Pomerantz 1984; Schegloff 2007). It should be noted that the concept of preference does not refer to speakers’ psychological motives and intentions, but rather to the design of turns associated with preferred or dispreferred activities (Heritage 1984a).
Requests and orders
327
Preference organization does not exclusively apply to the design of response turns or second parts; the first pair parts of certain social activities have also been found to be sequentially preferred practices. For example, it has been suggested that requests and offers are sequentially related and that offers are preferred activities over requests (Heritage 1984a; Lerner 1996a; Levinson 1983; Sacks 1996; Schegloff 1990, 2007). Similar to dispreferred response activities, dispreferred first actions may exhibit structural features of the turn design associated with dispreferred action formats, such as delays, accounts and mitigations. One important procedure that enables co-participants to collaborate to forward preferred actions and avoid dispreferred ones is the pre-sequence (Heritage 1984a; Lerner 1996a; Levinson 1983; Schegloff 1990, 2007). Pre-sequences are inserted before the first pair part of the adjacency pair and are considered to be preliminary to the first pair part (Schegloff 1980). Such expansions are for example preinvitations, e.g. what are you doing this evening?, or pre-requests, e.g. can you do me a favor?. Many pre-sequences are type-specific pre-sequences, i.e. they project a particular sequence type such as an invitation, announcement, etc. More to the point, pre-sequences explore the likelihood that the actions projected will not be responded to in a dispreferred way. For example, an utterance such as what are you doing tonight? can easily be interpreted by the recipient as a pre-invitation or pre-request. The recipient’s response to the first pair part of a pre-invitation may influence the outcome of the next action performed by the producer of the pre-invitation, and it might lead to an actual invitation, or to other talk if the recipient has prior engagements. According to Goodwin and Heritage (1990), pre-sequences provide specific advantages both to their producer, and to their recipients. First, pre-sequences enable parties to abort a projected interaction sequence in which conflict, disagreement, or rejection might emerge. Second, if the projected sequence is not aborted in this way, an affiliative outcome becomes more likely. For example, a pre-sequence that anticipates a request may elicit an offer from the hearer. Seen in this way, pre-sequences are conversational devices through which dispreferred, face-threatening actions and sequences can be systematically avoided in interaction (Heritage 1984a; Lerner 1996a; Levinson 1983; Schegloff 2007). By using pre-sequences to forestall possible rejection turns, co-participants minimize the occurrence of such turns and thereby promote social solidarity (Heritage 1984a). In sum, preferred and dispreferred responses are structured in a way to enhance the performance of preferred responses, and to inhibit the performance of dispreferred responses (Davidson 1984; Goodwin and Heritage 1990; Heritage 1984a; Pomerantz 1984; Schegloff 2007). The specific design of preferred and dispreferred action types is related to their affiliative and disaffiliative characteristics: the preferred responses to requests, invitations, and offers are affiliative actions and thus supportive of social solidarity, while dispreferred responses are largely disaffiliative and destructive of social solidarity (Goodwin and Heritage 1990; Heritage
328
Carmen Taleghani-Nikazm
1984a). That is, speakers follow patterns of behavior to interact with one another smoothly and to minimize the risk of confrontation. The conversation analytic approaches to requests in naturally occurring conversations have revealed the reflective relationship between speakers’ choices of specific formulation of request and contextual factors. Specifically, the studies suggest that the design of request turns, in other words, the way the request is linguistically and syntactically constructed, is tightly connected to speakers’ interpretation or knowledge of the interactional environment in which the request occurs. Speakers’ choices are not directly related to social factors (such as social status, age, gender) but to their assessment of the contingencies surrounding the requested activity (Curl and Drew 2008; Heinemann 2006; Lindström 2005; Taleghani-Nikazm 2006; Wootton 1997, 2005; Vinkhuyzen and Szymanski 2005). In a longitudinal study on children’s requests, Wootton (1997, 2005) demonstrates that there is an orderly connection between the grammatical form in which requests are made (in the form of an interrogative or imperative) and the interactional environment in which they occur. He notes that children at a very young age orient to what he calls “transcontextual knowledge,” (Wootton 1997: 177) which they bring to the occasion and systematically choose a specific form of request that appears to be sensitive to patterns of accountable alignment taken up in preceding talk and action. Curl and Drew (2008) examine the distributional patterns of two most commonly used forms of requesting, Would/could you … and I wonder if … and relate these particular requesting formats to the interactional circumstances in which they are used. They compare the occurrences of request formats that were made in ordinary telephone calls between family and friends with those requests made in after-hours calls to a doctor’s office. In their analysis they noticed a particular distributional pattern, namely that requests made in everyday phone calls were produced with modal verbs or imperatives, whereas requests made in phone calls made to a doctor’s office were often prefaced with I wonder if. Curl and Drew draw the conclusion that modal verbs are used in ordinary conversation, and I wonder if is used in institutional interactions. Furthermore, they note that the distributional pattern is related not solely to the sociolinguistic speech setting but rather with speakers’ orientations to known or anticipated contingencies associated with their request. Speakers’ selections between these two forms reflect their orientations to their entitlement to make the request and to the contingencies that may be involved in the recipient granting the requests. In wonder-prefaced requests, speakers avoid displays of entitlement to the requested object or actions. They instead show that they are not making any presumptions, that they do not know if a particular course of action is the appropriate one: they wonder if such and such can be done, if it is “possible”. Modal verbs are used to display speakers’ assessment of their entitlement to the object of the request (Curl and Drew 2008: 143). In this account, they show how linguistic structures are resources through which actions are managed
Requests and orders
329
and social identities are mobilized in interaction. Generally, the request design is related to speakers’ assessment of their entitlement to the requested object/activity and of the contingencies surrounding its granting. In contrast, in constructing requests using I wonder prefaced formats, speakers orient to the contingencies that may be associated with the request, contingencies that they may be unsure about (Curl and Drew 2008: 147). By using conventionalized request forms can/could you, callers claim the right to have their request granted whether in an everyday or an institutional setting (Curl and Drew 2008: 148). In her study of requests in Danish interactions between elderly care recipients and their home help assistants, Heinemann (2006) similarly analyzes the connection between particular formats of request in Danish and the interactional environment in which they occur. Focusing on the different request formats in Danish, namely positive and negative interrogative structures such as Could you pass the salt? and Couldn’t you pass the salt?, Heinemann demonstrates that choices between the two request formats are tightly connected to the request entitlement issues, i.e. whether the requester is entitled to make a request or not. Requests formatted as positive interrogatives are typically formulated in terms of the recipient’s willingness, they are mitigated, and the requested task is treated as non-routine. This type of request is frequently resisted and challenged by the recipient. In contrast, negative interrogatives are typically formulated in terms of the recipient’s ability, are unmitigated and the requested task is treated as routine. With this format of request, the care recipient orients to her request as one she is entitled to make. Vinkhuyzen and Szymanski (2005) examine how the non-granting responses of copy shop employees’ were shaped by the ways in which customers produced their requests originally. More specifically, they demonstrate that the linguistic formulation of the request had an impact on how employees construct their non-granting responses. In other words, the grammatical design of request determines whether the non-granting response is done in an affiliative or disaffiliative way. Whether the non-granting response is constructed and performed in such a way that reinforces social support or destroys interpersonal relationship. In their discussion they show that requests made by customers take two predominant forms: (1) customers format their requests as self-oriented declaratives stating a customer’s desire or need (e.g. I’d like/I need to make three copies of this), or (2) they format their requests as other-oriented interrogatives that inquire about the organization’s willingness or capability to make a photocopy such as can you make a copy … (Vinkhuyzen and Szymanski 2005: 95). Vinkhuyzen and Szymanski demonstrate that customers formulate their request grammatically as such that denying them fullservice must be done as a disaffiliative response. Request turn designs may also reveal speaker’s understanding of the disaffiliative nature of their action. For example, in a systematic analysis of grammatical construction and the sequential placement of turns produced in German request se-
330
Carmen Taleghani-Nikazm
quences, Taleghani-Nikazm (2006) focuses on speakers’ management of affiliative and remedial work when performing a request and demonstrates that speakers may utilize specific linguistic strategies (i.e. remedial work) to display their orientation to their dispreferred action of requesting. For instance, one common turn design feature of requests in German that is related to remedial work is accounts, that is some sort of explanation or justification for the requestive action. Similar to English conversations, the content of accounts for requests refer to no fault (Heritage 1984b) issues (for example, the lack of access to the requested object) that may serve as remedial work and are related to management of social solidarity. Overall, the above-mentioned interactionally based studies on requests and their variations suggest that their particular formulation (for example, as interrogative or imperative) is not only related to social factors such as speaker’s relationship but rather the speakers’ orientation to specifics of the context, for example, their knowledge or orientation to contingencies associated with their request.
2.
Description of the corpus
The present study is based on 18 hours of recorded German and Persian everyday conversations. The corpus consists of eight hours of audio-recorded telephone conversations in Iran where Farsi was the language of interaction. The German data corpus comprises 10 hours of audio-recorded telephone conversations between Germans in Germany and in the United States. The interactions included everyday naturally occurring conversations between adult relatives, friends, and acquaintances. The speakers were between 20 and 65 years of age. The German speakers came from different regions in Germany (Hamburg, Bremen, Berlin, Brandenburg, Munich, and Nuremberg), whereas the Persian speakers all came from Tehran. For the Persian data, a total of 21 speakers were recorded, of whom nine were females. The German data consists of 22 speakers, of whom 12 were females. About half of the German phone calls were made from the United States by Germans who were living in the United States at the time the phone calls were made8. For the present study, only instances of requests for service (action) performed by one of the interactants were examined. Furthermore, only request situations that had been assessed as being similar between German and Persian cultures were used. The resulting request sequences were those in which the interactional goal of the speakers was to get his or her co-participant to perform an action such as transferring an object, service or information for the benefit of the requester or a third party. The data were transcribed using the transcription convention developed by Gail Jefferson, along with some additional notation used by the author. In the transcript, the top line presents the original talk in German and Persian. The word-by-word gloss, including grammatical descriptions, is provided in the second line. The English translation is provided in italics in the third line of the turn.
Requests and orders
3.
331
An interactional approach to requests in Persian and German
As was discussed above, conversation analytical studies have shown that there is a range of linguistic forms for constructing requests available to the participants and that the speakers’ choices of a particular linguistic form for making requests are tightly connected to the contextual environment in which requests are made. In addition, the speaker’s particular grammatical composition of their requests displays their understanding of and orientation to the particularities of the social action (i.e. disaffiliative action) and the interactional environment in which the request is issued. The remainder of this chapter discusses requests in Persian and German. In particular, the following section provides a detailed analysis of the two different grammatical compositions of requests in Persian and German cultures: a) Requests formulated as interrogative with the modal verb can, for example can/could I … and can/could you … This particular request formulation has been described in some of the previous literatures as conventionalized indirect requests (see, for example, House and Kasper 1981) b) Requests in the form of imperative mood such as go! or look over here!. This request formulation has also been referred to as direct requests (Brown and Levinson 1987; House and Kasper 1981; Blum-Kulka et al. 1989) or orders. 3.1.
Requests formulated as interrogatives in Persian and German
3.1.2.
Switchboard requests (can I talk to X?) in telephone conversation openings
An investigation of both data sets, Persian and German, suggests telephone conversation openings as a common interactional environment, in which requests in the form of interrogatives with a modal verb occur. In both languages speakers utilize similar syntactic constructions to formulate a request to speak to another person can I talk to X?. Schegloff (1979: 30) has referred to this type of request in American English telephone openings as “switchboard request”. The first segment in this section exemplifies such requests in Persian. In this conversation, Ali, calls his sister’s house to speak with his nephew, Paymân, to arrange a game of Tennis with him. When Ali calls, Pari, Paymân’s sister answers the call. (1) Persian: mitoonam sohbat konam? (Can I talk to X?) 01 Ali : pay-par-paymân hast?9 pay-par-paymân is? is pay-par-paymân there? 02 Pari: bale yes
332
Carmen Taleghani-Nikazm
03 Ali : mitoonam sohbat konam,? can+I talk do+I,? can I talk to him? 04 Pari: bale gooshi khedmadetoon bashe, yes receiver serve+you(F) be, yes hold on please,
After exchanges of greetings and how are yous (not included in the example), Ali asks his niece whether Paymân is at home (line 1), thereby, displaying recognition of the recipient, namely Pari, as not the intended recipient. After receiving an affirmative answer from Pari (line 2), Ali performs a request in the form of an interrogative to talk to Paymân (line 3). Ali’s request is constructed with the modal verb tavanestan ‘to be able to/can’ and is formulated as a self-oriented interrogative, mitoonam ‘can I’, exploring the possibility for the caller to talk to a third party, namely Paymân. This construction of switchboard requests in the form of a selforiented interrogative with the modal verb “can” was also found in the German telephone conversation openings. The next segment exemplifies similar request construction, which is a “switchboard” request formulated as an interrogative with the modal verb können ‘can’ in German. In the following excerpt, Ingo calls to talk to his friend Kurt. (2) German: kann ich ihn mal sprechen? ‘Can I talk to X?’ Ring 01 Answerer: hallo,? hello,? 02 Ingo
: hier ist ingo, hallo. is der kurt da? here is ingo, hello. is def.m. kurt there? this is ingo, hello. is kurt there?
03 (.) 04 Answerer: ja:, yes:, 05 Ingo
: .hhh kann ich ihn mal sprechen bi[tte? .hhh can I him MP speak pl[ease? .hhh can I talk to him pl[ease? [ 06 Answerer: [mhm, [mhm, 07 Answerer: kurt,?
In line 2, following a self-identification and a greeting, Ingo asks whether Kurt is there (is der kurt da? ‘is Kurt there?’). After receiving a confirmation from the answerer (line 4), Ingo formulates a switchboard request to speak to his friend, Kurt (kann ich ihn mal sprechen … ‘can I talk to him …’). Note that Ingo’s request is constructed as a self-oriented question with the modal verb können ‘can’ asking about the possibility for him (the caller) to talk to his friend Kurt. Furthermore, Ingo’s request is accompanied by the politeness marker bitte ‘please’ displaying
Requests and orders
333
his understanding of the disaffiliative nature of his action. It has been suggested that the politeness marker bitte in a requestive situation may serve to soften the imposition of the action (House and Kasper 1981). Note also that in addition to the politeness marker bitte, the speaker uses the modal particle mal ‘just’ which also serves to smoothen the impact of the utterance on its recipient (House and Kasper 1981: 177–180). Unlike German, there are no similar modal particles in Persian requests. However, Persian speakers commonly use politeness markers and expressions when performing requests (see below). In sum, the cross-linguistic analysis of switchboard requests in Persian and German telephone openings revealed that Persian and German speakers frequently use a conventionalized request formulation of can I talk to X? (in Persian and German) to ask for the possibility to talk to a third party. Additionally, the analysis of the syntactic construction of switchboard requests and the interactional environment in which they occur (in both languages) suggest that this request formulation is related to the caller’s assessment of the requested activity and of the contingencies surrounding its granting. In other words, after establishing the proximity or accessibility of the intended party (is X there?) and formulating the request as a selforiented interrogative using a modal (can I talk to X) speakers display their orientation to the contingencies associated with the possibilities or likelihood for them (i.e. the caller) to talk to the intended party. 3.1.3.
Can you/could you … Request format in everyday conversation
In general, items or topics in the conversation may generate requests and in such situations speakers utilize various linguistic resources to formulate and perform their requests in relation to the sequential context. One particular request format that has been observed in both German and Persian data sets is the interrogative can you/could you … Consider data excerpt (3) taken from a Persian conversation between Ali and his nephew, Paymân. Ali calls Paymân to ask if he wants to play tennis on that day. Some background information is needed here. Tennis courts in Tehran belong to clubs, and normally one would need to reserve a court for a particular time to play tennis; however, there is also the opportunity to find a court available without prior reservation depending on the time of the day. This seems to be Paymân’s main concern. It is in this context that Ali asks Paymân to go over to the court and reserve a court personally (line 25). (3) Persian: Tennis court (can you …) 11
Ali
: mirim tennis,? mikhâi berim tennis? go+1stPl tennis,? want go+1stPl tennis? are we playing tennis,? do you want to play tennis?
12
Paymân : kei berim? when go+1stPl. when shall we go?
334
Carmen Taleghani-Nikazm
13
Ali
14
Paymân : kei? when?
15
: hâ? huh?
(.)
16
Ali
17
Paymân : sâat chand? time what? what time?
18 19
(0.5) Ali
20 21
: haft biâm? seven come+1stSg? shall I come at seven?
: haft oonjâ. seven there. I’ll be there at seven. (.)
Paymân : haft? seven?
22
(.)
23
Ali
24
Paymân : khooneye mâ bâshin âkhe oonjâ zamin gir home+of us be+2ndPl. Because there court catch be at our house because you cannot get a court
25
Paymân : nemiâd. n+come there.
26 27
: âhâ. mhm.
(0.7) Ali
28
: râst migi? truth say+2ndSg.? really? (0.5)
29
Paymân : bale. .hhh masalan bâyad yeh-yerob yes. .hhh for example have to one-a quarter yes. .hhh for example we have to go there: about a-
30
Paymân : bis daghighe zoodtar berim oonjâ:, twenty minutes earlier go+1stPl there:, fifteen to twenty minutes earlier,
31
Ali
32
: khob. okay. (0.5)
Requests and orders 33
335
Paymân : zamina bâshan courts be+3rdPl. courts to be available
34
(0.5)
35 -> Ali
: mitooni to beri begiri? can+2ndSg you go+2ndSg get+2ndSg? can you go and reserve one?
36
Paymân : man ke (.) .hhh e:h ma-ghosh-ghoshi shomâ me that (.) .hhh u:h ma-receiv-receiver youF well I (.) .hhh u:h ma-hold-hold on what
37
Paymân : sâate chand bâyad berin? (talking to his Mother) time what have to go+3rdPl? time do you have to leave?
38 (3.2) (Paymân’s Mother explains something in the background)
Our target turn is in line 35, where Ali asks Paymân to go and reserve a tennis court for them. Ali’s request is constructed as a question with a modal verb, namely mitooni ‘can you’. A close analysis of the construction of Ali’s request and its contextual environment reveals that this particular design of request is closely connected to the sequential position in which it occurs. In other words, it seems that Ali’s request was occasioned by the prior talk; Ali suggests a specific time, namely seven o’clock (line 16), Paymân displays problems with the suggested time through a series of return turns (lines 17–23), a concern that at that particular time they may not find an available court, and he suggests that one should go about fifteen to twenty minutes earlier (lines 29–30). Ali then asks his nephew whether he can go and reserve a court (mitooni to beri begiri?, line 35). One important piece of information that may be relevant for the analysis is that Paymân lives much closer to the tennis club than Ali, so if someone should go to reserve a court, it would be Paymân considering his proximity to the club. Note that Ali formulates his request as an other-oriented interrogative. By formulating his request in terms of the recipient’s ability (i.e. the modal), Ali displays his assessment of the situation, namely Paymân’s proximity to the club. Furthermore, by formulating his request as an interrogative, Ali orients to the recipient’s ability and likelihood to comply with the request, rather than Paymân’s willingness to perform the requested action. Similar request structures, i.e. can you/could you … were also found in the German corpus. Excerpt (4) is taken from a conversation between Dora and her mother (Mutti) in Germany. After a brief conversation about the family cat and how it has been gaining weight (not included in the transcript), Dora performs a request (lines 2–4): she asks her mother if she is able to tell their mutual friend, Miriam, that Dora has a certain set of pictures that she wanted to see.
336
Carmen Taleghani-Nikazm
(4) German: Fotos (can you …) 01
Mutti : .hh ha dieses viech .hh ha this animal .hh ha this animal
02 -> Dora : mh °naja° äh he .hh hum. ah kannst vielleicht der mi-na mh °well° uh he .hh hum. oh can+you maybe the mi-no mh °well° uh he .hh hum. oh can you maybe mi-no 03 -> Dora : ich schreib jetzt ihr selber mal der Miriam mal sagen, dass i write now to her myself PT the Miriam MP say that i’ll write her myself just tell Miriam that I have 04 -> Dora : ich fotos hab weil die will a eweng was sehen. i pictures have because she wants also a bit some see. pictures because she also wants to see some of them. 05
(0.8)
06
Mutti : auch die: photos von mir,? also the: pictures from me,? my pictures, too?
07
Dora : ha ha yeah yeah yeah yeah
08
(0.5)
09
Dora : ha= ha= ha=
10
Mutti : =ich muss sowieso’s mal die woch nunter ins =i must anyway MP this week down to =i have to go this week anyway down to
11
Mutti : theater eingladen, theater invited, invited to the theater,
12
Dora : mh mh mh mh mh mh
13
Mutti : weil wir mal wieder ins theater genga, because we MP again to the theater go, because we’re going to the theater again,
14
Dora : ha ha, mh mh, mh mh
15 16
(.) Dora : ach [da kannste oh [there can+you oh [there you can [
Requests and orders 17
Mutti :
337
[un da nimm ich se mal mit. [and there take i them MP with. [and i’ll take them with me.
Dora’s request turn in lines 2–4 is designed as an interrogative with the modal können ‘can’ (in the second person singular). Note that Dora’s request kannst vielleicht der Miriam mal sagen dass ich fotos hab ‘can you maybe tell Miriam that I have pictures’ (lines 1–3) is interrupted to insert the phrase na ich schreib jetzt ihr selber ‘no I’ll just write her myself’ which dismisses the previous component of her turn, namely beginning of a request (kannst vielleicht … ‘can you maybe …’). Dora, however, decides to continue with the request turn and completes it with providing an account weil die will a eweng was sehen ‘because she also wants to see some of them’ (line 4). Previous studies on requests (Blum-Kulka et al. 1989; Taleghani-Nikazm 2006) have suggested that accounts (i.e., some kind of explanation or justification) are a design feature of requests and that the content of accounts for requests commonly address some “uncontrollable” issue and or “lack of access” to the requested activity or object. Similar to the previous example from Persian, it appears that the other-oriented interrogative request format with a modal verb may be related to some contextual factors. Specifically, it seems that the issue of proximity may play a role in Dora’s request design. Note that in Dora’s request she asks her mother to tell their friend about the pictures, whereas in the insertion clause (which retrieves her incomplete request), she expresses that she’ll write10 to her. Hence, it appears that Dora does not have access to the friend in person (since she lives in a different city than her mother), but her mother is the one who will see the friend in the near future when she goes down to the theater, as she indicates in line 13. Dora’s request turn design shows her knowledge of the recipient’s closer access to the person and is formulated in terms of recipient’s ability to perform the request. In addition, similar to the previous data example from the Persian data set, the speaker here has the option or possibility to perform the activity herself, and it seems that, due to the recipient’s proximity to the person, the speaker chooses to ask the recipient to perform the service. To reiterate, the above section has provided a description of two particular request turn designs, namely requests formulated as self and other-oriented interrogatives with modal and contextual environments in which they typically occur in the Persian and German data sets. A comparison of these particular request formats and their sequential placement in both languages demonstrates how the speaker’s choice of a specific form of request is tightly connected to contextual circumstances in which they occur. A comparison of Persian and German telephone conversations revealed that the can I … request format commonly occurs as switch board requests where the caller, upon securing the intended recipient’s availability, asks the answerer to speak to the intended party. In the case of can/could you … request format, it was suggested that speakers formulate their request as an interrogative with a modal in situations in which the requester has no immediate access
338
Carmen Taleghani-Nikazm
to the requested activity whereas, the recipient has. It was suggested that speakers in both languages formulated their requests in terms of the recipient’s ability and likelihood of performing the requested action rather than their willingness (in all examples in the data the request was accepted). This specific design of request seems to be influenced by the underlying structure of social interaction, i.e. the management of self-and other relationships and the maintenance of social solidarity. 3.2.
Requests formulated as imperative (order) in Persian and German
An investigation of the Persian and German data revealed that speakers in both languages frequently design their requests in the imperative mood, thereby expressing orders. Imperatives in German and Persian have several inflected forms, which may change according to grammatical number. In general, orders have been described as inappropriate and impolite in certain social situations. For example, Brown and Levinson (1987) have described orders and requests as face-threatening acts, in other words, acts that threaten the recipient’s negative-face want by imposing some future act on the recipient, and thereby putting pressure on the recipient to perform the act. Hence, when performing orders, speakers employ politeness strategies to mitigate and soften their face-threatening acts. Other studies on orders have shown that the usage of imperative mood and its appropriateness depends on factors such as social relationships and contextual situations (Eslamirasekh 1993; Goodwin 1990; Lindström 2005; Salmani-Nodoushan 2008; Wootton 1997) A comparison of linguistic formulation and contextual placement of requests in the form of imperative (i.e., orders) in Persian and German suggests certain similarities in terms of turn design and the interactional environment in which imperatives were performed. In addition, the systematic analysis of the construction of imperatives demonstrates some culture-specific differences in speakers’ lexical choices. The discussion will begin with examples from the German data set and continues with imperatives in Persian. Throughout the discussion below the social activity of order will be referred to as imperative since this is the grammatical formulation through which orders in the data are being realized. The first example of requests in the form of imperative in German comes from a conversation between Karin, a German speaker who at the time was a student at an American university, and her sister-in-law in Germany, who is planning to come and visit her in the United States. Our target turns are in lines 3–5, where Karin asks her sister-in-law to bring her more plum butter from Germany (a food item that is not commonly found in the United States) when she visits her in winter.
Requests and orders
339
(5) German: Pflaumenmus (bring me …) 01
Karin : .hhh oka:y, .hhh oka:y, .hhh oka:y,
02
Rita : >okay< kar[in >okay< kar[in >okay< kar[in [ Karin : [du übrigens weisst du was wenn ihr [you by the way know+you you what when you [by the way you know what when you
03
04 -> Karin : ähm m winter kommt, (0.7) .hh da bringt mir uhm in winter come, (0.7).hh MP bring me uhm come to visit in winter, (0.7).hh MP bring 05 -> Karin : pflaumenmus pflaumenmus mit plum butter plum butter with me plum butter plum butter 06 07
08
(0.5) Rita : [hä:,? [hu:,? [hu:,? [ Karin : [e:h he he he he .HHH ich ha(h)b heute morgen den [u:h he he he he .HHH i ha(h)ve today morning the [u:h he he he he .HHH this morning i finished the
09
Karin : letzten pflaumenmus verspeist. ist schon ganz last plum butter ate. it is already completely last plum butter. it’s quite
10
Karin : traurig he he [he he he sad he he [he he he sad he he [he he he [he he he Rita : [kann man den kann man den auch [can one it can one it also [can you can you also
11
12
Rita : schicken,? [( send,? [( mail it,? [
) )
Prior to this segment, the speakers talk about espresso makers and coffee (not in the transcript) and exchange of possible pre-closings (Schegloff and Sacks 1973) in lines 1–2. It is in this context that Karin performs a request in the form of an imperative bringt mir pflaumenmus pflaumenmus mit ‘bring (you plural) me plum butter plum butter’ (lines 4–5). The imperative utterance in Karin’s turns is preceded by a series of prefatory components: the pronoun du ‘you’, a summons, a “misplacement marker” (Schegloff and Sacks 1973) übrigens ‘by the way’, and the
340
Carmen Taleghani-Nikazm
discourse marker weisst du was ‘do you know what’ which functions as an attention-getting device (Schegloff 1968) by addressing the recipient and getting her attention. Note that übrigens in combination with the address term marks the relationship of the current turn to the prior talk. In particular, it signals that the new turn is sequentially out of place (note the exchanges of possible pre-closing indicating a possible place for beginning a new topic/activity) and that which follows is topically related to some distant prior talk (Taleghani-Nikazm 2008). Note that earlier in the phone conversation, Karin and Rita talked about Rita and her family’s planned trip to the US in winter. They also talked about Karin preparing a list of items that she wishes her family to bring her from Germany. It has been shown that young children use imperatives in circumstances in which the child can presume a willingness on the part of her recipient to perform the requested activity (Wootton 1997). In this case, considering the contextual environment to be the prior talk about the family’s visit to the United States, it seems that Karin can presume a willingness on the part of her family, i.e. Rita, to grant her requested action. Furthermore, it may be argued that Karin constructs some grounds for her imperative by uttering wenn ihr ‘m winter kommt ’when you come in winter‘ just before her imperative. In doing so, she displays her understanding of the contingencies of granting her request and the basis for expecting her imperative to be in accord with the preferences of her recipient. The next data sample (6) provides another stretch of talk in which a speaker performs a request in the form of an imperative. This is a conversation between Markus and his grandmother. Markus is a student at an American university at the time of the conversation and calls his grandmother. In the excerpt, Markus’ grandmother explains to him that her neighbor brought her some apricots and that she is going to bake an apricot cake the next day since Nicki (Markus’ sister) likes it so much (line 1). Upon hearing this information, Markus mentions that he likes apricot cake as well (line 2). In response, his grandmother expresses her surprise (line 3). After a pause (line 5), some laughter and a micro-pause, Markus’ grandmother explains that she cannot mail him an apricot cake (lines 7–8). It is in this context that Markus performs a request formulated as an imperative (lines 9–10). (6) German: Aprikosenkuchen (give her …) 01
Oma
:nicki isst ja so gerne aprikosenkuchen= yea so like apricot cake= eating that so much=
02
Markus:=ja ich auch. =yes me too. =yes me too.
03
Oma
:du auch?= you too?= you too?=
Requests and orders 04
Markus:=ja =yes =yes
05 06
07
08
341
(0.2) Markus:[hehe[hehe[hehe[ Oma :[hehe(.)ja den kann ich dir leider nicht [hehe(.)yes that can i you unfortunately not [hehe (.) yes I unfortunately can’t mail that to Oma
:schicken= send= you=
09 -> Markus:=.hhhh ja, kannst einfrieren hehe gib der nicki =.hhhh yes,can freeze hehe give the nicki =.hhh yes, you can freeze it hehe give nicki 10 -> Markus:nur ein stück hehehe only one piece hehehe only one piece hehehe 11
Oma
:ja ’n stückchen kann ich dir einfrieren das ist ne gute yes a piece+SUF can I you freeze that is a good yeah I can freeze a small piece for you that is a good
12
Oma
:idee. (0.2) das mach ich. idea. (0.2) that do I. idea. (0.2) I’ll do it.
In lines 9–10, Markus first suggests to his grandmother to freeze a piece of the cake and then tells her to give Nicki, his sister, only one piece, gib der nicki nur ein stück ‘give Nicki just a piece’. In return, Markus’ grandmother first agrees with Markus’ suggestion expressing that she likes the idea (line 11). This is then followed by her agreeing to grant the request das mach ich ‘I’ll do it’ (line 12), thereby showing her understanding of Markus’ turn as a request. As in the prior example, the imperative occurs in a contextual environment in which the speaker presumes a willingness on the part of the recipient to grant the requested action. Markus’ comment that he likes apricot cake provides justification for his request and constitutes grounds for his choice of the imperative form of request. Later in the conversation, the grandmother mentions that she has frozen asparagus for him (not included in the transcript), thereby demonstrating that this is a common activity for her, i.e. freezing food items for her grandson while he is abroad. Note that Markus’ imperative is followed by some laughter. Studies on laughter in social interaction have suggested that laughter is indexical and that its interactional meaning emerges from its sequential placement within the talk (Jefferson 1984; Glenn 2003). It appears that the laughter marks Markus’ request as
342
Carmen Taleghani-Nikazm
playful or humorous. The laughter here may also be understood as a strategy to soften the request imposition (i.e. to freeze a whole cake for him and only give one piece of it to his sister, for whom the cake was originally intended). Excerpt (7) is another example of a request in the form of an imperative. In this conversation Klaus tells his friend, Markus, who is on an exchange program at an American university, to provide information about the price of a specific tennis racket (lines 5, 36 and 37). (7) German: Tennisschläger (find out …) 01
02
03
04
Markus: ich glaube [das i believe [that i believe [it [ Klaus : [MARKUS genau das wollte ich [MARKUS exactly that wanted i [MARKUS exactly that’s what i Klaus : noch bitten still ask wanted to ask you (0.2)
05 -> Klaus : erkundige dich ma:l nach ’em tennisschläger. inform yourself MP after a tennis racket. get some information about a tennis racket. 06
(0.5)
07
Markus: ja= yes= yes=
08
Klaus : =und zwar (0.2) nach einem wilson prostar =and MP (0.2) after a Wilson prostar =about (0.2) a Wilson prostar
09
Klaus : der edberg schläger. the edberg racket. the edberg racket.
10
Markus: also ein moment ich muss mal aufschreiben da. so a moment i must MP PRX+write there. just a moment there i have to make a note
11
(0.5)
12
Markus: .hh also? .hh so? .hh so?
13
Klaus : wilson. wilson. wilson.
Requests and orders 14 15
(.) Klaus : ne? right? right?
16
(0.2)
17
Markus: ja,? yes,? yea,?
18
Klaus : und der heisst prostar and the name+it prostar and it’s called prostar
19 20
21 22
23
24
(1.2) Markus: propropro(0.2) Klaus : st[ar st[ar st[ar [ Markus: [plural. ja? [plural. yea? [plural. right? (0.5)
25
Klaus : bitte? please? pardon?
26
Markus: plural? stars? p- mit s? plural? stars? p-with s? plural? stars? p-with s?
27 28
29 30
31
32
(0.5) Klaus : star star star (.) Klaus : [prostar [prostar [prostar [ Markus: [ja [yes [yea Markus: ja. yes. yea.
343
344
Carmen Taleghani-Nikazm
33
Klaus : nicht stars sondern star eigentlich not stars but star actually not stars but star actually
34
Markus: mhm mhm mhm
35
Klaus : aber das ist der der edberg schläger,.hhh but that is the the edberg racket, .hhh but that’s the the edberg racket, .hhh
36 -> Klaus : (.) w-we-erk-kundige dich mal wieviel der da (.) w-we-in-form yourself MP how much the there (.) w-we-find out how much it costs 37 -> Klaus : kostet und schreib mir das mal. cost+it and write me it MP there and write me 38 39
(0.8) Markus: ja okay yes okay yea okay
In line 5, Klaus tells his friend to get some information about a tennis racket (a Wilson Prostar) erkundige dich ma:l nach ’em tennisschläger ‘get some information about a tennis racket. After some repair and clarification on the name of the model (lines 6–35), Klaus performs another related request (lines 36–37); this time he tells his friend to get information about the price of the tennis racket erkundige dich mal wieviel der da kostet ’find out how much it costs there‘ (the adverb there refers to United States). Similar to his first request, Klaus formulates the second request as an other-oriented imperative. This request is followed by a third request schreib mir das mal ’write it for me (telling him to email him the price), which is also formulated as an imperative. Note that the request sequence in this segment does not begin with Klaus’ request turn in line 5. There are some elements in Markus’ previous turn which specifically project an upcoming request. Consider Klaus’ turn design MARKUS genau das wollte ich noch bitten ‘MARKUS exactly that’s what I wanted to ask you’. He begins his turn with a summons (which is produced louder than the surrounding talk) and the adverb genau ‘exactly’. These elements in the beginning of Markus’ turn may be understood as marking his turn as the beginning of a new sequence. Markus continues the turn with a pre-request, announcing that he wanted to request something from his friend. Klaus’ turn design foreshadows the particular type of action that will follow next, namely a request. It has been suggested that speakers may utilize specific linguistic components, produced as “prefatory” (Schegloff 1980: 116) to the projected action in order to signal that they are in the process of initiating a new activity. Klaus’ prefatory turn to his request specifically exhibits the type of activity he is about to perform, namely
Requests and orders
345
a request, thereby providing his recipient with some information as to what to expect in his upcoming utterance. It is in this context that Klaus performs an imperative. Considering the social relationship between the speakers (close friends) and the circumstances (one living abroad for a short period of time), it may be argued that Klaus can presume his friend’s willingness to get the information he needs, hence formulating his request in the imperative. Another noteworthy feature is observable in Klaus’ request sequence. As was noted earlier in the chapter, requests are dispreferred actions and when performing requests, speakers utilize strategies to manage performing a dispreferred action. In example (7), by producing the prefatory component to his request turn, Klaus allows his recipient of the current turn to inspect what is “in play”, thereby facilitating coordination of social activities in interaction (Taleghani-Nikazm 2006). In sum, the analysis of requests formulated as imperatives in German has shown that the occurrences of imperatives were mainly associated with the interactional environments in which they were produced. It was shown that in the case of imperatives, not only social factors such as the relationship between the speakers may play a role, but that their occurrences were tightly connected to the details and particularities of the circumstances in which they were used. The next part of this section analyzes examples of requests in the form of imperatives in Persian and compares with German usage. Excerpt (8) is taken from a telephone conversation between Reza and his wife Massi. Reza is at a friend’s house and calls home. After some telephone opening exchanges (lines 2–9) and a brief explanation that the conversation is being recorded (lines 9–10), Reza tells his wife to speak properly so her talk can be recorded with good quality (lines 12–13). Reza performs another request: he tells his wife that if someone from Mr. Tâheri’s (his boss11) home calls, she should tell them that he is on his way home (lines 15–18). (8) Persian: Speak properly 01
Ring
02
Massi: °°alo?°° °°hello? °° °°hello? °°
03
Reza : alo. hello. hello.
04
Massi: alo? hello? hello?
05
Reza : e:h Massi? e:h Massi? e:h Massi?
346
Carmen Taleghani-Nikazm
06
Massi: salâm, hello, hello,
07
Reza : salâm. hâlet khoobeh? hello. feeling+2ndSg. good+is? hello. how are you?
08
Massi: hâle shomâ feeling youF how are you?
09
Reza : e:h mersi mamnoon. bebin in alân man dâram sohbat e:h thanks thank you. look this now I AUX talk e:h thanks thank you. listen this now while I’m
10
Reza : mikonam dâreh eh rooye navâr zabt mishe do AUX eh on tape record doing talking it’s being recorded on a tape
11
Massi: bale, yes, yes,
12 -> Reza : kh-khoob sohbat bekon ke rooye navâr dorost zabt g-good talk do+2ndSg. so on tape right record talk properly so it can be recorded 13 -> Reza : beshe can right 14 -> Massi: [kohb. [ok. [ok. [ 15 Reza : [bebin e:h ageh az manzele âghâye Tâheri telefon [look e:h if from home+of Mr. Tâheri call [listen e:h if they call from Mr. Tâheri’s 16
Reza : kardan, made+2ndPl., home,
17
Massi: bale yeah yeah
18 -> Reza : begin ke man too râh hastam, tell+2ndPl. that I in way are, tell them that I’m on my way, 19
20
Massi: bale [bale yes [yes yes [yes [ Reza : [e:h miresam be manzel khob? [e:h arrive+1stSg. at home ok? [e:h I’ll be at home ok?
Requests and orders 21
22
347
Massi: bâshe bâ[she okay ok[ay okay ok[ay [ Reza : [âh … [âh … [âh …
Note that by explaining to his wife that the conversation is being recorded on a tape (lines 9–10), Reza provides some background information for his imperative, which constitutes some relevant context for his request/order. Another noteworthy feature of this request sequence is that, similar to previous examples of requests in the form of an imperative, the speaker provides an explanation or a justification for his request, namely that the voice is recorded in good quality. Massi replies to the imperative with khob ‘okay’ signaling her understanding of the request and granting performing of the activity. Reza performs another request in the form of imperative (line 18). The request is preceded by an ifclause ageh az manzele âghâye Tâheri telefon kardan, ‘if they call from Mr. Tâheri’s home’, providing a condition or circumstance for the requested service to be performed – a feature that was also observed in examples from the German data. Requests in the form of imperatives may be accompanied by some variety of appreciation term/phrase such as please. Excerpt (9) is taken from a phone conversation between two distant relatives, Ali and Ahmad. At the time of the phone conversation, Ahmad, who lives abroad, has called Ali at his home number to let him know that he is back in town, i.e. Tehran. Our target turn is in lines 13–14 where Ahmad performs a request in the form of an imperative followed by an appreciation token lotfan ‘please’. (9) Persian: bedin lotfan (give please) 01
Ali
02
Ahmad : hatman [yâ be Mariam zang mizanam, definitely [either to Mariam ring make, definitely [I’ll either call Mariam, [ Ali : [âreh [yeah [yeah
03
04 05
: beharhâl e:h bebinim shomâ ro. anyway e:h see+1stPl. youF prep. anyway e:h let’s see each other.
(0.5) Ali
: bâshe. okay. okay.
348
Carmen Taleghani-Nikazm
06
Ahmad : yâ:n be khodetoon. or to youF. or you.
07
(.)
08
Ali
: bâshe. okay. okay.
09
Ahmad : shomâre telefoni hastesh ke be man bedin ke number+of phone is+3rdSg that to me give+2ndPl. that is there a phone number that you could give so
10
Ahmad : man betoonam bahât I can with+youIF I can call you
11
Ali
: hamin shomâre manzel dige behtarine bâ oothis number+of home actually best with thathis number of home actually is the best and tha-
12
Ali
: hamoon lufthânsâ ke behtarine bâ Mariam. that lufthansa that best with Mariam. that of lufthansa that it’s best reaching Mariam.
13 -> Ahmad : oh lufthânsâ ro j-jadidesho nadâram behem oh lufthansa Prep n-new one don’t have to+me oh lufthansa’s phone number the n-new one I don’t 14 -> Ahmad : bedin lotfan. give+2ndPl. please. have. what is it. 15
16 17
Ali
: hafsado haftâdo do, seven-hundred seventy two, seven-hundred seventy two, (0.7)
Ahmad : kho:b, oka:y, oka:y,
Prior to Ahmad’s request, Ali suggests seeing each other. Ahmad, in turn, agrees and states that he either calls Mariam (i.e. at her work number, namely Lufthansa, the German airline) or he will call Ali (lines 2–6). Upon Ali’s agreement of the arrangement (line 8), Ahmad asks Ali if there is a phone number at which he can reach him (lines 9–10). In response, Ali tells Ahmad that the Lufthansa number, i.e. Mariam’s work number, is the best way to reach them (lines 11–12). It is in this context that Ahmad performs an imperative. Ahmad begins the turn with a changeof-state token (Heritage 1984b) oh ‘oh’ marking receipt of new information and announces that he does not have Lufthansa’s new number. Ahmad’s account for his request is immediately preceded by the request. In doing so, the speaker provides an explanation, more specifically, background information for the upcoming re-
Requests and orders
349
quest. Furthermore, it seems that not only the provided account but also the entire stretch of talk prior to Ahmad’s request constitutes grounds for the use of an imperative. In particular, considering the particularities of the circumstances it can be understood that Ahmad’s imperative is produced in a context in which he presumes Ali’s willingness to provide him with the requested information, since it was Ali who claimed Lufthansa number to be the best way to contact them. Ahmad then continues his turn by telling Ali to give him the (new) number (lines 13–14). Note that he adds a politeness marker (House and Kasper 1981) lotfan ‘please’ to his request thereby displaying his orientation to the dispreferred nature of his activity. It is worth observing that Ahmad and Ali’s relationship is rather formal. This can be also noted in the speakers’ usage of formal you in Persian shomâ (line 1) and the plural ending of khodetoon ‘yourself’ (line 6), thereby using polite language forms. By adding the marker lotfan, Ahmad displays his understanding of their relationship and the delicate nature of the activity he has just performed. Excerpt (10) is another example of request in the form of an imperative accompanied by a politeness marker. Reza calls his parent’s house, and his brother-inlaw, Mansoor, answers the call. After some talk, Reza asks if their family friend, Davood, is still there and has not left yet for they are supposed to go to a funeral. After finding out that the friend is still there, Reza tells his brother-in-law to tell the friend to wait so they can go together (lines 5–6). (10) Persian: Ghorboonat begoo (please tell …) 01
Reza
02 03
:
[in agha davood raft ya oonjast,? [this Mr. Davood went or there+is,? [has Mr. Davood left or is he still there,? (0.7)
Mansoor: davoo::d, injast narafte. Davoo::d, here+is not+gone. Davoo::d, is here hasn’t left.
04
(0.7)
05 -> Reza
: pa ghorboonet12 agar hast behesh begoo vaste ke then I please if is+3rdSg. to+him say wait so then please if he’s there tell him to wait so
06 -> Reza
: baham berim. together go+1stPl. we can go together.
07 08
09
(0.5) Mansoor: bashe ba[she okay ok[ay okay ok[ay [ Reza : [man alan miam pain. [I now come down. [I’ll come down.
350
Carmen Taleghani-Nikazm
Reza’s request turn begins with the resumptive element pa (an abbreviated form of pas) ‘then’ referring to the information he received in the prior turn (line 3) that the family friend is still there. The next element in Reza’s request turn is the expression ghorboonet ‘please’. This expression and its variation literally means ‘I am ready to be sacrificed for you’, and it can be used interchangeably with the equivalents of ‘please’ and ‘thank you’ in formal and informal conversation (Moosavie 1986). Reza continues his turn by producing the conditional-clause agar hast ‘if he is there’ thereby expressing the contingency of granting his request (note that this was his question in line 1 and reason for the call). Reza performs a request in the form of an imperative behesh begoo vaste ke baham berim ‘tell him to wait so we can go together’. Note further that the request is followed by an account for the requested service. In terms of the context in which this imperative occurs, it can be noted that the caller tells the recipient to give his message to the third party after securing the other parties’ presence at their home. It can be argued that Reza presumes the recipient’s willingness to perform the service since they have been waiting for him to get home so they can leave for the funeral. Furthermore, by asking whether the third party is still at his home, Reza provides the recipient with some context or information about his upcoming action. Requests in the form of an imperative may also be preceded by some other polite expression such as bizahmat ‘without trouble’. Excerpt (11) exemplifies this. Reza is at his boss’ house taking care of some business when he calls an acquaintance. The acquaintance (Alireza), however, is not home yet, even though he had told Reza to call at that time. Reza tells the wife to tell him to return his call at Mr. Tâheri’s house (lines 6–7). (11) Persian: bizahmat begin (without trouble tell …) 01
Reza
02
Fariba: nakheir nistan shomâ? no not+3rdPl. youF? no he’s not here who is speaking?
03
Reza
: bande keshani hastam goftan alân man khoonam ke I Keshani am said+3rdPl. now I home+1stSg that this is Keshani he said he’ll be home now so that
04
Reza
: zang bezanam beheshoon. ring make+1stSg. to himF. I call him.
05
Fariba: ah miân hamin alân. ah come+3rdPl. this now. ah he’ll be home in just a minute.
06 -> Reza
: alireza khan tashrif daran,? Alireza Mr. presence has+3rdPl.,? Is Mr. Alireza there,?
: khob bizahmat begin zang okay without taking any trouble tell+2ndPl. ring okay please tell him to
Requests and orders 07 -> Reza
08
351
: bezanan khooneye aghaye Tâheri make+2ndPl. home of Mr. Tâheri at Mr. Tâheri’s home
Fariba: chashm hatman okay definitely okay sure
After finding out that the acquaintance is not at home yet, Reza leaves him a message with his wife to call him at Mr. Tâheri’s house. Note that the acquaintance had told Reza to call him at home, and evidently he has not arrived home yet. Upon providing an account for his call (the third party had told Reza to call home at that time), which the wife responds to by assuring him that her husband will be home very soon, Reza produces an imperative begin zan bezanan khooneye aghaye Tâheri ‘tell him to call at Mr. Tâheri’s home’. Reza uses the imperative in a circumstance in which he can presume the wife’s willingness to give her husband the message. Furthermore, the account preceding the imperative provides contextual background and justification for Reza’s imperative. Reza’s request is preceded by the polite expression bizahmat ‘without trouble’. In doing so, Reza orients himself to the dispreferred nature of his activity, namely requesting the wife of an acquaintance to perform a service. Considering the formal nature of the relationship between Reza and the acquaintance’s wife, it seems that the polite expression bizahmat may serve the function of softening the imposition of the request. To sum up, a common feature in these imperatives is that they are prefaced by some components or types of talk that may project some specific type of action. More specifically, it seems that the prefatory talk to the imperative utterance provides the recipient with some information or grounds for their upcoming request that is formulated in the imperative form. Furthermore, it was suggested that imperatives were used in circumstances in which the speakers, due to details of the prior talk, assumed recipient’s willingness to perform the activity. By formulating their request as imperatives, speakers do not give their recipients an option not to perform the action. Another feature that most of the examples share is that the speakers provide some sort of account for their requested action. It has been suggested that accounts in request sequences may function as remedial work and that speakers frequently provide background information in their account turns in order to make their request not only more understandable, but also more acceptable, thereby avoiding any threat to the social relationship between themselves and their recipients (Taleghani-Nikazm 2006).
352 4.
Carmen Taleghani-Nikazm
Concluding remarks
Using the framework of conversation analysis, the present study has analyzed the grammatical construction and interactional placement of requests in the form of interrogatives and imperatives. Examining requesting as a sequentially and/or contextually implemented social activity, the study has demonstrated how the distribution of the different syntactic constructions used by Persian and German speakers to perform requests is systematically related to the contextual situation and the sequential placement of the request. Moreover, the cross-linguistic examination of the data sets suggests that the speaker’s selection of one particular request form, for example an interrogative with modal can, over another, is distinctively linked to particular types of interactional environments. Persian and German speakers commonly choose the self-oriented interrogative can I talk to X? to perform switchboard requests in telephone openings. In particular, it was observed that this construction of request typically occurred in an interactional environment in which the speaker first established the intended party’s proximity and formulated the request in terms of the recipient’s willingness to perform the requested action. Furthermore, the study has shown that the other-oriented interrogatives with modal can you … in German and Persian cultures were produced in circumstances in which the requested action may be performed by the speaker.However, considering the circumstances and the practicality (speakers not having immediate access), the speaker chooses to ask the recipient to perform the requested action. It was suggested that speakers’ particular grammatical composition of their requests displays their understanding of and orientation to the contextual knowledge of the social situation and the interactional environment in which the request is produced; speakers formulated their requests in terms of the recipient’s ability and likelihood of performing the requested action. The speaker’s choice of an imperative construction seems to be linked to some contextual factors, such as the speaker’s assumption about the recipient’s possible willingness to perform the requested action that may have been demonstrated in the details of the prior talk and the emerging context. Overall, this conversation analytical study and its findings is part of a larger research program (Curl and Drew 2008; Heinemann 2006; Lindström 2005; Wootton 2005) which involves the investigation of interactional environments in which speakers select one grammatical composition of a particular social activity rather than another. In addition, the current study is an attempt to address Levinson’s (2006) call for more cross-linguistic and cross-cultural research that focuses on the structure of conversation and explores commonalities in terms of language and social interaction in different cultures. This cross-linguistic study has demonstrated that, unlike what has been claimed by previous studies on request and its varieties, speakers’ grammatical composition of social action may not solely be influenced by the social factors such as age, social status, and gender, but it may be distinctly
Requests and orders
353
associated with the details of the context and interactional environment in which requests are issued. In addition, the request turn design may be governed by universal pragmatic principles underlying the social act of request. It has been shown that speakers in both cultures utilize strategies such as providing an account for their requests explaining or justifying their requestive action which may be understood as an affiliative move addressing the dispreferred nature of requests. Finally, the analysis suggested that the sequential placement and content of accounts for requests in Persian and German conversations are organized by the underlying structure of social interaction, i.e. by the management of self- and other relationships and the maintenance of social solidarity.
Notes 1. For studies on the linguistic realization of requests in various cultures see: Blum-Kulka and Olshtain (1984); Blum-Kulka (1987) for Hebrew; Blum-Kulka, S., House, J. and Kasper, G. (1989) for American English, Canadian French, Hebrew, Argentinian Spanish, Russian, German, Thai; Brown and Levinson (1987) for English, Tamil and Tzeltal; Eslamirasekh (1993); Moosavie (1986) for Persian; House and Kasper (1981); House (1989) for German; Fukushima (1996) for Japanese; Garcia (1993) for Spanish; Koike (1989, 1994) for Spanish and English; Leech (1983) for English; Márquez Reiter (2000) for Uruguayan Spanish and British English; Pair (1996) for Spanish and Dutch; Sifianou (1993) for Greek; Trosborg (1995) for Danish and English; Van Mulken (1996) for French and Dutch. 2. For their cross-cultural study of request, Blum-Kulka et al. (1989) developed a coding scheme in which requests are made up of two parts, namely the “head act” which is the core utterance that fulfils the function of requesting and the various “peripheral” elements, which mitigate or aggravate the force of requesting. When occurring within the request head act these elements and their functions are referred to as “Internal” modifications. Requests may also be modified by “external” devices localized not within the same “head act” but within its immediate context (see also Faerch and Kasper 1984, footnote 4). 3. House and Kasper’s (1981) data corpus consisted of 24 situations in which the role relationship between the interlocuters (i.e. authority or lack of authority) and social distance (i.e. formality and informality) varied. 4. After viewing their collected data, House and Kasper (1981) developed a schema of 8 levels of directness, with level 1 being the most indirect and level 8 the most direct. For example, they categorized “mild hints” as the most indirect of requests. In their study, mild hints refer to utterances that do not explicitly express their illocutionary force; therefore, the recipient of such an utterance (who is actually the analyst(s), considering the data collection method) must discover what the utterance (locution) implies. 5. The Cross-Cultural Speech Act Realization Project (CCSARP) begun in 1984 was a collaborative effort among linguists to empirically study the speech act of requests and apologies in various languages in terms of cross-cultural variation. Studies performed in coordination with the CCSARP have typically used a discourse completion test (DCT),
354
6.
7.
8.
9.
10. 11.
12.
Carmen Taleghani-Nikazm
or, a written questionnaire that incorporates varying degrees of social distance and domi nance. In addition to designing the DCT, the CCSARP developed a coding scheme in order to standardize the terminology used across the languages. Blum-Kulka et al. (1985) describe alerters as elements that serve as attention getters, which may affect the social impact of the utterance. They may either act as a downgrader that may mitigate the act or as an upgrader that intensifies its degree of coerciveness. For a detailed discussion on concerns about using DCTs and whether we can assume that the participants’ responses are reflective of what would occur in natural discourse, see Golato (2002). The speakers in these phone conversations were graduate students from German universities who were on an exchange program in a German graduate program at an American university. These speakers of German had resided in the United States for two years and two months at the time of recording. During their graduate studies, they were in close contact with other German native speakers in their department and on campus. When examining their recorded conversations, no systematic influence of their stay in the United States upon their interactions was observed, other than their usage of certain specific English vocabulary. Persian is normally written in the Arabic script and is read from right to left, whereas English is read from left to right. In order to avoid problems in transcribing (such as marking overlap), I opted for using the standard Roman script for utterances in Persian (Batani 1993). Note that the verb “to write” here refers to writing an email. It is important to note here that Mr. Tâheri owns a small trading business and Reza works in the office where his responsibilities range from delivery items (packages, documents) and taking care of small business and services for Mr. Tâheri, which may also include taking care of some private business for Mr. Tâheri. The Persian idiomatic expression ghorboonet and its variations literary mean ‘may I sacrifice myself for you’ and is used interchangeably with khahesh mikonam ‘please’.
Appendix Transcription conventions Selected Jeffersonian transcription conventions as described in Atkinson and Heritage (1984: ix-xvi). Note that punctuation is not used to mark conventional grammatical units but to capture characteristics of speech delivery: [ ] =
(0.5) (.)
start of simultaneous talk by two or more speakers (overlap talk) end of overlap talk latching between turns: An utterance by one speaker starts immediately after the end of another speaker’s utterance without the normal intervening beat of silence. silence; length of silence is time relataive to the speed of the surrounding talk micro pause (less than 1 tenth of a second)
Requests and orders
.hh hh haha/hehe (h) (hh) WORD word (word) ( ) >word
5
M:
ne:. die is sehr sehr gut no:. it is very very good no:. it is very very good
Annette is offering Markus food and right after he turned down the offer, he produces a compliment. In this and similar situations, compliments seem to be used to mitigate the dispreferred action and to potentially minimize face-threatening acts (Golato 2005: 95–105) and to build rapport.6, 7 In preferred environments, compliments are used as seconds to self-deprecations (and hence may have been “solicited” by the self-deprecating speaker (Lambert 1998: 168; Golato 2005: 107–108)) and as responses to tellings, informings, and announcements. Moreover, they can be produced as part of expressions of thanks or as part of acceptances (both in second and in first pair parts (Golato 2005: 116–120)). A final observation is that they occur in turns in which speakers indicate that they have just noticed something, i.e., noticings (data sample 1 is an example of a noticing). Which function a compliment plays in interaction is very much intertwined with the local context, i.e., with the sequential structure, in which it is produced. 3.1.2.
Oh and mmmh as compliments
In addition to the semantic features of compliment turns described in section 3.1.1, compliments can also be composed of non-verbal elements as in the data sample below. Data sample 3: [Fischessen: 0:04:26:00 nudellas] S is putting bowl of pasta on the table 1
D:
((whistles, looks at S, raises eyebrows))
2
B:
O::::H [es gibt nudellas O::::H [there are noodellas8 O:::H: [we are having noodellas [
366 3
Andrea Golato U:
4
[mhmm:: [mhmm:: (.)
5
S:
6
U:
7
[vielleicht brauchn wir das hier [maybe need we this here [maybe we need this here [ [nudellas? [noodellas? (.)
8
U:
nutellas? nutellas?
9
S:
für den fisch. for the fish.
In this excerpt from a dinner table conversation, Sybille (S) is placing food on the table. All coparticipants are complimenting Sybille on the food. David (D) does so entirely non-vocally by whistling and raising his eyebrows. In overlap with David, Bernhard (B) produces an oh plus an announcement of the food item being served, and in line 3, Uschi (U) produces a gustatory mhmm (Gardner 1997; Wiggins 2001; 2002).9 In the following, I will analyze such appreciatory sounds (oh) and gustatory markers (mmmh) in detail. 3.1.2.1. Oh Compliment turns can either have an oh as the only element of the turn, or else feature oh combined with other compliments. The question that arises is what interactional function is associated with the production of an oh. Is an oh plus a verbal compliment simply an intensified compliment, or does the oh serve an additional function? Within the field of functional linguistics, German oh is typically analyzed as an interjection (Kühn 1979; Wahmhoff and Wenzel 1979; Ehlich 1986; Kucharczik 1989; Zifonun, Hoffmann and Strecker 1997). In this body of research, it is described as underlining the expressive nature of an utterance or else as a marker of surprise. However, when analyzing oh in the context of conversational sequences, its function can be specified further. A conversation analytic study (Golato, under review) has compared German oh (i.e., ohs not used in compliment sequences but in other turns) with ach/achso10 and drawn comparisons with prior research on English oh (James 1972, 1974; Heritage 1984; Schiffrin 1987; Local 1996; Maynard 1997; Heritage 1998; Hutchby 2001; Heritage 2002; Bolden 2006). This study has shown that like English oh, German oh is typically used as a change-ofstate token; however, German oh is only used in those instances when the change of
Appreciatory sounds and expressions of embodied pleasure used as compliments
367
state is of an emotional kind. For example, it is used in responses to good and bad news and in turns expressing surprise (both positive and negative)11. In all instances, an oh turn communicates an emotional stance of the speaker, such as joy, pleasure, physical pain, unhappiness, disgust, etc. Depending on the type of emotion and depending on the sequential position of the oh (i.e., whether it is a standalone token or a turn-initial token, or whether it is part of an agreeing or disagreeing second pair part), its phonetic realization varies (for a detailed description of the varying intonation contours and their meaning, see Golato, under review). I argue (Golato, under review) that oh serves as a vehicle for embodying and expressing the emotion felt by the speaker. This emotion is not reported on, but instead is portrayed as being experienced at the moment when the oh is uttered. In contrast, the tokens ach/achso are used when cognitive changes-of-state are being expressed (e.g., moving from being uninformed to being informed, moving from non-understanding to understanding, etc.). Hence, in German different tokens are employed as markers of changes-of-state, whereas in English only one token is used. Having seen in the previous paragraph that oh is used in non-compliment sequences as a vehicle for expressing the ongoing experience of an emotional stance, we can turn to the discussion of oh in compliment sequences. Not all compliments contain an oh. As mentioned above, compliments can also be used in sequences that primarily serve other functions (such as reprimands, requests, etc.). In these compliments, single-turn oh and oh-prefaced utterances do not occur since the primary action is not complimenting or communicating one’s emotional stance. In addition, ohs do not occur in turns in which the compliment is produced as a summative statement of a prior experience. For instance, speakers complimenting a card game partner at the end of a round for having played well will produce compliments without ohs. Similarly, guests who produce compliments as part of leavetakings do not produce oh. However, compliments do contain an oh when the complimenter is experiencing the emotion while he or she is expressing it (e.g., speakers complimenting a card game player for playing well, at the moment when an excellent play is made). Data sample 3, here reproduced as data sample 4, illustrates this: Data sample 4: [Fischessen: 0:04:26:00 nudellas] S is putting bowl of pasta on the table 1
D:
((whistles, looks at S, raises eyebrows))
2
B:
3
U:
O::::H [es gibt nudellas O::::H [there are noodellas O:::H: [we are having noodellas [ [mhmm:: [mhmm::
4
(.)
368
Andrea Golato
As Sybille is placing the food on the table and at the moment he has clear visual access to the food, Bernhard produces an O::::H. It is produced more loudly and lengthened in relation to the surrounding talk; it starts out low, rises and then falls again, resulting in a bell-shaped intonation contour. This intonation contour seems typical of compliments in noticings. This oh is produced at the first noticing of the food, and as such involves a change-of-state from not knowing what is for dinner to knowing what will be served while simultaneously expressing joy at the prospect of eating the food. The same can be seen in the following excerpt in which Lina (who bakes bread frequently) places a loaf of bread wrapped in a towel on the table and announces that this is a gift for her guests Anton and Carmen. It becomes clear from the recording (not displayed) that Lina has given other bread loaves to Anton and Carmen before. Lina’s husband and another guest, Erna, are also present. Data sample 5 [M2U-29_21.44_weil3.brot]
1
L:
2
C:
3
L:
=> 4
C
5
A:
6
C:
7
L:
8
*L is placing the bread on the table *hier (.) das is für euch anton und carmen da hab ich-* *here (.) this is for you anton and carmen there have i-* [oach das is lie:b. [oh that is ni:ce. [ [*L is lifting the towel off the bread [*diesmal ein >ein ein< helles, *this time a >a a < light one, *C moves eyebrows together *OOH OOH *E breaks into a broad smile mmmh? [*mmmH [ [he he he he [ [das is ein drittel roggn ein drittel dinkel. also [this is a third rye a third spelt. so diesmal en eher helleres? this time a rather lighter one?
Lina’s announcement that she has another loaf of bread for Anton and Carmen is greeted by Carmen with a news receipt ach (slightly colored towards an o) and an expression of gratitude. However, when Lina is unwrapping the bread and when Carmen has first visual access to the food, she produces two loud ohs and her facial expression changes to one of experiencing pleasure. That is, an informing about an impending gift is first treated with an oach and the action of having offered a gift is being assessed with das is lie:b ‘that is ni:ce’. The object to be given away, however, has not been assessed as the recipients have not had (visual)
Appreciatory sounds and expressions of embodied pleasure used as compliments
369
access to it (Auer and Uhmann 1982; Pomerantz 1984). Lina’s action of unwrapping and presenting the bread is designed to elicit compliments from the coparticipants. And, indeed, at the moment when she first has visual access to it, Carmen produces an assessment of the gift item, thereby complimenting Lina. Gaining visual access to the bread constitutes a change-of-state from not being (visually) informed about the item, to being informed about it. This changeof-state has an emotional component, namely admiration for the baker, Lina, and joy at the prospect of eating the food. This emotion is also embodied through Carmen’s change in facial expression. Put differently, the oh is uttered at the moment the speaker has first access to the assessable and at the moment she is experiencing joy. Anton too produces a compliment in form of an mmmh (to be discussed below) and Erna breaks into a broad approving smile. We can see here that Carmen is upgrading her actions from a news receipt to an expression of gratitude, and then to a compliment, which consists of an oh and a full embodiment of (anticipated) pleasure. Ohs are not only produced in conjunction with compliments on food items, but also with a wide variety of other assessables. Moreover, it is not necessary for the complimenter to have visual access to the assessable in question. Both of these characteristics can be demonstrated with the next compliment paid in a telephone conversation between Kirsten and her friend Carla, who recently purchased and moved into a new home which she is currently describing to Kirsten: Data sample 6: [Kirsten, Tape 2, B47] 1
C:
2
sa:ch weißte so’n so’n so’n haus we:ll you know such a such a such a house swell you know such a such a such a house sneunzehnhundert ge[baut nineteenhundred bu[ilt built in nineteenh[undred [ [ja. [yes.
3
K:
4
C:
un dreimeterfünfzich hohe deckn and threemeterfifty high ceilings an ceilings three meters fifty high
=> 5
K:
o::h is aber toll o::h is PRT great o::h really great ((annoyed tone)) _______|_______ | |
6
C:
ja das is toll aber äh d- die wä- aber d-das zu yes that is great but uh- t- the wa- but t-that to yes that’s great but uh t- the wa- but wallpapering all
370
Andrea Golato
7
8
K:
tapeziern is we(h)niger toll [he wallpaper is le(h)ss great[he that is le(h)ss great [he [ [he [he [he ich mir vorstelln i me imagine
he he he he he he he he he he he he he ka(h)nn he he he he ca(h)n he he he he i ca(h)n
Carla provides a description of her house. The first installment (the age of the house) is greeted with a continuer ja. ‘yes.’, while the second installment (the high ceilings) is greeted with an embodied compliment. That is, upon hearing this news, Kirsten produces a drawn-out oh which expresses her admiration, followed by a second compliment is aber toll ‘is really great’. Carla responds by first confirming the compliment (acknowledging that high ceilings are nice), however, she then rejects it, by providing a circumstance in which having high ceilings is less than desirable. Her annoyed tone when acknowledging the compliment is already an indication that she may not entirely agree with Kirsten’s positive assessment. In this sequence, a non-edible assessable was complimented with an oh to which the compliment giver had no visual access. Thus, the access to the assessable consists here of the verbal information and the imagination of the compliment giver. To sum up, ohs are used in compliments when a speaker is undergoing an emotional change-of-state, that is, when speakers are expressing a positive emotion at the moment they first have access to the assessable. Ohs are more than simply an appreciatory sound, but rather are expressions of embodied pleasure in that during their production, speakers can be seen to use facial expressions and bodily movements to indicate their joy, surprise or pleasure. In addition, oh turns and ohfronted turns are seen frequently in noticings and as part of expressions of thanks. 3.1.2.2. mmmh The mmmhs described in this section are typically called (de)gustatory expressions and have already been analyzed in Australian English (Gardner 1997, 2002), American English (Wiggins 2001, 2002) and German (Golato 2005). Even without phonetic analysis, it is clear that they are phonetically distinct from other types of mhms (as for example the mhms described by Gardner 2002). For instance, gustatory mmmhs and confirming mhums ‘uhums’ are two distinct sounds since the former are uttered as one lengthened sound with a rise-falling intonation, while the latter contain a glottal stop in the middle (Gardner 2002: 65; Golato 2005). Speakers use gustatory mmmhs in order to embody pleasure in the activity in which they are engaged (Wiggins 2001, 2002). Typically, such activities are eating and drinking (Wiggins 2001; Gardner 2002; Wiggins 2002; Golato 2005), smoking, or sex (Gardner 2002: 75). The tokens are also used in anticipation of such activities.
Appreciatory sounds and expressions of embodied pleasure used as compliments
371
The following excerpt exemplifies the use of an mmmh used as a compliment. Here, Brit (B) and her son Jonas (J) are having breakfast with Brit’s friends Erna (E) and Tim (T). Erna is eating a roll with homemade elderberry jam. Data sample 7: [Brit044_min35_mit.butter] 1
J:
… in der küche irgendwo.= in the kitchen somewhere.=
=> 2
E:
=mmmh mmm=
3
B:
=is gut oder?= =is good right?=
4
E:
=das schmeckt super. =this tastes super.
5
B:
6
T:
mh[m uh [hm [ [((clearing his throat and looking at E))
The prior question-answer sequence comes to an end in line 1. In line 2 Erna, who has just taken the first bite of a roll with homemade elderberry jam, produces two gustatory mmmhs in a row, stressing the second one. These mmmhs embody the pleasure she is experiencing while eating the jam. The mmmhs are simultaneously a compliment to the hostess, who orients to the turn as a compliment in that she responds with one of the typical ways of accepting a compliment in German, namely through a second assessment with a response pursuit marker (Jefferson 1980; Harren 2001) which triggers a second round of compliments (see section 3.2. below as well as Golato 2002, 2005). The data segment also illustrates that gustatory mmmhs can be produced fairly soon after a speaker has gained access to the assessable as they can be uttered while the speaker is ingesting the food (Gardner 1997; Wiggins 2001; Gardner 2002; Wiggins 2002). In contrast, talking and eating at the same time would carry a potential stigma, not to mention the risk of choking (Wiggins 2002). However, a gustatory mmmh is not only produced during the evaluation process, i.e., when the speaker is first tasting or seeing the assessable, but can also be part of a summative statement at the end of a meal. In the following segment, Markus, Christine and Annette are having breakfast and Markus has just finished a piece of cantaloupe that was offered to him by the hostess Annette: Data sample 8 [Frühstück in Texas Tape 1, A556] 1 => 2
(1.0) M:
m::::: m::::: m:::::
372
Andrea Golato
3
C:
uups [he ((C spilled tea)) oops [he oops [he [ [ach [well [well
4
A:
=> 5
M:
lecker tasty tasty
6
A:
7
C:
[( [( [ [( [(
8
M:
9
C:
10
A:
) ) ) )
*M is putting hand on tea pot möchts *noch bis ( ) want+you *still lit ( ) do you want some more ( ) ne: ( no: ( no: (
) ) )
*turning to M *da is noch mehr du kannst gerne noch en stück essen there is still more you can please another piece eat there’s more you are welcome to eat another piece
In line 1, Markus is eating the last bit of cantaloupe and then utters the gustatory token mmmh, thereby expressing his pleasure in the food. This expression of pleasure constitutes a positive assessment, which he further verbalizes in line 5. Since Markus is positively evaluating a food item offered by the hostess, he can be said to compliment the hostess. After a short insertion sequence which deals with Christiane spilling her last bit of tea, Annette responds to the compliment by treating it as a veiled request and offers additional fruit to Markus in line 10. 3.1.3
Ohs changing to mmmhs
As mentioned above, ohs are change-of-state tokens that express an emotion; thus, oh-fronted compliments occur as an expression of the speaker’s positive stance at the moment the speaker is experiencing or taking this stance. Furthermore, mmmhs are regularly produced with ingestible items. It thus should come as no surprise that there are also compliments that contain a shift from an oh to an mmmh as in the data sample below. Bernhard (the host) and Freddie are on the porch, discussing what to have to drink. Bernhard’s response to Freddie’s request for a glass of white wine is displayed in line 1.
Appreciatory sounds and expressions of embodied pleasure used as compliments
373
Data sample 9: [Doppelkopf, 0:00:54:00] 1
B:
weißwein. white wine12 is auch immer gut. white wine. white wine is also always good. white wine. white wine is always a good choice. (0.2) ((B went to fridge in kitchen or basement to get the (0.2) wine, F turned around at the same time, and spots the (0.2) roses ))
=> 2
F:
3 4
habt ihr schöne ro:sen hier. have you+pl. nice ro:ses here. you’ve got nice ro:ses here. (0.2)
F:
nein. no. wow. ((F walks over to roses in flower pot smells the tallest rose)) _|_ | |
5 6
7
(2.5) F:
o::::hm, a::::hm, a::::hm, (0.5)
In line 1, Bernhard receipts Freddie’s request for a glass of white wine and moves off camera into the house to fetch the drink. At the very moment that Bernhard turns around to move into the house, Freddie turns in the other direction and spots several flower pots of roses across the yard. He pays Bernhard a compliment on the roses (lines 2 and 4), but does not receive any uptake because unbeknownst to Freddie, he is alone on the porch at that time. Freddie then walks over to one of the flower pots, and at the moment that he is inspecting the first rose close up, he produces a second compliment that starts out with an oh but ends in an hm. I suspect that upon noticing in detail one of the roses, Freddie produces the emotional change-of-state token to express his positive stance towards the flower. While doing so, he then was able to smell the highly fragrant flower and changed the token in mid-production to a gustatory one. Again, he receives no uptake as he is without a conversational partner. He notices this fact a few seconds later. The flowers have apparently left such a positive impression on him though that later in the conversation he compliments the hosts on them. In sum, gustatory mmmhs are produced as compliments only with items that are ingestible (either edible or inhalable), while ohs are used as compliments for food and non-food items alike. Both can be used as stand-alone items or can be combined with other compliments. Both ohs and mmmhs are means of expressing an
374
Andrea Golato
emotion/sensation as it is being experienced; however, ohs are only placed in sequential positions when the assessable becomes first available for an assessment, while mmmhs can also be used in summative statements. When used in the latter environments, mmmhs are typically treated as pre-requests. Recall that I described German compliments in their design as non-speaker centered and fairly referential, both in terms of their syntactic and semantic design. Appreciatory sounds and gustatory expressions afford German speakers the opportunity to express the compliment in a more speaker-centered fashion, i.e., in terms of the speakers’ likes and experiences. 3.2.
Compliment responses
Compliment responses have been studied in much detail for both German (Schwitalla 1995; Lambert 1998; Golato 2002, 2005; Werthwein 2009), for different varieties of English (Pomerantz 1978; Wolfson 1981, 1984; Knapp, Hopper and Bell, 1984; Herbert 1986; Holmes 1986; Herbert 1989; Herbert and Straight 1989; Sims 1989; Herbert and Straight 1995; Werthwein 2009), and for English speaking learners of German (Huth 2006). In the previous section, I pointed out that compliments can serve different functions in interaction. As it turns out, the compliment function has an impact on the compliment response. Put differently, depending on the function of the compliment, the compliment is responded to in different ways. When compliments are produced in the course of other actions, specifically, when they are produced in dispreferred first pair parts (such as a criticism), the coparticipants typically respond to the dispreferred action and not to the compliment itself (Golato, 2005: 185–186). Data segment 10: [Quartet Material, 4/12/94] 1
Mike: okay, (0.6) (hit it)
2 Bob:
((music 10.0[))-----------------[ [(That’s the place),
5
Bob:
-------------] ] (Mike)/(now),] thats beautiful sound
6
Bob:
but, (.) try not to retard.=
7
Marg: =he didn’t.
8
(0.2)
3 4
=>
9 => 10 11
Bob:
°he didn’t?°=
Marg: =that was gr(h)(h)eat= Bob:
=oh I’m sorry
Appreciatory sounds and expressions of embodied pleasure used as compliments
375
In line 5, Bob pays Mike a compliment as part of an overall criticizing turn. Neither Mike nor the other participants reacts to the compliment. Instead, Marge is orientating to the criticism the compliment was embedded in by condradicting and challenging Bob. And in line 10, we can see Marge criticizing Bob by complimenting Mike. And again, Mike does not respond to the compliment but Bob reacts to the criticism by apologizing to Mike. When compliments are produced as part of dispreferred second pair parts, such as rejections to offers or invitations, the coparticipant either ignores the compliment and responds to the rejection, or the coparticipant confirms the compliment and then reissues the offer or invitation (Golato, 2005: 186–189). This can be seen in the following data segment: Data segment 11: [Frühstück in Texas Tape 1, B230] 1
C:
2
A:
3
eh [hä ha he eh [ha ha he eh [ha ha he [ [*A gazes at M [*(da is noch) melone [*(there is still melon [*(there’s still) some melon (.)
4
A:
wenn de noch willst if you still want if you want some more
5
M:
ne:. die is sehr sehr gut no:. it is very very good no:. it is very very good
6
(1.0)
7
M:
is das jetz honey dew? is that now honey dew? now is that honey dew?
8
A:
9
C:
mhm. (ich weiß [nich glaube) uhum. (i know [not believe) uhum. (I don’t [know think so) [ [ja ((clears throat)) [yes [yeah
Markus’s declination of Annette’s offer includes a compliment (line 5). We can see that Annette does not respond to this compliment as evidenced by the silence. Moreover, Markus does not treat the compliment response as relevantly absent as he switches topics in line 7. However, when compliments are produced in preferred
376
Andrea Golato
first or second pair parts, the complimenting action is responded to (Golato, 2005: 189–191). When interactants receive a compliment, they are put under two simultaneous but conflicting constraints (Pomerantz 1978). On the one hand, compliments are assessments since the speaker is positively evaluating some state of affairs or (an aspect of) a person. The preferred second pair part to an assessment in both German and English is an agreement with the assessment (Auer and Uhmann 1982; Pomerantz 1984). In addition, compliments are also said to be supportive actions such as offers, gifts, praises, etc., and as such their preferred second pair part is an acceptance. However, there is a constraint on participants to avoid self-praise (Pomerantz 1984: 89–90). Given these two constraints, a variety of compliment responses are observable in both English and German. Pomerantz notes that the most frequent response types in her corpus are “in-between” acceptances and rejections in their design. In American English, compliment responses of this type include, for instance, questioning the compliment (Those tacos were good! – You liked them … – I loved them), evaluation shifts such as downgrades (Oh it was just beautiful – Well thank you … uh I thought it was quite nice) and qualifications (Good shot. – Not very solid, though), referent shifts (i.e., reflecting the compliment away from self, You’re a good rower, honey – These are very easy to row. Very light), compliment returns (Ya sound (justiz) real nice – Yeah you soun’ real good too), comment histories (I love that suit. – I got it at Boscov’s), and reinterpretations (I like those pants – You can borrow them anytime) (Pomerantz, 1984: 89–90).13 These compliment response types have also been observed for German (Golato 2002, 2005; Werthwein 2009). In addition, both the German and American English corpora also feature compliment rejections and acceptances. While the rejections are produced in much the same way in both English in German, it is in the realm of compliment acceptances that there are differences between the two languages. Specifically, acceptance types found for German have not been found for English and vice versa: Germans accept compliments with assessments of the compliment assertion, with agreements in the form of confirmations, and with a same strength assessment followed with an agreement pursuit (for examples, see Golato 2002, 2005).14 In contrast, English speakers are seen to use appreciation tokens (i.e., thank you) and agreements in the form of second assessments. In terms of the form of compliment responses in German, Werthwein (2009) notes that the emotion and subjective stance of the compliment recipient is communicated mostly through certain verbs (finden ‘to find’, denken ‘to think’), interjections and particles. In addition to describing the form and function of compliment responses, Werthwein (2009: 175–187) also provides a very detailed analysis of the prosodic features of compliment responses. She concludes that prosodic cues and their illocutionary force are in line with the emotional stance of the compliment recipient.
Appreciatory sounds and expressions of embodied pleasure used as compliments
3.2.2.
377
Compliment responses to ohs and mmmhs
In terms of the responses they elicit, oh and mmmh behave differently; this difference is due to the types of sequences in which oh and mmmh compliments are embedded. Recall that oh never occurs in compliments that are produced as part of a dispreferred turn. As pointed out in section 3.2., when a compliment is produced in the course of a dispreferred action, the coparticipants respond to that dispreferred action and not to the compliment. Given that oh-fronted compliments do not occur as part of dispreferred actions, coparticipants always treat them as compliments. As such, they receive the entire range of compliment responses described in the literature review on compliment responses. Looking back at the data samples in this paper, we observe mainly responses of the “in-between” kind, i.e., an account (data sample 5) and a confirmation followed with a qualification (data sample 6). In data sample 1, no apparent response is given audibly, and unfortunately, the compliment recipient’s face is blocked by a lamp, hence it is impossible to discern if she is giving a non-vocal response. Other oh compliments in the collection contain a variety of acceptances and a rejection. In contrast to ohs, mmmhs can be part of some dispreferred actions, specifically requests. As pointed out above, speakers produce mmmhs not only when they first have access to the ingestible, but also when they have almost finished it. In these situations, coparticipants regularly treat the mmmh as a pre-request. That is, the responses that mmmhs in these positions receive are offers, as was seen in data sample 8, lines 8–10. In all other instances, i.e., when they are not used in dispreferred turns, mmmhs receive the entire range of compliment responses (acceptances, as in data sample 7, line 3, but also rejections, and inbetween responses). 3.3.
Compliments in multiparty interaction
Obviously, compliments are not only paid in dyadic conversations, but also in multi-party interactions. How third-parties react when one speaker has paid another a compliment depends upon their own access to the assessable, and upon the function the compliment serves in interaction (Golato, 2005). Just as in general assessment sequences (Auer and Uhmann 1982; Pomerantz 1984), in compliment sequences coparticipants can only provide an assessment if indeed they have (had) access to the assessable and are thus in a position to evaluate it. If they do not have access, this might itself become topicalized or else the coparticipant might assess the assessment assertion (e.g., That’s nice, Good for you, etc.). If it is possible to provide the third party of a compliment response with access to the assessable, compliment givers regularly do so as in data sample 12. This is the same as data sample 7 above in which Erna is eating homemade elderberry jam, though the sample contains additional talk:
378
Andrea Golato
Data sample 12: [Brit044_min35_mit.butter] 1
J:
… in der küche irgendwo.= in the kitchen somewhere.=
=> 2
E:
=mmmh mmm=
3
B:
=is gut oder?= =is good right?=
4
E:
=das schmeckt super. =this tastes super.
5
B:
6
T:
mh[m uh[hm [ [((clearing his throat and looking at E))
7
(2.0) ((E is chewing and nodding, reaching for jam))
8
E:
9
T:
10
*passing jam to T *sollteste mal testen. *should+you PRT test. *you should try it. *reaching for the butter mm *hm uh *hm (1.0)
11
E:
[vor allem] schmeckt das sehr gut mit butter. [above all] tastes it very good with butter. [it tastes] especially good with butter. [ ] [(du ich- ] [(hey i- ]
12
B:
13
E:
(>das hätt ich nich gedacht is wirklich gutthat had+Subj. I not thought is really goodI wouldn’t have thought so it’s really good 12
D:
*taking his first bite *oooh
13
U:
eh he
14
B:
very good.
Bernhard is noisily eating the food and then produces an ooooh indicating his pleasure. Since he is positively evaluating something for which Uschi can take credit (i.e., having selected and purchased the liqueur in question), he can be said to be complimenting Uschi. Annette, who had already been eating the ice-cream, agrees with a confirmation marker. Uschi, who has also tried the food item, greets the compliments by downgrading the assessable and offering this downgrade for agreement (line 3). Bernhard changes this assessment slightly (changing strong to tart), but immediately qualifies this utterance by stating that he likes the tartness. Annette provides an account for the tartness in line 5, which leads to several repair initiations on the word lime (left out). At this point, David is taking his first bite of the food and he too provides a compliment to Uschi. He, like Bernhard before him, does so with an ooooh and Bernhard agrees with him (in line 14). David’s compliment is produced much later than the other second compliments in the sequence. It is produced at a point when David has first access to the assessable. It is thus better
Appreciatory sounds and expressions of embodied pleasure used as compliments
381
viewed as a second first compliment than a second aligning compliment. (Note too, that it is produced after a compliment response had already been uttered by the compliment recipient.) The second example in which an oh is used in second position is data segment 3, reproduced here as data sample 14: Data segment 14: 1
D:
((whistles, looks at S, raises eyebrows))
2
B:
3
U:
O::::H [es gibt nudellas O::::H [there are noodellas O::::H [we are having noodellas [ [mhmm:: [mhmm::
David is producing a first compliment in line 1 which is entirely non-verbal. Bernhard produces an oh-fronted compliment in line 2. Given Sybille’s movements, who is placing the food on the table, Bernhard only gains visual access to the food a moment prior to when he utters the oh. Again, ohs in second position are only produced in situations in which the second oh producer had gained access to the assessable just prior to the oh; unlike mmmhs, they are never produced when the second compliment speaker has had access for a while and is still enjoying the food. Given this sequential placement and interactional function of oh, it might be more appropriate to analyze an oh-compliment produced after another speaker already has uttered a compliment as a second first compliment, rather than as a second aligning compliment. Unlike other sections of this paper, the discussion in the present section has focused entirely on second compliments in German multiparty interaction. To my knowledge, there is to date no study on second compliments in English or any other language. Future research would have to address whether the observations noted for German also hold for other languages.
4.
Concluding discussion
This paper has described how oh and mmmh are used in German compliment sequences in dyadic and multi-party conversation. It has been shown that both oh and mmmh are used in first compliment turns when participants express their pleasure as they are experiencing it and are often accompanied by gestures or facial expressions indicating pleasure. These compliments thus function as embodiments of pleasure. Oh was described as a token that expresses an emotional change-of-state. Speakers use stand-alone oh or oh in combination with other compliments when they first gain access to the assessable and are expressing their admiration at that
382
Andrea Golato
particular moment. Put differently, oh is a signal that the positive evaluation is taking place at the moment when the oh is uttered. Ohs assess both food and non-food items that speakers have either direct access to or are being told about. Mmmh is used only with ingestible objects (food, drink, odors) and it is employed either immediately when its speakers have gained access to the assessable or else after they have enjoyed the assessable for a while. Under the latter circumstance, the mmmhcompliment is typically treated as a pre-request in that the recipient offers a second helping of the ingestible in question. In other sequential slots, mmmh and oh behave similarly in terms of compliment responses in that they receive all of the responses that verbal compliments also receive. Similarly, in multi-party interaction, compliments realized with oh and mmmh put the same constraints on coparticipants to produce a second aligning compliment as verbal compliments do. These second aligning compliments can be realized with an mmmh (if the assessable is a food item), but they are not realized as oh-compliments. When oh-compliments occur after another person has already paid a first compliment, the oh-compliment is better analyzed as a second first compliment with which its speaker marks his/ her own independent evaluation of the assessable. The findings presented in this paper contribute to research in three different domains. They first and foremost contribute to research on compliments and compliment responses. The paper shows that compliments are not only expressed with lexical items, phrases or clauses, but that they can also be conveyed with response tokens or interjections. In addition to their evaluative nature, these tokens also communicate additional information. For example, an oh claims that the speaker has gained access to the assessable at the moment the oh is uttered, while an mmmh indicates an evaluation taking place at any moment during which an assessable is enjoyed by the compliment giver. Moreover, both oh and mmmh are used in order to embody pleasure in the object which is being evaluated. The analysis has shown that all parties to the interaction orient to these particular functions of oh and mmmh. Lastly, it was argued that appreciatory sounds and gustatory expressions afford German speakers the opportunity to express a compliment in more speakercentered fashion, i.e., in terms of the speakers’ likes and experiences. The findings also contribute to research on the use of tokens and interjections in interaction (Jefferson 1984; Gardner 1995, 1997; Sorjonen 2001; Gardner 2002; Drescher 2003; Golato and Fagyal 2006; Wilkinson and Kitzinger 2006; Betz and Golato 2008; Golato and Betz 2008; Barske 2009; cf. also Norrick, this volume). When used in the sequential contexts discussed in this paper, oh and mmmh are not simply uncontrolled reactions to outside events or utterances, but instead have a very specific communicative and discursive function. They are placed in very specific sequential contexts, and coparticipants orient to them as they also orient to other types of compliments. While oh and mmmh may appear to be uncontrolled reactions, they are actually interactional achievements, or as Wilkinson and Kitzinger put it “reaction tokens are not spontaneous, visceral eruptions, but rather are
Appreciatory sounds and expressions of embodied pleasure used as compliments
383
designed to appear as-if-visceral; that they are little performances of viscerality” (Wilkinson and Kitzinger 2006: 161). An additional contribution of the research presented in this paper concerns the meaning and function of oh. Prior research has identified specific functions of individual German tokens. For instance, German okay is relevant in the organization of sequences (Barske 2009), achso “oh I see” signals understanding (Golato and Betz 2008; Golato 2010), achja indicates remembering (Betz and Golato 2008), and ach marks an informing (Golato and Betz 2008; Golato 2010). In the present paper, we have seen that ohs also indicate a change-of-state, but rather than indicating a cognitive change-of-state (as ach), they mark an emotional change-of-state. In this regard, the paper also makes a contribution to the growing body of conversation analytic research on emotions (e.g. Jefferson and Lee 1981; Selting 1995; Christmann and Günthner 1996; Günthner 1997; Maynard 1997; Whalen and Zimmerman 1998; Fiehler 2002; Drescher 2003; M. H. Goodwin 2006; Wilkinson and Kitzinger 2006; C. Goodwin 2007; Local and Walker 2008). In contrast to research on emotions conducted within different methodological frameworks, conversation analytic research views emotions not primarily as internal or psychological experiences but rather as social expressions “which are manifested, interpreted, and processed together communicatively in the interaction for definite purposes by the persons involved” (Fiehler, 2002: 79). The chapter showed how the emotions of admiration and/or pleasure and reactions to them are jointly constructed by the participants. Moreover, the analyses showed that displays of this particular emotion can be used for different interactional goals depending on the sequential position in which the expression is placed and on how it is responded to.
Appendix: Transcription conventions . ? , ? : = – underlined CAPITALIZED superscript zero (( )) [ > < < >
Falling intonation Rising intonation Slightly rising intonation Rising pitch in the next syllable/phrase Lengthened speech Latching speech Cut off word Stressed syllable Higher volume Beginning and end of quieter speech Vocal effects accompanying speech Beginning of overlap of speech Speech faster than surrounding talk Speech slower than surrounding talk
384
Andrea Golato
(number) (.) (text) ( ) hhh .hhh PRTL * and text in italics
Duration of silence in seconds Micro pause Transcriber’s best guess at what is being spoken Inaudible speech Audible out-breath (laughing voice) Audible in-breath Particle not translated into English Non-verbal actions occurring without speech or at around the same time as the speech in the line immediately below
Notes 1. I would like to thank Emma Betz, Thorsten Huth, and Carmen Taleghani-Nikazm for providing me with additional data excerpts and Peter Golato and the CA research group at the University of Wisconsin, Madison for comments on an earlier version of this paper. All remaining errors are my own. 2. For a list of studies see Golato (2005), for a review of additional, more recent studies, see Werthwein (2009). 3. As the discussion will show, the gustatory marker mmmh is associated with joy in relation to eating and drinking (Wiggins, 2001; Gardner 2002; Wiggins 2002; Golato 2005), smoking, or sex (Gardner 2002: 75). The appreciatory sound oh however can be used to express emotions of a variety of kinds (Golato, under review) in relation to any number of activities and objects. 4. For descriptions of compliments and compliment responses in German based on discourse completion tasks and/or interviews, see Nixdorf (2002), Mulo Farenkia (2004, 2005, 2006), Probst (2003). 5. For a detailed discussion of preference organization, see, e.g., Pomerantz (1984) and Schegloff (2007). 6. This also seems to be one of the reasons why compliments are used in surveys (Gathman, Maynard and Schaeffer 2008). 7. But see Lambert (1998: 170–171) for an example, in which a compliment sequence is used by participants as a vehicle for negotiating power, dominance, and distance. Similarly, Holmes (1995: 119) noticed that compliments can have a “darker side” and be heard as sarcastic or as a put-down. 8. Nudellas is a play on word, possibly to make it sound Italian. Nudellas sounds like a cross between nudeln ‘noodles’, and nutella, a well-known hazelnut and chocolate spread – a connection Uschi and David vocalize in a subsequent repair sequence (not displayed). 9. Wiggins (2002) uses the term gustatory marker while Gardner (1997) uses the term degustatory marker. Since Wiggins analyzed the marker in greater detail, I adopted her labeling. 10. There is a functional difference associated with the tokens ach and achso. While ach merely marks the receipt of prior information, achso marks the understanding of that prior information (Golato and Betz 2008; Golato 2010). 11. For detailed conversation analytic work on surprise, see Wilkinson and Kitzinger (2006).
Appreciatory sounds and expressions of embodied pleasure used as compliments
385
12. I have no explanation for the use of English in this particular segment. 13. For a description of these response types in other varieties of English, see the studies listed at the beginning of this section. 14. Studies employing discourse completion tasks, such as Nixdorf (2002), Golato (2003), and Mulo Farenkia (2004), have yielded appreciation tokens as German compliment responses. However, see the discussion in Golato (2003) that tries to show that this data elicitation technique taps native speakers’ intuitions or prescriptive knowledge rather than actual target language use. Werthwein (2009: 174) also found appreciation tokens in her corpus of spontaneously spoken German data (9.6 % of response types). However, her data were elicited by her conducting unguided conversations with subjects. Throughout each conversation which lasted between 30–60 min, the researcher paid 2–13 compliments. This seems to be a somewhat high number, given that in my 2005 corpus, 36 hours of naturally occurring data yielded a total of 62 compliments – less than two compliments per hour (disregarding for a moment that we do not produce compliments per hour/minute, see Schegloff (1993)). While subjects were unaware of the goal of Werthwein’s study, some subjects commented on having noticed that she paid compliments frequently but thought that was her conversational style. Further research will have to show whether the frequency of compliments in Werthwein’s study led to a higher occurrence of appreciation tokens, or whether appreciation tokens are a viable but infrequent response type in naturally occurring data. 15. Note, however, that in data segment 8, this was not the case. Tim’s “missing” compliment may not have been heard as relevantly absent since he only gained access to the assessable after the compliment sequence had come to a close and the interaction had moved on to other conversational matters in which all participants were involved. 16. There is no direct translation of ne, but in function it is similar to tags or right.
References Atkinson, J. Maxwell and John Heritage (eds.) 1984 Structures of Social Action. Studies in Conversation Analysis. Cambridge: Cambridge University Press. Auer, Peter 1991 Vom Ende deutscher Sätze. ZGL 19: 139–157. Auer, Peter 1993 Zur Verbspitzenstellung im gesprochenen Deutsch. [The verb in first position in spoken German]. Deutsche Sprache 3: 193–222. Auer, Peter and Susanne Uhmann 1982 Aspekte der konversationellen Organisation von Bewertungen. [Aspects of the conversational organization of assessments]. Deutsche Sprache 10: 1–32. Barske, Tobias 2009 Same token, different actions: A conversation analytic study of social roles, embodied actions, and ok in German business meetings. Journal for Business Communication 46(1): 120–149.
386
Andrea Golato
Betz, Emma and Andrea Golato 2008 Remembering relevant information and withholding relevant next actions: The German token achja. Research on Language and Social Interaction, 41(1): 58–98. Bilmes, Jack 1988 The concept of preference in conversation analysis. Language in Society 17: 161–181. Bolden, Galina 2006 Little words that matter: Discourse markers “so” and “oh” and the doing of other-attentiveness in social interaction. Journal of Communication 56: 661–688. Burkhardt, Armin 1982 Gesprächswörter. Ihre lexikologische Bestimmung und lexikographische Beschreibung. In: Wolfgang Mentrup (ed.), Konzepte zur Lexikographie. Studien zur Bedeutungserklärung in einsprachigen Wörterbüchern, 138–171. Tübingen: Niemeyer. Chen, Rong 1993 Responding to compliments. A contrastive study of politeness strategies between American English and Chinese speakers. Journal of Pragmatics 20(1): 49–75. Christmann, Gabriela B. and Susanne Günthner 1996 Sprache und Affekt. Die Inszenierung von Entrüstungen im Gespräch. Deutsche Sprache 24: 1–33. Drescher, Martina 2003 Sprachliche Affektivität. Darstellung emotionaler Beteiligung am Beispiel von Gesprächen im Französischen. Tübingen: Max Niemeyer. Ehlich, Konrad (ed.) 1986 Interjektionen. Tübingen: Niedermeyer. Eisenberg, Peter 1999 Grundriss der deutschen Grammatik. Der Satz. Vol. 2. Stuttgart/Weimar: J. B. Metzler. Fiehler, Reinhard 2002 How to do emotions with words: emotionality in conversation. In: Susan R. Fussell (ed.), The Verbal Communication of Emotions. Interdisciplinary Perspectives, 79–106. Mahwah, NJ: Lawrence Erlbaum. Ford, Cecilia E. and Barbara A. Fox 1996 Interactional motivations for reference formulation: He had. This guy had, a beautiful, thirty-two o:lds. In: Barbara A. Fox (ed.), Studies in Anaphora, 145–168. Amsterdam/Philadelphia: John Benjamins. Fox, Barbara A. 1987 Discourse Structure and Anaphora. Cambridge: Cambridge University Press. Fox, Barbara A. (ed.) 1996 Studies in Anaphora. Vol. 33. Amsterdam/Philadelphia: John Benjamins. Gardner, Rod 1995 On Some Uses of the Conversational Token Mm. Unpublished PhD. Dissertation, University of Melbourne.
Appreciatory sounds and expressions of embodied pleasure used as compliments
387
Gardner, Rod 1997 The conversation object Mm: A weak and variable acknowledging token. Research on Language and Social Interaction 30(2): 131–156. Gardner, Rod 2002 When Listeners Talk: Response Tokens and Listener Stance. Amsterdam/Philadelphia: John Benjamins. Gathman, E. Cabell Hankinson, Douglas W. Maynard, and Nora Cate Schaeffer 2008 The respondents are all above average: compliment sequences in a survey interview. Research on Language and Social Interaction, 41(3): 271–301. Golato, Andrea 2002 German compliment responses. Journal of Pragmatics 34: 547–571. Golato, Andrea 2003 Studying compliment responses: A comparison of DCTs and recordings of naturally occurring talk. Applied Linguistics 24(1): 90–121. Golato, Andrea 2005 Compliments and Compliment Responses: Grammatical Structure and Sequential Organization. Amsterdam/Philadelphia: John Benjamins. Golato, Andrea 2010 Marking understanding versus receipting information in talk: achso and ach in German interaction. Discourse Studies 12(2): 147–176. Golato, Andrea under review German Oh and Ach/Achso: change-of-state tokens for marking emotional and cognitive changes of state. Golato, Andrea and Emma Betz 2008 German ach and achso in repair uptake: A resource to sustain or remove epistemic asymmetry. Zeitschrift für Sprachwissenschaft 27: 7–37. Golato, Andrea and Zsuzsanna Fagyal 2006 Two countours, two meanings: The intonation of jaja in German phone conversations. In: Speech Prosody, 3rd International Conference Dresden, May 2–5, 2006. Abstract Book and CD-Rom Proceedings: TUDpress Verlag der Wissenschaft. Goodwin, Charles 2007 Participation, stance and affect in the organization of activities. Discourse & Society 18: 53–73. Goodwin, Marjorie Harness 2006 Participation, affect, and trajectory in family directive/response sequences. Text & Talk 26(4): 515–544. Günthner, Susanne 1997 The contextualization of affect in reported dialogues. In: Susanne Niemeier and René Dirven (eds.), The Language of Emotions, 247–275. Amsterdam/ Philadelphia: John Benjamins. Harren, Inga 2001 “Ne?” in Alltagsgesprächen – Interaktive Funktionen und Positionierung in Turn und Sequenz. Unpublished M.A. Thesis, Universität Oldenburg, Oldenburg, Germany. Herbert, Robert K. 1986 Say “Thank you” – or something. American Speech 61(1): 76–88.
388
Andrea Golato
Herbert, Robert K. 1989 The ethnography of English compliments and compliment responses: A contrastive sketch. In: Wieslaw Oleksy (ed.), Contrastive Pragmatics, 3–35. Amsterdam/Philadelphia: John Benjamins. Herbert, Robert K. 1991 The sociology of compliment work: An ethnocontrastive study of Polish and English compliments. Multilingua 10(4): 381–402. Herbert, Robert K. and H. Stephen Straight 1989 Compliment-rejection versus compliment-avoidance: Listener-based versus speaker-based pragmatic strategies. Language and Communication 9(1): 35–47. Heritage, John 1984 A change-of-state token and aspects of its sequential placement. In: J. Maxwell Atkinson and John Heritage (eds.), Structures of Social Action. Studies in Conversation Analysis, 299–345. Cambridge: Cambridge University Press. Heritage, John 1998 Oh-prefaced responses to inquiry. Language in Society 27: 291–334. Heritage, John 2002 Oh-prefaced responses to assessments: A method of modifying agreement/ disagreement. In: Cecilia E. Ford, Barbara A. Fox and Sandra. A. Thompson (eds.), The Language of Turn and Sequence, 196–224. Oxford: Oxford University Press. Holmes, Janet 1986 Compliments and compliment responses in New Zealand English. Anthropological Linguistics 28(4): 485–508. Holmes, Janet 1988 Paying compliments: A sex-preferential positive politeness strategy. Journal of Pragmatics 12: 445–465. Holmes, Janet 1995 Women, Men and Politeness. London and New York: Longman. Holmes, Janet and Dorothy F. Brown 1987 Teachers and students learning about compliments. TESOL Quarterly 21(3): 523–546. Hutchby, Ian 2001 “Oh,” irony and sequential ambiguity in arguments. Discourse & Society 12: 123–141. Hutchby, Ian and Robin Wooffitt 1998 Conversation Analysis. Principles, Practices and Application. Cambridge: Polity Press. Huth, Thorsten 2006 Negotiating structure and culture: L2 learners’ realization of L2 complimentresponse sequences in talk-in-interaction. Journal of Pragmatics 38(12): 2025–2050. James, Deborah 1972 Some aspects of the syntax and semantics of interjections. Chicago Linguistic Society 8: 162–172.
Appreciatory sounds and expressions of embodied pleasure used as compliments
389
James, Deborah 1974 Another look at, say, some grammatical constraints on, oh, interjections and hesitations. Chicago Linguistic Society 10: 242–251. Janney, Richard W. and Horst Arndt 1992 Intracultural tact versus intercultural tact. In: Richard J. Watts, Sachiko Ide and Konrad Ehlich (eds.), Politeness in Language. Studies in its History, Theory and Practice, 21–41. Berlin and New York: Mouton de Gruyter. Jaworski, Adam 1995 “This is not an empty compliment!” Polish compliments and the expression of solidarity. International Journal of Applied Linguistics 5(1): 63–94. Jefferson, Gail 1980 The abominable Ne?. An Exploration of post-response pursuit of response. In: Peter Schröder and Hugo Steger (eds.), Dialogforschung. Jahrbuch 1980 des Instituts für deutsche Sprache, 53–88. Düsseldorf: Pädagogischer Verlag Schwann. Jefferson, Gail 1984 Notes on a systematic deployment of the acknowledgement tokens “yeah” and “Mm hm”. Papers in Linguistics 17: 197–216. Jefferson, Gail and John R. Lee 1981 The rejection of advice: Managing the problematic convergence of a ‘troublestelling’ and a ‘service’ encounter. Journal of Pragmatics 5: 399–422. Knapp, Mark L., Robert Hopper and Robert A. Bell 1984 Compliments: A descriptive taxonomy. Journal of Communication 34(4): 12–32. Kotthoff, Helga 1989 So nah und doch so fern. Deutsch-amerikanische pragmatische Unterschiede im universitären Milieu. Informationen Deutsch als Fremdsprache 16(4): 448–459. Kotthoff, Helga 1993 Interkulturelle deutsch-“sowjetische” Kommunikationskonflikte. Informationen Deutsch als Fremdsprache 20(5): 486–503. Kucharczik, Kerstin 1989 Sprecher- und hörerseitige Verwendungen der Interjektion HM. In: Sabine Kowal and Roland Posner (eds.), Zeitliche und inhaltliche Aspekte der Textproduktion, 168–191. Berlin: Institut für Linguistik. Kühn, Peter 1979 Aha! Pragmatik einer Interjektion. Deutsche Sprache 4: 289–297. Lambert, Margitta 1998 Zu unterschiedlichen Realisierungen des Komplimentmusters in Altentagesstätten- bzw. Altenheimkommunikation. In: Reinhard Fiehler and Caja Thimm (eds.), Sprache und Kommunikation im Alter, 161–174. Opladen/Wiesbaden: Westdeutscher Verlag. Local, John 1996 Conversational phonetics: Some aspects of news receipts in everyday talk. In: Elizabeth Couper-Kuhlen and Margret Selting (eds.), Prosody in Conversation. Interactional Studies, 177–230. Cambridge: Cambridge University Press.
390
Andrea Golato
Local, John and Gareth Walker 2008 Stance and affect in conversation: On the interplay of sequential and phonetic resources. Text & Talk 28(6): 723–737. Lorenzo-Dus, Nuria 2001 Compliment responses among British and Spanish university students: A contrastive study. Journal of Pragmatics 33: 107–127. Manes, Joan 1983 Compliments: A mirror of cultural values. In: N. Wolfson and E. Judd (eds.), Sociolinguistics and Language Acquisition, 96–102. Rowley, MA/London/ Tokyo: Newbury House Publishers. Manes, Joan and Nessa Wolfson 1981 The compliment formula. In: Florian Coulmas (ed.), Conversational Routine. Explorations in Standardized Communication Situations and Prepatterned Speech, 115–132. The Hague/Paris/New York: Mouton Publishers. Maynard, Douglas W. 1997 The news delivery sequence: Bad news and good news in conversational interaction. Research on Language and Social Interaction 20(2): 93–130. Mulo Farenkia, Bernard 2004 Kontrastive Pragmatik der Komplimente und Komplimenterwiderung. Aachen: Shaker Verlag. Mulo Farenkia, Bernard 2005 Kreativität und Formelhaftigkeit in der Realisierung von Komplimenten: Ein deutsch-kamerunischer Vergleich. Linguistics online 22(1/05). http://www.linguistik-online.de/22_05/mulo.html). Mulo Farenkia, Bernard 2006 Beziehungskommunikation mit Komplimenten. Ethnographische und gesprächsanalytische Untersuchungen im deutschen und kamerunischen Sprachund Kulturraum. Frankfurt: Peter Lang. Nixdorf, Nina 2002 Höflichkeit im Englischen, Deutschen, Russischen. Ein interkultureller Vergleich am Beispiel von Ablehnungen und Komplimenterwiderungen. Marburg: Tectum Verlag. Pomerantz, Anita 1978 Compliment responses. Notes on the co-operation of multiple constraints. In: Jim Schenkein (ed.), Studies in the Organization of Conversational Interaction, 79–112. New York, San Francisco, London: Academic Press. Pomerantz, Anita 1984 Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes. In: J. Maxwell Atkinson and John Heritage (eds.), Structures of Social Action, 225–246. Cambridge: Cambridge University Press. Potter, Jonathan 2002 Two kinds of natural. Discourse Studies 4(4): 543–548. Probst, Julia 2003 Ein Kompliment in Ehren … Aspekte eines “höflichen” Sprechaktes in mehreren Sprachen. In: Nicole Baumgarten, Claudia Böttger, Markus Motz and Julia Probst (eds.), Übersetzen, interkulturelle Kommunikation, Spracherwerb und Sprachvermittlung – das Leben in mehreren Sprachen. Festschrift
Appreciatory sounds and expressions of embodied pleasure used as compliments
391
für Juliane House zum 60. Geburtstag. Zeitschrift für interkulturellen Fremdsprachenunterricht, Vol. 8, 210–225. Ort: Verlag. Online: available at . Schegloff, Emanuel A. 1984 On some questions and ambiguities in conversation. In: J. Maxwell Atkinson and John Heritage (eds.), Structures of Social Action. Studies in Conversation Analysis, 28–52. Cambridge: Cambridge University Press. Schegloff, Emanuel A. 1988 Presequences and indirection. Applying speech act theory to ordinary conversation. Journal of Pragmatics 12: 55–62. Schegloff, Emanuel A. 1993 Reflections on quantification in the study of conversation. Research on Language and Social Interaction 26(1): 99–128. Schegloff, Emanuel A. 1996 Some practices for referring to persons in talk-in-interaction: A partial sketch of a systematics. In: Barbara A. Fox (ed.), Studies in Anaphora, Vol. 33, 437–485. Amsterdam/Philadelphia: John Benjamins. Schegloff, Emanuel A. 2007 Sequence Organization in Interaction. A Primer in Conversation Analysis, Vol. 1. Cambridge: Cambridge University Press. Schiffrin, Deborah 1987 Discourse Markers. Cambridge: Cambridge University Press. Schwitalla, Johannes 1995 Kommunikation in der Stadt. Vol. 4. Berlin/New York: Walter de Gruyter. Schwitalla, Johannes 1997 Gesprochenes Deutsch. Eine Einführung. Berlin: Erich Schmidt Verlag. Selting, Margret 1995 Prosodie im Gespräch. Tübingen: Niemeyer. Sims, Anntarie L. 1989 The compliment sequence. The Southern Communication Journal 54: 171–184. Sorjonen, Marja-Lena 2001 Responding in Conversation: A Study of Response Particles in Finnish. Vol. 70. Amsterdam/Philadelphia: John Benjamins. Spencer-Oatey, Helen, Patrick Ng, and Li Dong 2000 Responding to compliments: British and Chinese evaluative judgements. In: Helen Spencer-Oatey (ed.), Culturally Speaking: Managing Rapport through Talk across Cultures, 98–120. London and New York: Continuum. Valdés, Guadalupe and Cecilia Pino 1981 Muy a tus ordenes: Compliment responses among Mexican-American bilinguals. Language in Society 10(1): 53–72. Wahmhoff, Sibylle and Angelika Wenzel 1979 Ein HM ist noch lange kein HM – oder was heisst klientenbezogene Gesprächsführung? In: Jürgen Dittmann (ed.), Arbeiten zur Konversationsanalyse, 258–297. Tübingen: Niemeyer. Werthwein, Daniela Christina 2009 Already got a compliment today? Wie Australier und Deutsche verbal auf Komplimente reagieren. St. Augustin: Asgard.
392
Andrea Golato
Whalen, Jack and Don H. Zimmerman 1998 Observations on the display and managment of emotion in naturally occurring activities: the case of “Hysteria” in calls to 9–1–1. Social Psychological Quarterly 61: 141–159. Wiggins, Sally 2001 Construction and action in food evaluation. Journal of Language and Social Psychology 20(4): 445–463. Wiggins, Sally 2002 Talking with your mouth full: Gustatory mmms and the embodiment of pleasure. Research on Language and Social Interaction 35(3): 311–336. Wilkinson, Sue and Celia Kitzinger 2006 Surprise as an interactional achievement: Reaction Tokens in Conversation. Social Pscychology Quarterly 69(2): 150–182. Wolfson, Nessa 1981 Compliments in cross-cultural perspective. TESOL Quarterly, 15(2): 117–124. Wolfson, Nessa 1983 An empirically based analysis of complimenting in American English. In: Nessa Wolfson and Elliot Judd (eds.), Sociolinguistics and Language Acquisition, 82–95. Rowley/London/Tokyo: Newbury House Publishers. Wolfson, Nessa 1984 Pretty is as pretty does: A speech act view of sex roles. Applied Linguistics 5(3): 236–244. Wolfson, Nessa and Joan Manes 1980 The compliment as a social strategy. Papers in Linguistics 13(3): 391–410. Ylänne-McEwen, Virpi 1993 Complimenting behavior: A cross-cultural investigation. Journal of Multilingual and Multicultural Development 14(6): 499–508. Yuan, Yi 1996 Responding to compliments: A contrastive study of English pragmatics of advanced Chinese speakers of English. Proceedings of the Annual Boston University Conference on Language Development 20(2): 861–872. Zifonun, Gisela, Ludger Hoffmann and Bruno Strecker 1997 Grammatik der deutschen Sprache. 3 Bände. Berlin/New York: de Gruyter.
13.
Politeness and impoliteness Jonathan Culpeper
1.
Introduction
Thirty or so years ago politeness was a specialist, even somewhat esoteric topic, primarily located in pragmatics. Indeed, Penelope Brown and Stephen Levinson’s 1978 book – the book that was to become so central to the area – was in fact bundled with another piece on questions and published as part of a collection edited by Esther Goody. Today, seven of the most cited articles published by the Journal of Pragmatics involve politeness or impoliteness, according to the publisher’s website. Furthermore, the field now has its own dedicated journal: the Journal of Politeness Research. The surge in politeness studies has had profound effects on the study of pragmatics. As O’Driscoll (2007: 465) points out, Brown and Levinson’s “great achievement has been to put socio-pragmatic concerns at the forefront of pragmatic research and the affective aspects of communication firmly on the pragmatics map”. Moreover, the multidisciplinary nature of politeness studies, something which undoubtedly has contributed to its popularity, has been strengthened. Although the conceptual heart of the field is still located in pragmatics, models of politeness have been applied and sometimes refined in diverse disciplines, including psychology (especially social psychology), anthropology, sociology, cultural studies, literary studies and behavioural organisation. Section 2 immediately following is devoted to politeness and is the largest section in this chapter. After some consideration of definitions of politeness in section 2.1, the discussion is structured according to two waves in the politeness literature. The first, in section 2.2, involves the pioneers who created the “classic” models of politeness in the 1970s and 1980s, and also the scholars who applied their model and offered specific criticisms of it mostly in the 1980s and 1990s. What is articulated here is generally a more pragmatic view of politeness. The second, in section 2.3, involves the scholars who in the 1990s and 2000s argued for the rejection (or at least radical revision) of the classic models, some of whom also proposed alternative models. What is articulated here is generally a more socio-cultural view of politeness (a few alternative models also attend to core aspects of pragmatic theory in their proposals). Section 3 of this chapter focuses more closely on the new burgeoning subfield of impoliteness. And finally, the concluding section of this chapter returns to the definition of politeness and suggests what might be referred to as the attitudinal view of politeness.
394
Jonathan Culpeper
2.
Politeness
2.1.
Definitions of politeness
What might politeness encompass? For somebody who has been invited to dinner in England, politeness might include remembering to use please when you want something passed, complimenting the cook on the food and definitely not burping. Each of these three things involves complexities that work on politeness attempts to account for. The word please is the “magic word” that British parents impress upon their children to use with all requests, and it looms large in the British psyche. But how is it actually used by adults? Aijmer (1996: 166–8) provides some evidence. It matters how the rest of the request is worded: please is most likely to be used in conjunction with an imperative (e.g. “please make me a cup of tea”) or with could you (e.g. “could you please make me a cup of tea”), but much less likely to be used with can you or will you. Differences in situation would influence whether you use the word please. Please tends to be used in relatively formal situations, and in business letters and written notices. It is particularly frequent in service encounters, notably telephone service encounters. So, if the dinner were a formal invitation, please would more likely be used. Complimenting the cook on the food may seem a straightforwardly nice thing to do, but it is not: you place the recipient of the compliment in a rather tricky position. If they simply accept the compliment, they may sound rather immodest, but if they simply reject it, they may offend the person who made it. Consequently, responses to compliments tend to weave a path between these two positions. A response such as “it’s kind of you to say that” suggests that the compliment is (at least in part) a product of the complimenter’s kindness and not necessarily a true reflection of the value of the food. Finally, even burping cannot always with certainty be seen as the antithesis of politeness. Cultural considerations clearly come into play here. In some cultures (e.g. on the Indian subcontinent), burping may be acceptable, or even a sign of appreciation of the food – a compliment. Culture also keenly influences all aspects of politeness. The use of the word please is more typical of British culture than North American, being used about twice as frequently (Biber et al. 1999: 1098). This is not to say that American culture is less polite. There are other ways of doing politeness, and those other ways might be evaluated as polite by North Americans, just as using please in certain contexts might be evaluated as polite by British English people. (Im)politeness is in the eyes and ears of the beholder. Politeness, then, involves “polite” behaviours. What those behaviours, linguistic and non-linguistic, consist of, how they vary in context, and why they are considered “polite” are some of the key areas of politeness study. What exactly is politeness? This is one of the most intractable questions in the field, to which a multitude of answers have been proposed. Bargiela-Chiappini (2003: 1464) comments: “Despite the variety of studies which focus on linguistic politeness […] the
Politeness and impoliteness
395
field still lacks an agreed definition of what ‘politeness’ is.” Although this state of affairs is not conducive to the development of certain aspects of the field, work undertaken to develop definitions and approaches to politeness can at least help deepen one’s appreciation of the issues. Let us survey some of those definitions and approaches. The classic, and most frequently cited, politeness studies lean heavily towards a pragmatic view of politeness. Specifically, these studies have concentrated on how we employ communicative strategies to maintain or promote social harmony: [The role of the Politeness Principle is] to maintain the social equilibrium and the friendly relations which enable us to assume that our interlocutors are being cooperative in the first place. (Leech 1983a: 82) … politeness, like formal diplomatic protocol (for which it must surely be the model), presupposes that potential for aggression as it seeks to disarm it, and makes possible communication between potentially aggressive parties. (Brown and Levinson 1987: 1) Politeness can be defined as a means of minimizing confrontation in discourse – both the possibility of confrontation occurring at all, and the possibility that a confrontation will be perceived as threatening. (R. Lakoff 1989: 102)
Thomas summarises the research agenda of scholars like the above engaged in the study of pragmatic politeness: All that is really being claimed is that people employ certain strategies (including the 50+ strategies described by Leech, Brown and Levinson, and others) for reasons of expediency – experience has taught us that particular strategies are likely to succeed in given circumstances, so we use them. (Thomas 1995: 179)
In the dinner table scenario, an example would be the choice of linguistic strategy in order to achieve the goals of both being passed something and maintaining harmonious social relations, despite inconveniencing the target of our request. For example, one’s experience may lead one to decide that “could you pass the salt please” is more expedient at a formal dinner event than “pass the salt”. The socio-cultural view of politeness emphasises the social context. More specifically, the emphasis is on either (or more often both) social norms or the constructions of participants (i.e. the notions which participants use to understand each other rather than which researchers use to understand participants). Regarding social norms, this view of politeness is neatly summed up by Fraser (1990: 220): Briefly stated, [the socio-cultural view] assumes that each society has a particular set of social norms consisting of more or less explicit rules that prescribe a certain behavior, a state of affairs, or a way of thinking in context. A positive evaluation (politeness) arises when an action is in congruence with the norm, a negative evaluation (impoliteness = rudeness) when action is to the contrary. (Fraser 1990: 220)
Politeness, in this sense, subsumes notions such as “good manners”, “social etiquette”, “social graces” and “minding your ps and qs”. For example, parents teaching their children to say please typically proscribe requests that are not accompanied by that word. Note that social norms are sensitive to context: the social
396
Jonathan Culpeper
politeness norms that pertain to a family dinner are rather different from those pertaining to an invited formal dinner occasion. In fact, there are some situations where communicative behaviours are not subject to politeness prescriptions; in other words, situations in which behaviours which might be viewed as “impolite” are unrestricted and licensed. Often, such situations are characterised by a huge power imbalance, as might be the case in army recruit training. But not necessarily so: Harris (2001), for example, describes the sanctioned impoliteness that takes place in the UK’s House of Commons, giving Opposition MPs opportunities to attack the Government that they might not have had in other contexts. Regarding the constructions of participants, let us turn to Richard Watts, whose work on politeness, spanning more than two decades, culminated in his 2003 book, the most important work on politeness in recent years. He writes: We take first-order politeness to correspond to the various ways in which polite behaviour is perceived and talked about by members of socio-cultural groups. It encompasses, in other words, commonsense notions of politeness. Second-order politeness, on the other hand, is a theoretical construct, a term within a theory of social behaviour and language usage. (Watts et al. 2005a [1992]: 3; see also Watts 2003, Eelen 2001; my emphasis) A theory of politeness2 should concern itself with the discursive struggle over politeness, i.e. over the ways in which (im)polite behaviour is evaluated and commented on by lay members and not with ways in which social scientists lift the term “(im)politeness” out of the realm of everyday discourse and evaluate it to the status of a theoretical concept in what is frequently called Politeness Theory (Watts 2003: 9).
What Watts refers to as second-order politeness (or politeness2) is the stuff of the pragmatic view previously mentioned, or what in this chapter constitutes the first wave of politeness research. What he refers to as first-order politeness (or politeness1), which constitutes the second wave, is like the social norm view of politeness in that it connects with “commonsense notions of politeness”, but it is more specific in that it argues that politeness exists in the articulations of lay members and not researchers. One problem with approaching politeness in this way is that researchers are reticent to define politeness precisely, because we are to be guided by the definitions of participants. But one may wonder how we are to recognise a participant’s definition of politeness as such, if we have nothing to guide us. In fact, Watts (2003:14) does identify one “fundamental aspect of what is understood as “polite” behaviour in all […] cultures” and that is displaying “consideration” for others, something which is also partly reflected in the pragmatic view of politeness (Goffman 1967: 11 also notes the importance of “considerateness”). These different definitions of politeness have largely evolved as a consequence of different agendas. Watts is concerned with “developing a theory of social politeness” (2003: 9, et passim); the pragmatic approach has a different agenda: The starting point of pragmatics is primarily in language: explaining communicative behaviour. By studying this we keep our feet firmly on the ground, and avoid getting lost too easily in abstractions such as “face” or “culture”. The basic question is: What did s
Politeness and impoliteness
397
mean [to convey] by saying X?. It is useful to postulate the Politeness Principle (PP), I claim, not because it explains what we mean by the word “politeness” (an English word which in any case doesn’t quite match similar words in other languages), but because it explains certain pragmatic phenomena […] (Leech 2003: 104–5)
However, whilst there are important differences, not least of all ontological, there is common ground between these two views, and perhaps more than the quotation from Leech above in particular seems to acknowledge. Pragmatic choices are not made in a vacuum but in the light of repeated experience of social situations (and their associated norms) which may lead one to expect certain kinds of interaction to happen, to be able to hypothesise what others’ expectations are and to know how to meet them. And once interaction has started we monitor how participants are constructing and orienting to politeness and adjust our pragmatic choices accordingly. Meanings, including understandings of politeness, thus emerge in the flux of social interaction. I will return to these definitional issues in the final section of this chapter, but first let us survey the various approaches to politeness in more detail. 2.2.
First-wave approaches to politeness
2.2.1.
The classic models
The classic theories of politeness draw, as one might guess, on the classic pragmatic theories, notably, Conversational Implicature (e.g. Grice 1975) and Speech Act Theory (e.g. Austin 1962; Searle 1969). The bulk of the work in politeness studies has been based on or related to Brown and Levinson (1987). Before attending to that, I will outline an alternative theory. Maxim-based politeness Robin Lakoff (1973) was the first to posit a maxim-based view of politeness. In brief, she proposes that there are two rules of pragmatic competence, one being “be clear”, which is formalised in terms of Grice’s (1975) Cooperative Principle, and the other being “be polite”, which is formalised in terms of a Politeness Principle. The latter Politeness Principle consists of the following maxims: (1) Don’t Impose, (2) Give Options, and (3) Make your receiver feel good. Lakoff notes that sometimes the need for clarity would clash with the need for politeness, as later would also Leech (e.g. 1983a). But unlike Leech (e.g. 1983a), she goes further and argues that “it is more important in a conversation to avoid offense than achieve clarity” (1973: 297). There are indeed many occasions in which conveying a potentially offensive message implicitly is a means of upholding politeness. However, we must be careful not to assume that implicitness or indirectness always conveys politeness. For example, the utterance “you must have shit for brains” is more likely an implicit way of conveying impoliteness.
398
Jonathan Culpeper
Leech (1983a) is a much more developed maxim-based approach to politeness. Leech (1983a; see also 1977) posits the Politeness Principle, which is involved in “trade-offs” with the Cooperative Principle (Grice 1975). In fact, it lends the Cooperative Principle much explanatory power: the Cooperative Principle accounts for how people convey indirect meanings, the Politeness Principle accounts for why people convey indirect meanings. Let us illustrate how this might work with an example from a play (this is analysed in Leech 1992): [Context: The waiter, as the most tactful communicator, has been chosen to convey some bad news to Crampton, namely, that he is Philips’ father.] Waiter:
Crampton: Waiter:
[smoothly melodious] Yes, sir. Great flow of spirits, sir. A vein of pleasantry, as you might say sir … The young gentleman’s latest is that you’re his father. What! Only his joke, sir, his favourite joke. Yesterday I was to be his father … (G.B. Shaw’s You Never Can Tell, II, 248)
The waiter does manage to express the information about Crampton’s parentage. However, in order to avoid upset (i.e. to maintain politeness), he sacrifices the Maxim of Quality in pretending that this information is untrue, merely a joke. Leech (1983a:81) defines the Politeness Principle as follows: “‘Minimize (other things being equal) the expression of impolite beliefs […] (Maximize (other things being equal) the expression of polite beliefs’)” [there is a corresponding, but less important, positive version]. Note that the Politeness Principle is not confined to dealing with impolite beliefs. Leech’s Maxims allow for the minimisation of impolite beliefs and the maximisation of polite beliefs. For example, the direct command “Have a drink” would appear to be impolite in restricting the hearer’s freedom of action, but in fact it maximises the politeness of the belief that the target would wish to have a drink. The Politeness Principle consists of the following maxims: 1) (a) 2) (a) 3) (a) 4) (a) 5) (a) [(b) 6) (a) [(b)
TACT MAXIM (in impositives and commissives) Minimize cost to other [(b) Maximize benefit to other] GENEROSITY MAXIM (in impositives and commissives) Minimize benefit to self [(b) Maximize cost to self] APPROBATION MAXIM (in expressives and assertives) Minimize dispraise of other [(b) Maximize praise of other] MODESTY MAXIM (in expressives and assertives) Minimize praise of self [(b) Maximize dispraise of self] AGREEMENT MAXIM (in assertives) Minimize disagreement between self and other Maximize agreement between self and other] SYMPATHY MAXIM (in assertives) Minimize antipathy between self and other Maximize sympathy between self and other] (Leech 1983a: 132)
Politeness and impoliteness
399
The Tact maxim is discussed at some length (1983a: chapter 5). Leech states that tact will be influenced by the following social parameters (i) (ii) (iii) (iv)
the greater the cost A to h, the greater the horizontal social distance of h from s, the greater the authoritative status of h with respect to s, the greater will be the need for optionality, and correspondingly for indirectness, in the expression of an impositive, if s is to observe the Tact Maxim. (1983a: 127)
The first three parameters are also key to Brown and Levinson’s (1987) model, so I will not elaborate on them here. Leech is careful not to claim that these maxims apply universally to all cultures, but instead suggests that the Politeness Principle maxims may be weighted differently in different cultures (1983a:150). For example, the Tact Maxim might be a strong feature of some British cultures, Modesty of some Japanese cultures and Generosity of some Mediterranean cultures. Face-based politeness In the field of politeness, Brown and Levinson’s work (1987) is the best known and the most researched. In their work they attempt to relate the following aspects: face, facework and acts that threaten face, sociological variables influencing face threat, and five general ways (or “superstrategies”) of counterbalancing face threat with (at least some) specific linguistic strategies. I shall cover these aspects in the following paragraphs. What is face? Notions such as reputation, prestige, and self-esteem, all involve an element of face. The term is perhaps most commonly used in English in the idiom “losing face”, meaning that one’s public image suffers some damage, often resulting in humiliation or embarrassment. Such reactions are suggestive of the emotional investment in face. Although the concept of face seems to hail from China (cf. Hu 1944; Ho 1976), much modern writing on face draws upon the work of Goffman (e.g. 1967). Goffman defines it thus: the positive social value a person effectively claims for himself by the line others assume he has taken during a particular contact. Face is an image of self delineated in terms of approved social attributes. (1967: 5)
This is echoed in Brown and Levinson’s scheme, but there are important differences. One such difference is that they posit two related components of face. “Positive face” is defined (clearly echoing Goffman) thus: “the want of every member that his wants be desirable to at least some others … in particular, it includes the desire to be ratified, understood, approved of, liked or admired” (1987: 62). “Negative face” is defined as “the want of every competent adult member that his actions be unimpeded by others” (1987: 62). Note that face is couched in terms of
400
Jonathan Culpeper
psychological “wants”, and also that they assume these to be universal (“every member wants to claim for himself” (1987: 61)). Facework, according to Goffman, is made up of “the actions taken by a person to make whatever he [sic] is doing consistent with face” (1967: 12). Any action that impinges to some degree upon a person’s face (typically, orders, insults, criticisms) is a face threatening act (hereafter, FTA). People are generally motivated to avoid FTAs, and are willing to incur costs in order to save face (Brown 1970). Facework can be designed to maintain or support face by counteracting threats, or potential threats, to face. This kind of facework is often referred to as redressive facework, since it involves the redress of a FTA. Brown and Levinson’s discussion of politeness is confined to this kind of redressive facework. Brown and Levinson (e.g. 1987: 70) claim that politeness can be distinguished according to the type of face addressed, positive or negative. Thus, requests are typically oriented to negative face (they typically impose on one’s freedom of action) and criticisms to positive (they typically detract from the positive values one lays claim to). This distinction has been retained or stated as necessary by many other researchers (e.g. Baxter 1984; Craig et al. 1986; Penman 1990; Tracy 1990). Brown and Levinson also make a distinction – although not one that they elaborate on – between FTAs that primarily threaten the hearer’s face and those that primarily threaten the speaker’s face (1987: 65–8). Among the former they include orders, requests, threats, criticism, contradictions, and the mention of taboo topics; among the latter, expressing thanks, unwilling promises and offers, apologies, the breakdown of physical control over one’s body, and confessions. The selection of politeness is dependent in part on who has their face threatened, self and/or other, though, as Brown and Levinson (1987: 286) hint, it is not always the case that facework is targeted at the recipient of the apparent greatest face threat. Brown and Levinson make the assumption that it is of “mutual interest” (1987: 60) for interactants to cooperate by supporting each other’s face: “In general, people cooperate (and assume each other’s cooperation) in maintaining face, such cooperation being based on the mutual vulnerability of face” (1987: 61). A threat would lead to a counter-threat. Thus, the speaker has a vested interest in maintaining the hearer’s face, since this will enhance the probability of reciprocal facework. Brown and Levinson (1987) argue that an assessment of the amount of face threat of a particular act involves three sociological variables defined thus (summarised from 1987: 74–8): 1) Distance (D) is a symmetric social dimension of similarity/difference between the speaker and the hearer. It is often based on the frequency of interaction. The reciprocal giving and receiving of positive face is symptomatic of social closeness. 2) Relative Power (P) of the hearer over the speaker is an asymmetric social dimension. It is the degree to which a participant can impose his/her own plans and self-evaluation. Deference is symptomatic of a great power differential.
Politeness and impoliteness
401
3) Absolute Ranking (R) refers to the ordering of impositions according to the degree to which they impinge upon an interactant’s face wants in a particular culture and situation. Negative face impositions can be ranked according to the expenditure (a) of services (including the provision of time) and (b) of goods (including non-material goods like information, as well as the expression of regard and other face payments). Positive face impositions can be ranked according to the amount of “pain” suffered by the other, based on the discrepancy between the other’s self-image and that presented in the FTA.
For example, asking a new colleague for a cup of tea is more face threatening than asking a long standing colleague (the distance variable); asking one’s employer for a cup of tea is more face threatening than asking a colleague (the power variable); and asking for a glass of vintage port is more face threatening than asking for a glass of water (the ranking variable). Brown and Levinson argue that these three variables subsume all other factors that can influence an assessment of face threat. They suggest that numerical values could be attached to each variable, and that the variables can be summed up to provide an act’s weightiness (W) or expected amount of face threat according to the following formula: Wx = D(S,H) + P(H,S) + Rx (Brown and Levinson 1987: 76). The point of calculating face threat, according to Brown and Levinson, is that it will lead to “a determination of the level of politeness with which, other things being equal, an FTA will be communicated” (1987: 76). They do not, however, attempt to apply this formula in a quantitative analysis of face threat (see section 2.2.2, for studies by other scholars that did attempt this). Brown and Levinson proposed five superstrategies (general orientations to face) that are systematically related to the degree of face threat. A rational actor – a “Model Person” (Brown and Levinson 1987) – will select an appropriate superstrategy to counterbalance the expected face threat. In general, an actor would not select a strategy associated with a greater risk of face threat for an act of less risk (as a sort of insurance policy), since the FTA might be assumed to be greater than it is. This would be counter-productive because it is the speaker’s “intention to minimize rather than overestimate the threat to the hearer’s face” (1987: 74). The individual superstrategies are briefly outlined below (the first superstrategy is associated with lowest face threat, and the last with the most). 1) Bald on record: The FTA is performed “in the most direct, clear, unambiguous and concise way possible” (Brown and Levinson 1987: 69); in other words, in accordance with Grice’s Maxims (1975). No attempt is made to acknowledge the hearer’s face wants. This strategy is typically used in emergency situations (e.g. shouting “Get out” when a house is on fire), when the face threat is very small (e.g. “Come in” said in response to a knock at the door), and when the speaker has great power over the hearer (e.g. “Stop complaining” said by a parent to a child). 2) Positive politeness: The use of strategies designed to redress the addressee’s positive face wants. The speaker indicates that in general they want to maintain some of the hearer’s positive face wants, by, for example, treating the hearer as a member of the same group or by expressing liking for the hearer’s personality. The sphere of relevant redress is not restricted to the imposition incurred in the FTA itself. The idea is that the general
402
Jonathan Culpeper
appreciation of the hearer’s wants will serve to counterbalance the specific imposition. It is a sugaring of the pill technique. The expression of positive politeness as a motivated strategy of face threat redress is marked by exaggeration. A general spin-off of positive politeness techniques is that they act as “a kind of social accelerator” (1987: 103), since in using them one indicates a wish to be closer to the addressee. 3) Negative politeness: The use of strategies designed to redress the addressee’s negative face wants. The speaker indicates respect for the hearer’s face wants and the wish not to interfere with the hearer’s freedom of action. Negative politeness is avoidance-based and characterised by: self-effacement, formality and restraint, with attention to very restricted aspects of H’s self-image, centring on his want to be unimpeded. Face-threatening acts are redressed with apologies for interfering or transgressing, with linguistic and non-linguistic deference, with hedges on the illocutionary force of the act, with impersonalizing mechanisms (such as passives) that distance S and H from the act, and with other softening mechanisms that give the addressee an “out”, a face-saving line of escape permitting him to feel that his response is not coerced. (1987: 70) In contrast with positive politeness, negative politeness focuses on the redress of the particular face threat caused by an act. It is a softening of the blow technique. A further contrast is that a spin-off of negative politeness techniques is that it increases social distance, it acts as a “social brake”. 4) Off-record: The FTA is performed in such a way that “there is more than one unambiguously attributable intention so that the actor cannot be held to have committed himself to one particular intent” (Brown and Levinson 1987: 69). In other words, it is performed by means of an implicature (Grice 1975). The speaker’s face threatening intention can be worked out by means of an inference triggered by the flouting of a maxim. Such implicatures may be denied. For example, “I’m thirsty”, said with the goal of getting a cup of tea, flouts the Maxim of Relation (Grice 1975). In a suitable context the hearer may be able to infer that the speaker is asking for a cup of tea, but, if challenged, the speaker could always deny this. 5) Withhold the FTA: The speaker actively refrains from performing the FTA. As Craig et al. (1986: 442) point out, “an option every communicator has is not to talk”.
Both positive and negative politeness strategies require “redressive action”, that is to say, action that is taken in order to “give face” to the hearer in an attempt to counterbalance the expected face damage of the FTA. Such redressive action need not be verbal: giving a box of chocolates or holding a submissive posture could be positive and negative politeness strategies respectively. But much of Brown and Levinson’s work is devoted to the linguistic realisations for positive and negative politeness. Each output strategy is a means of satisfying the strategic ends of a superstrategy. The output strategies given by Brown and Levinson (1987: 101–211) are listed below, each with an example I have constructed.
Politeness and impoliteness
403
Positive politeness output strategies: x x
x
x
x
x
x
x
x
x x x x x
x
Notice, attend to Hearer (H) – “You’ve had your hair cut.” Exaggerate (interest, approval, sympathy with H) – “That was so awful, my heart bled for you.” Intensify interest to H (exaggerate facts, tell stories in present tense) – “I open the door, guess what I see? A sea of mess.” Use in-group identity markers (in-group address forms, dialect, jargon, slang, ellipsis) – “Liz, my friend, come here.” Seek agreement (select a safe topic on which agreement is expected) – “Nice weather today.” Avoid disagreement (token agreement, white lies, hedging opinions) – “Yes, it’s kind of nice.” Presuppose/raise/assert common ground (small talk, shift deictic centre from S to H, presuppose H’s knowledge, wants, attitudes) – “I know how desperate you are to go.” Joke (utilise shared knowledge, put H “at ease”) – “So you’re free to do me a favour tomorrow.” (said to an obviously busy individual) Assert or presuppose S’s knowledge of and concern for H’s wants – “I know you want to help, I’ve got a job you can do.” Offer, promise – “Come over for dinner tomorrow.” Be optimistic – “You don’t mind if I borrow your computer?” Include both S and H in the activity – “Let’s have a drink.” Give (or ask for) reasons – “The sun is shining. Shall we do the garden?” Assume or assert reciprocity – “It’s your turn to make the tea. It would do us both good.” Give gifts to H (goods, sympathy, understanding, cooperation) – “I’ve recommended you to the boss.”
Negative politeness output strategies: x x x
x x x x x x x
Be conventionally indirect – “Can you please pass the salt.” Question, hedge – “Actually, I wondered, if you could help?” Be pessimistic (use the subjunctive, negative, and remote-possibility markers) – “I don’t suppose there would be any chance of a cup of tea?” Minimise the imposition – “Could I borrow your pen for a second?” Give deference – “I’ve been a real fool, could you help me out?” Apologise – “I don’t want to trouble you, but …” Impersonalise S and H – “It would be much appreciated, if this were done.” State the FTA as a general rule – “Late comers will not be served.” Nominalise – “Your failure to appear did not make a favourable impression.” Go on-record as incurring a debt, or as not indebting H – “I’d be forever grateful, if you’d help.”
404 2.2.2.
Jonathan Culpeper
Application and critical assessment
Leech (1983b and 1992) demonstrated how his 1983a model might be applied, in the one case to literary prose and in the other to a play, but applications of the entire model have been virtually non-existent (though see Brookins 2010). Instead, specific aspects of his model, such as his treatment of irony or banter, have fed into the development of new models, and, more generally, his work has had an impact on theoretical discussions. As far as specific criticisms, the most frequent are twofold. One is that he offers little guidance on the application of the maxims, with the exception of the Tact maxim. Do the sociological variables (e.g. power, social distance) apply equally to the other maxims? The other is that the list of maxims is open-ended. What is to stop the proliferation of maxims? Regarding the latter criticism, this is one that Brown and Levinson (1987: 4–5) themselves make of Leech, arguing that Leech’s model is too unconstrained to permit counterexamples and claiming their own model to be more parsimonious. Leech’s (pers. com.) riposte includes the observation that Brown and Levinson’s model is equally subject to proliferation: their lists of linguistic output strategies are open-ended. Turning to Brown and Levinson (1987), I will review each of the aspects of the model noted in the previous section. Recent discussion has focused on the precise definition of “face” (see in particular Bargiela-Chiappini 2003; Arundale 2006), and much of this has been a reaction to Brown and Levinson’s (1987) idea that face can be described in terms of universal individualistic psychological “wants”. Brown and Levinson (1987: 61) claim that their notion of face is “derived from that of Goffman and from the English folk term”. However, with Goffman, it is not just the positive values that you yourself want, but what you can claim about yourself from what others assume about you. How you feel about yourself is dependent on how others feel about you, and so when you lose face you feel bad about how you are seen in other people’s eyes. This social interdependence has been stripped out of Brown and Levinson’s reductive definition. Positive face is about what you as an individual find positive; negative face is about not imposing upon you as an individual. But this seems to ignore cases where the positive attributes apply to a group of people (e.g. a winning team), or where an imposition on yourself is not the main concern, but rather it is how you stand in relation to a group (e.g. whether you are afforded the respect associated with your position in the team). From a cultural perspective, researchers have argued that Brown and Levinson’s emphasis on individualism is a reflection of Anglo-Saxon culture, and not at all a universal feature. Matsumoto (1988), for example, points out that Japanese culture stresses the group more than the individual (see also Ide 1989; Ide and Ueno this volume; Gu 1990; Mao 1994; Nwoye 1992; Wierzbicka 2003 [1991], for similar points): What is of paramount concern to a Japanese is not his/her own territory, but the position in relation to the others in the group and his/her acceptance by those others. Loss of Face is associated with the perception by others that one has not comprehended and acknowl-
Politeness and impoliteness
405
edged the structure and hierarchy of the group. […] A Japanese generally must understand where s/he stands in relation to other members of the group or society, and must acknowledge his/her dependence on the others. Acknowledgement and maintenance of the relative position of others, rather than preservation of an individual’s proper territory, governs all social interaction. (Matsumoto 1988: 405)
Ide (e.g. 1989, 1993) argues that for some cultures politeness is not simply a matter of the individual’s strategic choice in redressing FTAs (something which relates to “volition”), but a matter of working out your position in a group and the social norms and acting accordingly. The latter Ide refers to as “wakimae”, a notion that bears strong similarities with Watts’s (1989, 2003) notion of “politic behaviour”, which I will discuss later. As I observed in the previous section, Brown and Levinson’s (1987) work is limited to acts that threaten face, and facework (politeness work) that attempts to redress those threatening acts. What about acts that simply enhance face? An important merit of Leech’s Politeness Principle is that it is not confined to the management of potentially “impolite” acts (i.e. FTAs), such as asking somebody to do something for you, but also involves potentially “polite” acts (Leech 1983a: 83) (or face enhancing acts), such as a compliment out of the blue. Leech’s Politeness Principle allows for the minimisation of impolite beliefs and the maximisation of polite beliefs. This helps account for why, returning to an earlier example, the direct command “Have a drink” at a social occasion, which would appear to be impolite in brusquely restricting the hearer’s freedom of action, in fact maximises a polite belief such as that the hearer would like and would benefit from a drink but is too polite to just take one. And what about acts that simply attack face – threats, insults, put-downs, sarcasm, mimicry and so on? This is no problem for Leech, but lies outside the scope of Brown and Levinson (they did not follow up Goffman’s 1967: 24–6 comments on “aggressive facework”). Clearly, politeness is not the issue here but rather “impoliteness”, an area I will attend to in section 3. Recent “relational” approaches to politeness, to be reviewed in section 2.3.2, locate potentially polite behaviours within a framework that encompasses all kinds of facework. Facework is generally more complex than Brown and Levinson’s (1987) model suggests. There is no neat separation between positive and negative facework, as several researchers have pointed out (see the discussion of mixed output strategies below). An interruption, for example, imposes on the other and thus may seem typical of negative face threat, but the interruption may also imply that the person who interrupted did not care to hear what the target was saying (i.e. they signal that it has little value for them), which would be typical of positive face threat. In my view, whilst an act may primarily have implications for one kind of face, it will often have secondary implications for other kinds. Furthermore, many of the acts that Brown and Levinson list as primarily threatening the hearer’s face can have ramifications for the speaker’s face. Orders, threats, criticisms and contradictions
406
Jonathan Culpeper
may carry potentially unfavourable, and hence self face threatening, attributions (e.g. authoritarianism and hostility), that may merit facework attention. Coupland et al. (1988) provide a good illustration of this. They studied politeness phenomena in nurses’ talk to institutionalised elderly people, and, in particular, attempts to redress the face threat involved in requesting that medicines be taken. Facework apparently directed at the positive face of the patients, such as the in-group identity markers “my darling, Edith, love” (1988: 259), is described by the authors as “mere verbalisations” (1988: 260). Facework apparently directed at the negative face of the patients, such as attempts to minimise the imposition of a drink by referring to it as “little” (1988: 261), is described as “more rather than less likely to impose and threaten face” (1988: 261), since it trivialises the imposition. Coupland et al. conclude that facework was partly designed to maintain the face of the nurses themselves. By producing the trappings of politeness the nurses could redress actions that threaten their face “both as a caring individual and as a competent professional” (1988: 260–261). Brown and Levinson’s (1987) theory of the mutual vulnerability of face is insufficient as an explanation of cooperation in facework. People are actively involved in maintaining and enhancing their own faces, and not merely hoping for reciprocal facework. Also, self-interest motivates cooperative behaviour for a number of reasons, not just mutual vulnerability. For example, cooperative social behaviour can promote an image of friendliness, kindness, helpfulness, etc., and this may well be an important identity claim of the self. Note that Brown and Levinson’s book was published in a sociolinguistics series, exemplifying interactional sociolinguistics. It is no surprise then that the methodological flavour of the dominant sociolinguistics paradigm, that of Labov, with its emphasis on quantification, affected subsequent politeness studies. Numerous researchers began administering questionnaires or Discourse Completion Tasks (favourite though not the only methodologies) to quantify the kind of politeness strategies used by people of different relative power, social distance and so on (see Spencer-Oatey 1996, for many references). Research has strongly supported Brown and Levinson’s claims for the power variable: the more relatively powerful speaker being associated with less politeness (e.g. Falbo and Peplau 1980; Baxter 1984; Holtgraves 1986; Brown and Gilman 1989; Holtgraves and Yang 1990; Lim and Bowers 1991; Leichty and Applegate 1991). Research has also generally supported claims about the rank variable: the greater imposition being associated with more politeness (Cody et al. 1981; Lustig and King 1980; Brown and Gilman 1989; Holtgraves and Yang 1992). However, evidence for claims about the distance variable, that politeness increases with distance, has been inconsistent. Baxter (1984) and Brown and Gilman (1989) found that more politeness was used in closer relationships. However, the basis of these studies is now being questioned. SpencerOatey (1996) demonstrated that researchers varied widely in what is understood by power or social distance. In fact, these variables were sometimes subsuming other
Politeness and impoliteness
407
independent variables. Baxter (1984), for example, showed that affect (i.e. whether there is liking or disliking between participants) was getting muddled up with social distance, but in fact is an independent variable. It is also worth noting here that Wolfson’s (1988) “bulge theory” of social distance flies in the face of Brown and Levinson’s claims. According to bulge theory, our responses to the extremes on the social scale, strangers and intimates, are similar; it is acquaintances, co-workers and casual friends who are different, because their social relationship is less certain and thus requires more negotiation. In politeness terms, a graph with increasing degrees of politeness on the y axis and of social distance on the x axis would have a bell curve and not a straight positive linear relationship, as predicted by Brown and Levinson. Furthermore, Brown and Levinson concede that they “underplay the influence of other factors” (1987: 12) in determining the seriousness of face threat. For example, the rights and obligations, the presence of a third party, formality or mood may be important. Regarding combinations of variables, Brown and Levinson assume that the effects of power, distance and rank on the perceived degree of face threat are independent of each other and additive; that is to say, each variable would have the same effect regardless of the other variables. This understanding is implicit in their formula for the calculation of face threat. However, there is some empirical evidence which suggests that the combination of these variables is not that simple. For example, Holtgraves and Yang (1990) found a relationship between power and distance such that when the speaker and hearer differed in power, distance had little effect. Brown and Levinson admit that they “may have been in error to set up the three super-strategies, positive politeness, negative politeness, and off record, as ranked unidimensionally to achieve mutual exclusivity” (1987: 18), but, they argue that there is an “absence of definitive evidence” that they “got the ranking wrong” (1987: 20), and, as mentioned above, resist the idea that strategies can be mixed (for example, positive politeness markers occurring in negative politeness strategies such as indirect requests) (1987: 17–20). However, a number of researchers have challenged that ranking. Blum-Kulka (1985) suggests that off-record strategies could be less polite than negative politeness strategies, since it is impolite to require a superior to calculate the force of an off-record request. Baxter (1984) suggests that positive politeness may presuppose negative politeness and should therefore occupy a higher position in the hierarchy. Moreover, several researchers (e.g. Scollon and Scollon 1981; Craig, Tracy and Spisak1986; Tracy 1990; Lim and Bowers 1991) argue that, because positive and negative politeness are different in type, they cannot be ranked unidimensionally. For example, Lim and Bowers (1991) argue that strong threats to positive face are not usually counterbalanced by negative facework. They give the following example, “I’m sorry to say this. I know that I don’t have any right to criticise you, but you did a poor job”. The negative politeness strategies here – the apology and assertion of the hearer’s rights to immunity from imposition – do not ameliorate the criticism. Brown and Gilman’s
408
Jonathan Culpeper
(1989) solution was to have a single category of redress in which acts of positive or negative politeness may be mixed, but need not be. Craig, Tracy and Spisak (1986: 452–3) claim that it is possible for a strategy to involve both negative and positive politeness. An example they give from their data is “Do me a favour”, which they argue is a kind of apology in Brown and Levinson’s terms (in so far as there is an admission that the “favour” will impinge on the other) (negative politeness), but also a kind of reciprocal doing of favours (positive politeness). In general it is difficult to assess the evidence, since it is not clear what the underlying politeness dimension linked to the superstrategies actually is. Like the superstrategies, the output strategies for positive and negative politeness are not linked to any clear underlying dimension of politeness. As Shimanoff (1977) notes, sometimes they are defined functionally (e.g. “Be pessimistic”) and sometimes linguistically (e.g. “Question, hedge”). Thus, the status of each strategy and how it relates to the other strategies is not clear. Craig, Tracy and Spisak (1986) pose a number of pertinent questions related to this issue. Can individual output strategies be scaled? Are all negative politeness strategies more redressive than any positive politeness strategy? Which output strategy will one select to serve a particular superstrategy? Furthermore, the output strategies can perform other functions apart from face redress. For example, they can be used, Brown and Levinson admit, as a social “accelerator” or “brake”. Neither are they mutually exclusive. For example, Craig, Tracy and Spisak (1986: 446) point out that “certain verbal hedges (negative politeness strategy 2; ‘I wonder whether’ or ‘I was wondering’) also appear to count as conventional indirectness (negative politeness strategy 1)”. Regarding the relationship between the output strategies and politeness, one issue is how politeness is affected by the quantity of strategies. Brown and Gilman (1989) assess the amount of redress in terms of the number of “codable” output strategies, claiming that incidental remarks made by Brown and Levinson justify this approach (e.g. “In general, the more effort S expends in face maintaining linguistic behaviour, the more S communicates his sincere desire that H’s face wants be satisfied”, 1987: 93). Elsewhere, however, Brown and Levinson (1987: 22) explicitly attack this approach: “politeness is implicated by the semantic structure of the whole utterance, not communicated by ‘markers’ or ‘mitigators’ in a simple signalling fashion which can be quantified”. The point here is that politeness is not just determined by a particular strategy. It is determined by a particular strategy in a particular context and the participants’ particular assessment of the whole. One final point to mention regarding the output strategies, though it is not one usually mentioned in the literature, is that Brown and Levinson pay very little attention to prosodic, paralinguistic or non-verbal resources for conveying politeness. In a suitable context, a smile could be a means of attending to somebody else’s face; or an intonation contour might be crucial in interpreting the “politeness” of a speech act (see, for example, Wichmann 2004 on please-requests). This deficit is in fact a general one across the entire literature. The single
Politeness and impoliteness
409
exception of note is Arndt and Janney (1987), who argue that ‘utterances become “meaningful” – by which we mean interpretable – only through the interaction of verbal, prosodic, and kinesic actions in context’ (1987: 248). They argue that people, as opposed to social situations and their norms of appropriacy, are the locus of politeness, and that we should “focus on cross-modal emotive behaviour as a means by which politeness is negotiated” (1987: 377). 2.3.
Second-wave approaches to politeness
2.3.1.
Rejection
In the previous section, we saw how numerous studies, particularly dating from the 1980s and 1990s, have addressed particular aspects of Brown and Levinson’s (1987) politeness theory. By the 2000s, the tide of criticism had reached the point where the very foundations of Brown and Levinson’s model were being challenged. The landmark publication initiating the second wave is probably Gino Eelen’s A Critique of Politeness Theories (2001), though some foundations were laid in Watts et al. (2005b [1992]). It should be noted here that some of these critical challenges simply reflect the fact that thinking about how communication and social and interpersonal interaction works had moved on since the 1970s, when the seeds of Brown and Levinson (1987) and Leech (1983a) were sown.1 Let us review some of the fundamental criticisms. With respect to Brown and Levinson’s (1987) politeness theory, some are: (1) Ignoring the lay person’s conception of politeness, as revealed through their use of the terms polite and politeness, and instead postulating a facework theory as a theory of politeness; (2) Claiming to be universal (a particular issue with regard to their conception of “face” applied across diverse cultures); (3) Basing the politeness model on an inadequate pragmatic model, which is biased towards the speaker and the production of language and which fails to account for key ways in which politeness is understood; and (4) Failing to articulate an adequate conception of context, despite the key importance of context in judgments of politeness. Further criticisms are given in Eelen (2001: 245–6), one of which is the failure to adequately deal with impoliteness, something that I shall attend to in section 3. The following paragraphs will expand on each of the above four points in turn. We have already seen in section 2.1 how Watts (e.g. 1992, 2003; Watts et al. 2005b) led criticisms of the classic pragmatics-based politeness theories, arguing that they articulated a pseudo-scientific theory of particular social behaviours, labelling them politeness (so-called politeness2), whilst ignoring the lay person’s conception of politeness as revealed through their use of the terms polite and pol-
410
Jonathan Culpeper
iteness to refer to particular social behaviours (so-called politeness1). This criticism strikes at the epistemological and methodological foundations of politeness, as it concerns where our understanding of politeness comes from, the observers or the participants. Similar criticisms have been made of other areas of social science. Broadly speaking, this particular critical thrust reflects a shift in emphasis, at least in some paradigms, from “etic” approaches to an “emic” ones, to use the terminology coined by Kenneth Pike (1967 [1954]). A particular characteristic of politeness1 approaches is that notions of politeness are not taken as given, but are assumed to be subject to dispute, to be negotiated – and indeed emerge – in interaction. In other words, politeness is discursive. I will discuss the discursive approach to politeness in section 2.3.2. It should be noted that Watts and his colleagues are not dismissing politeness2 approaches. They are mainly concerned with politeness2 approaches masquerading as politeness1 approaches, or vacillation between the two. Consequently, they argue for distinct terminology and definitions, as well as careful monitoring (see also, Eelen 2001: 30–32). Of course, being explicit about the basis of one’s approach, notions and claims makes for good social science. This is not to say that it is easy: all politeness2 approaches encompass at least some politeness1 notions – if they did not, a manual on car mechanics would have a more or less equal claim to be a textbook on politeness. Brown and Levinson have received much criticism for proposing that their model is universal. In this respect, one might note that the title of their book is Politeness: Some Universals in Language Usage. However, it is not as if Brown and Levinson (1987) were unaware of cultural differences. In their view, “any comparative social theory must be at once based on universal principles and yet have culture-internal application” (1987: 242). If we throw out universal concepts or more radically any kind of generalization, how can we compare the politeness of one culture with that of another, if each is defined solely within its own terms? It would be the equivalent of comparing apples with oranges and concluding that they are different; whereas applying dimensions of variation (e.g. the absence/ presence of seeds, edibility, sweetness) gives us a handle on the differences. In fact, this is one of the potential weaknesses of the politeness1 approaches: how do we compare politeness in one culture with that of another, if we base our approach solely on members’ constructs? Having said all this, it cannot be denied that Brown and Levinson’s particular “universal principles” turn out not to be so universal. As I noted in the previous section, much criticism has focused on their definition of face, which they claimed to be universal (e.g. 1987: 244). Its basis in the psychological “wants” of individuals and construction as two components, positive and negative, has led to difficulties when applied across various cultures. Recent approaches to politeness have in fact tended to shift back towards Goffman (e.g. Arundale 2006), and/or considerably elaborate and supplement the notion of face, as we will see is the case for Spencer-Oatey’s (e.g. 2000) framework. One final point to mention on the topic of universality is that Leech’s (1983a) politeness
Politeness and impoliteness
411
framework is sometimes lumped together with Brown and Levinson (1987) and criticised for being universalist (see for example, Wierzbicka 2003 [1991]). Whilst Leech (1983a) does concentrate on the elaboration of the Tact maxim, which is perhaps stereotypical of British culture, it is a gross misreading to claim that that work argues for universality. Such claims were part of the motivation for writing Leech (2007), a publication in which he repudiates such claims and also takes the opportunity to update his framework. He spells out his position thus: It is premature to talk of universals of politeness, but my position is as follows. Probably the scales of value in [section] 5.5 above are very widespread in human societies, but their interpretation differs from society to society, just as their encoding differs from language to language. I suggest this is the basis on which a well-founded cross-cultural pragmatic research could proceed. The question to ask is: given these scales of value, what socio-cultural variants of them are found in particular cultures, and what pragmatic linguistic forms of language are used to encode these variants? (2007: 200–201)
Classic politeness theories are built on classic speech act theory (e.g. Austin 1962; Searle 1969) and Gricean implicature (Grice 1975), which, separately or together, do not offer an adequate account of communication, or of politeness in particular. Speech act theory is discussed in relation to single short utterances with single functions, single speakers and single addressees. This ignores the multi-functionality and complexity of discourse situations (see, for example, Thomas 1995: 195–204), and the fact that speech acts are often constructed over a number of turns (see, for example, Geis 1995). Brown and Levinson (1987: 10) recognise that the adoption of speech act theory as a basis for their model has not been ideal: speech act theory forces a sentence-based, speaker-oriented mode of analysis, requiring attribution of speech act categories where our own thesis requires that utterances are often equivocal in force.
The sort of decontextualised speech acts they use do not reflect the indeterminacies of utterances and the face threatening ramifications they may have for any of the participants in a particular speech event. Brown and Levinson’s work includes no extended examples. Gricean implicature underpins the traditional approaches to politeness. Brown and Levinson (e.g. 1987: 95) postulate that deviations from the maxims coupled with inferential work result in the intended message, which can include a politeness motive. In other words, politeness involves recovering a particularised implicature concerning the speaker’s polite intentions (see also Brown 1995). Broadly speaking, Leech (1983a) pursues the same line. Central, then, is the idea that intentions exist a priori in the minds of speakers and that it is the recovery of a speaker’s “polite” intentions by hearers that leads to the understanding of politeness. But the Gricean (e.g. 1957, 1969, 1975) model of communication has been challenged in various respects, including its emphasis on the reflexivity of (communicative) intentions, the relative neglect of conventionality and the fact that it ignores collective or “we-intentions” (Haugh 2008: 101, where many supporting
412
Jonathan Culpeper
references can be found). Even more problematic for the classic theories of politeness is that politeness does not always rely on the recognition of the speaker’s intentions. Politeness can also be conveyed without deviations from the Cooperative Principle. Merely saying “good morning” to my colleague at the beginning of the working day may be considered “polite”, but this does not involve a deviation from the Cooperative Principle triggering the recovery of the speaker’s intention – it is more a case of performing a routine expected by the hearer, given the social norms. A number of researchers have accounted for the fact that politeness can be expected, normal, not noticed and thus not a deviation from the Cooperative Principle (e.g. Escandell-Vidal 1998; Jary 1998; Terkourafi 2001; Watts 2003). This kind of politeness is labelled “anticipated politeness” by Fraser (1999, cited in Terkourafi 2001). The distinction between anticipated politeness and inferred politeness is elaborated by Haugh (2003) (see also Terkourafi 2001: 121–7). In a nutshell, “politeness is anticipated when the behaviour giving rise to politeness is expected, while it is inferred when the behaviour giving rise to politeness is not expected” (Haugh 2003: 400). Gricean implicature also underpins the notion of directness (cf. Searle 1975). The main dimension along which Brown and Levinson’s (1987) super-strategies for performing politeness are ordered is directness, defined in terms of complying with or deviating from Grice’s Cooperative Principle. Thus, at one end we have bald on record and at the other end we have off record (or even avoiding the face threatening act altogether), and in between we have negative and positive politeness. Leech’s (1983a: 108) comments on the relationship between politeness and indirectness are frequently cited: indirect utterances, such as ‘Could you possibly answer the phone’, tend to be more polite because they increase optionality for the hearer, whilst decreasing illocutionary force (cf. Leech 1983a: 108). However, Leech’s account is more nuanced, as he also points out that indirectness might increase impoliteness when expressing “impolite beliefs”. He comments “because ‘You have something to declare’ is an impolite belief, the more indirect kinds of question [e.g. ‘Haven’t you something to declare’] are progressively more impolite, more threatening, than the ordinary yes–no question” (1983a: 171). Nevertheless, Blum-Kulka’s (e.g. 1987), work on requests in various cultures, suggests a different picture. The politest way of making a request is by “appearing to be indirect without burdening the hearer with the actual cost of true directness” (1987: 143–144). This can be achieved by using conventionally indirect strategies, which short-circuit the necessary inferencing to arrive at the illocutionary point of the utterance. But whilst the argument is neat, this does not work for all cultures. Wierzbicka ([1991] 2003: 34) reports that “in Polish, politeness is not linked with an avoidance of imperative, and with the use of [conventionally indirect] interrogative devices, as it is in English”, and that the “flat imperative” “constitutes one of the milder, softer options in issuing directives” (2003:36). The point is that the interpretation of directness varies according to culture; more directness is not al-
Politeness and impoliteness
413
ways interpreted as less politeness (particularly, it seems, in less individualistic cultures) (see also Field 1998 on directives in a native American culture). More fundamentally, the notion of directness, largely shaped by Searle’s (1975) classic work on indirect speech acts, is highly problematic. There is little agreement on the status of direct/indirect speech acts, how indirect speech acts work or where to draw the line between the three usual categories of directness – direct, conventionally direct and non-conventionally direct (see Aijmer 1996: 126–8, for a brief overview of some of the issues). A particular limitation is that key interpersonal information, such as relative power, social distance, and so on, is missing from Searle’s account. This information alone can trigger, for example, the perception that somebody is making a request or order, circumventing any Gricean inferencing. Holtgraves (1994), for instance, found that knowing that a speaker was of high status was enough to prime a directive interpretation, in advance of any remark having been actually made (see also Ervin-Tripp et al. 1987 and Gibbs 1981, for the general importance of social context in speech act interpretation). This leads us to our final consideration in this section, namely, the fact that context is inadequately handled in the classic politeness models, being reduced to a handful of social variables which do not reflect the complexity of real world interactive events (for an excellent illustration of this, see Turner’s 2003 discussion of power). The sheer complexity of context is daunting, encompassing not only aspects of the world relevant to communication, but also their cognitive representation, their emergence in dynamic discourse, different participant perspectives on them and their negotiation in discourse, and so on. Myers (1991: 44–5), for example, notes that the values for all variables are not given, but are constructed in interaction. For each variable “there is not one value, but a tension between at least two interpretations of the situation” (Myers 1991: 44). To be fair to Brown and Levinson, they did acknowledge part of the problem, stating that values on their variables “are not intended as sociologists’ ratings of actual power, distance, etc., but only as actors’ assumptions of such ratings, assumed to be mutually assumed, at least within certain limits” (1987: 74–6) [original emphasis]. But they did not back this up with a suitable methodology (one which is more qualitative in nature and thus able to handle the complexity), and subsequent researchers often chose to ignore it. Research on social context has now moved on, leading to paradigm shifts across the social sciences. With respect to the development of sociolinguistics, we have seen a (at least partial) shift away from the Labovian variationist model. Research on context was given a particular boost by the papers published in Duranti and Goodwin (1992). I noted in section 2.2.2 that power had been found to be particularly important with respect to politeness. The focus on politeness and power has continued in the literature, fuelled by significant developments in the way power is conceptualised (cf. Fairclough 1989; Bourdieu 1991). Power and politeness are the focal points of two recent monographs, Holmes and Stubbe (2003) and Locher (2004), and various articles (e.g. Harris 2003; Mullany 2004; Tiisala 2004).
414 2.3.2.
Jonathan Culpeper
Replacement
Many specific improvements to the classic politeness models have been proposed, as have a number of new models that can claim to be radically different at least in some respects. I will focus on the latter, and discuss them under three headings: (1) discursive, (2) relational, and (3) frame-based. Two points need to be stressed. One is that I use these headings to group work that has distinctive similarities, but I fully acknowledge that there are also some differences. The other is that some works claim a stake in more than one approach; the groups are not mutually exclusive. Discursive A particularly noteworthy line of work is sometimes referred to as “post-modern” or “discursive” (e.g. Eelen 2001; Locher 2004, 2006; Locher and Watts 2005; Mills 2003; Watts 2003, 2005a [1991]; Watts et al. 2005b [1992]2). As a far as the label post-modern signals a concern with cultural and individual relativism and a dislike of universalising generalisations, it is accurate, but it brings some unwanted baggage with it, and so I will deploy the label discursive. Indeed, a key feature of all this work is that it emphasises that the very definitions of politeness itself are subject to discursive struggle. As one might guess from this, discursive work leans towards politeness1, and indeed shares some of its characteristics (see, for example, the discussion of politeness1 in Eelen 2001: 32–43). Although works apparently belonging to the discursive approach vary somewhat, their characteristics typically include: x
x x
x
x x
x
x x
the claim that there is no one meaning of the term “politeness” but it is a site of discursive struggle; the centrality of the perspective of participants; an emphasis on situated and emergent meanings rather than pre-defined meanings; the claim that politeness is evaluative in character (that it is used in judgements of people’s behaviours); an emphasis on context; the claim that politeness is intimately connected with social norms, which offer a grasp on the notion of appropriateness (note here the connection with the socio-cultural view of politeness discussed in section 2.2); the reduction of the role of intention in communication (it is rejected, or at least weakened or re-conceptualised); a focus on the micro, not the macro; and a preference for qualitative methods of analysis as opposed to quantitative.
Politeness and impoliteness
415
Most of these characteristics are also typical of discursive approaches generally. For example, of particular note is the development of the discursive approach in social psychology, spurred on especially by Derek Edwards and Jonathan Potter (e.g. 1992). Discursive psychology similarly emphasises emergent and contingent interactional meanings, is anti-cognitivist, and uses qualitative methods. Creating a “theory” of politeness seems not to be the objective for the discursive politeness approach (cf. Watts 2005b: xlii). A consequence of focussing on the dynamic and situated characteristics of politeness is that politeness is declared not to be a predictive theory (Watts 2003: 25), or, apparently, even a post-hoc descriptive one (Watts 2003: 142). As Terkourafi comments (2005a: 245): “What we are then left with are minute descriptions of individual encounters, but these do not in any way add up to an explanatory theory of the phenomena under study”. Furthermore, if “the analyst cannot legitimately attribute meaning, one wonders what, then, does constitute a legitimate role for the analyst” (Holmes 2005: 115, original emphasis; Haugh 2007b: 297 makes a similar point). However, to be fair to the key practitioners of discursive politeness approach, such as Richard Watts and Miriam Locher, and returning to a point made in the previous section, there is no prohibition on using a conceptual apparatus of the politeness2 kind. For example, Locher (2006: 262; my emphasis) writes that the discursive approach to politeness stresses that we first of all have to establish the kind of relational work the interactants in question employ to arrive at an understanding of the then-current norms of interaction.
Watts (e.g. 2003) and Locher (e.g. Locher 2004; Locher and Watts 2005) embrace the relational approach, which I shall discuss shortly. The concepts that comprise the relational approaches are second-order. For example, Locher (2006: 256) states that the concept of “politic behaviour” is a second-order term, although she claims that it can be “equalled with appropriateness in lay people’s perceptions”. Indeed, note that the relational approach need not be pursued discursively. In point of fact, the relational approach taken by Spencer-Oatey (e.g. 2000), offering a powerful second-order descriptive framework for cross-cultural comparisons, would not match all of the discursive characteristics listed above. Furthermore, Watts (e.g. 2003) and Locher (e.g. Locher 2004; Locher and Watts 2005) both embrace the notion of a (cognitive) “frame”, a second-order concept, which again I will discuss shortly. In practice then, I have argued that Locher and Watts’s work knowingly combines approaches of different orders. What is not so clear is whether the label discursive applies to the whole resulting combination or just the more discursive aspect. In my view, the relational and frame aspects so clearly do not fit discursive characteristics, as listed above, that they could not be considered discursive. In contrast, Haugh (2007b) understands the discursive approach they articulate to embrace relational work, something which then affords him ample opportunity to point out why their approach is incoherent. Whatever the case, as a whole, Locher
416
Jonathan Culpeper
and Watts’s approach is not purely discursive. This may also partly be the reason why some of their apparently discursive politeness analyses do not always seem to pursue the discursive agenda to the full. In the absence of participants deploying and debating explicit evaluations of (im)politeness in the discourse that has taken place, some data for politeness analyses are selected on the basis of claims by the researcher, supported by implicit evidence, that they involve politeness (or rather a weaker claim of “potential politeness”), much in the same way as data analyses of naturally occurring conversation in studies of politeness2. Interestingly, Watts (e.g. 1989, 2003) embraces relevance theory (Sperber and Wilson 1995) as an explanatory framework. Mills (2003) also argues, though not uncritically, that relevance theory can make a contribution to the discursive approach (see also Christie 2007: 278–279). Given that relevance theory is a “grand” theory of universal application, this would seem a rather odd move. It has been used to account for politeness by a number of scholars (e.g. Escandell-Vidal 1996; Jary 1998; Christie 2007). In particular, relevance theory can account for the anticipated versus inferred distinction: so few cognitive effects arise from anticipated politeness (behaviour following social norms) that it is not relevant enough to spend inferential effort; but when there are sufficiently large cognitive effects to reward processing effort, inferred politeness can take place. What makes it attractive to scholars pursuing the discursive approach is that it emphasises the hearer and does not have generalised norms of behaviour as a starting point, instead focusing on specific situated behaviours “it provides an extremely powerful interpretive apparatus” (Watts 2003: 212). However, relevance theory has three problems for politeness-related studies. Firstly, the relevance theory account of communication still involves the recognition of speakers’ intentions. It does not suit politeness to place a relatively restricted notion of intention at its centre. Secondly, Haugh (2003: 46) points out that the notion of cognitive effects has not been sufficiently characterised: [T]here is no distinction made between cognitive effects which have “positive effect” (such as feelings of approval or warmth and so on), and those which have “negative effect” (such as antagonism or alienation and so on). For example, there is no distinction made between showing that one thinks well of others (which can give rise to politeness), and showing that one thinks badly of others (which can give rise to impoliteness).
Thirdly, no publication has shown how relevance theory can produce effective analyses of stretches of naturally-occurring discourse, a limitation Watts (2003: 212) concedes: “[o]ne major problem with RT is that it rarely, if ever, concerns itself with stretches of natural verbal interaction” (2003: 212). As an antidote to the classic politeness theories, discursive politeness work has been a valuable corrective. In particular, it has drawn attention to the fact that (im)politeness is not inherent in particular forms of language in the sense that a judgement of politeness is solely determined by the usage of particular language,
Politeness and impoliteness
417
arguing instead that it is a matter of the participants’ evaluations of particular forms as (im)polite in context. It has also been influential: indeed, many recent studies have at least some of the characteristics of the discursive approach. However, even the leading politeness discursive work has but some characteristics of a purely discursive approach and not all. So, what might a purely discursive approach to politeness look like? The obvious, well-established methodological approach which emphasises the understandings displayed by participants themselves is conversation analysis. Few politeness studies have adopted this approach (but see the notable example of Piirainen-Marsh 2005). Perhaps the reason for this is that conversation analysis has nothing to say about politeness itself: we still need to interpret a participant’s understanding of politeness as relevant to the theory of politeness with which we as analysts are approaching the data. This point is elaborated in Haugh (2007b). Haugh (2007b) argues that the Conjoint Co-Constituting Model of Communication (Arundale e.g. 1999, 2006) has much to offer as a framework for the study of politeness, “as it is consistent with a conceptualization of (im)politeness as being interactionally achieved in a collaborative, non-summative manner through interaction by participants, whilst carefully avoiding the ontological trap of conflating the analysts’ and participants’ perspectives” (2007b: 309). Interestingly, this approach encompasses a wider context than the local structures of conversation analysts, including for example aspects of identity and the history of particular identities. Also, in this approach evidence for politeness evaluations involves the analyst not just in looking at explicit comments made by participants, but also at “the reciprocation of concern evident in the adjacent placement of expressions of concern relevant to the norms invoked in that particular interaction” (2007b:312). The idea of “reciprocation of concern” is an important one, and in fact will be echoed in another framework discussed later in this section. Although the proof of the pudding is in the eating – and we have yet to see the power of this interactional approach in extended analyses of data – it clearly deserves attention. Relational Christie (2005: 4) points out that three of the five articles constituting the very first edition of the Journal of Politeness Research: Language, Behaviour, Culture stress that politeness is some form of “relational work”. Whether one uses the term “relational work” (Locher and Watts 2005), “relational practice” (Holmes and Schnurr 2005) or “rapport management” (Spencer-Oatey 2000, 2005), they have in common a central focus on interpersonal relations, rather than a central focus on the individual performing “politeness” which is then correlated with interpersonal relations as variables. Also, they include impolite behaviour within the scope of relational work, whilst avoiding presenting it as a simple dichotomy with politeness. I will outline Locher and Watts’s (e.g. Locher 2004, 2006; Locher and Watts 2005; Watts 2003) framework, and then that of Spencer-Oatey (e.g. 2000, 2005).
418
Jonathan Culpeper
Locher and Watts state that “relational work can be understood as equivalent to Halliday’s (1978) interpersonal level of communication” (2005: 11), and further that “[r]elational work is defined as the work people invest in negotiating their relationships in interaction” (2008: 78). Relational work is not switched off and on in communication but is always involved. Locher (2004: 51) writes that “[t]he process of defining relationships in interaction is called face-work or relational work”, and states a preference for the term “relational work” because “it highlights the involvement of at least two interactants”. The concept of face is central to relational work, though not as defined by Brown and Levinson (1987) but by Goffman (1967: 5). Face is treated as discursively constructed within situated interactions. Relational work covers “the entire continuum from polite and appropriate to impolite and inappropriate behaviour” (Locher 2004: 51; see also Locher and Watts 2005: 11). In this perspective, Brown and Levinson’s work is not a theory of politeness but “a theory of facework, dealing only with the mitigation of face-threatening acts”, and fails to “account for those situations in which face threat mitigation is not a priority, e.g., aggressive, abusive or rude behaviour” (Locher and Watts 2005: 10). Watts (2005b: xliii see also Locher and Watts 2005: 12; Locher 2004: 90) offers a diagram which usefully attempts to map the total spectrum of relational work, reproduced below as Figure 1.
Figure 1. Relational work (Watts 2005b: xliii)
Politeness and impoliteness
419
Relational work in this perspective incorporates the issue of whether behaviour is marked or not. Markedness here relates to appropriateness: if the behaviour is inappropriate, it will be marked and more likely to be noticed. Unmarked behaviour is what Watts (1989, 2003) refers to in his earlier work as “politic behaviour”: “[l]inguistic behaviour which is perceived to be appropriate to the social constraints of the ongoing interaction, i.e. as non-salient, should be called politic behaviour” (Watts 2003: 19), and is illustrated by the following examples: A: Would you like some more coffee? B: Yes, please. M: Hello, Mr. Smith. How are you? S: Hello David. Fine thanks. How are you? (Watts 2003: 186, emphasis as original) Politeness, on the other hand, is positively marked behaviour. Watts (2003: 19) writes that “[l]inguistic behaviour perceived to go beyond what is expectable, i.e. salient behaviour, should be called polite or impolite depending on whether the behaviour itself tends towards the negative or positive end of the spectrum of politeness”. By way of illustration, we can re-work Watts’s examples accordingly: A: Would you like some more coffee? B: Yes, please, that’s very kind, coffee would be wonderful. M: Hello, Mr. Smith. It’s great to see you. We missed you. How are you? S: Hello David. I’m fine thanks. It’s great to see you too. How are you? As mentioned earlier, politic behaviour echoes Ide’s (e.g. 1989, 1993) notion of “wakimae”, which involves working out your position in a group and the social norms and acting accordingly. A framework that moves beyond a focus on the individual’s strategic choice in redressing FTAs (cf. Brown and Levinson 1987) is in a better position to accommodate cross-cultural variation. Some researchers (e.g. Meier 1995) see politeness as a matter of doing what is appropriate, but Watts is clearly right in allowing for the fact that people frequently do more than what is called for. In Watts’s and Locher’s view, politeness is associated with a marked surplus of relational work. One problem here, however, is that Watts’s definitions, as given in the previous paragraph, suggest a hard line between politic behaviour and politeness: if it is not one, it is the other. This seems unrealistic; surely there is a scale between politic behaviour and politeness capturing degrees of difference between relatively “normal” behaviours and situations, such as greetings and leavetakings in expected contexts, and those which are more creative (see Leech 2007: 203, for a similar comment). Indeed, contrary to the definitions, the dotted lines in Figure 1 suggest fuzziness. The other key dimension in Figure 1 is negative versus positive behaviour, presumably referring to a participant’s evaluation of the behaviour. As noted in section 2.2, these labels are not in themselves sufficiently precise to capture a notion like politeness.
420
Jonathan Culpeper
Spencer-Oatey’s (e.g. 2000, 2002, 2005, 2007, 2008) model of “rapport management” arose from critical dissatisfaction, particularly in the cross-cultural sphere, with Brown and Levinson’s model: Taking these arguments [concerning the inadequacy of Brown and Levinson’s model] into consideration, I propose a modified framework for conceptualizing face and rapport. I maintain that Brown and Levinson’s (1987) conceptualization of positive face has been underspecified, and that the concerns they identify as negative face issues are not necessarily face concerns at all. I propose instead that rapport management (the management of harmony–disharmony among people) entails three main interconnected components: the management of face, the management of sociality rights and obligations, and the management of interactional goals (2008: 13)
Spencer-Oatey’s rapport management framework consists of three types of face, “quality”, “relational” and “social identity”,3 and two types of “sociality rights”. These are summarised in Table 1. Table 1.
Categories in the rapport management framework
Face (defined with reference to Goffman (1967: 5): “the positive social value a person effectively claims for himself [sic] by the line others assume he has taken during a particular contact” (2008: 13)
Quality face (related to the self as an individual): “We have a fundamental desire for people to evaluate us positively in terms of our personal qualities, e.g. our confidence, abilities, appearance etc.” (2002: 540)
Sociality rights defined as the “fundamental social entitlements that a person effectively claims for him/herself in his/her interactions with others” (2008: 13)
Equity rights: “We have a fundamental belief that we are entitled to personal consideration from others, so that we are treated fairly: that we are not unduly imposed upon, that we are not unfairly ordered about and that we are not taken advantage of or exploited” (2008: 16)
Relational face (related to the self in relationship with others): “[s]ometimes there can also be a relational application; for example, being a talented leader and/or a kind-hearted teacher entails a relational component that is intrinsic to the evaluation” (2008: 15) Social identity face (related to the self as a group member): “We have a fundamental desire for people to acknowledge and uphold our social identities or roles” (2002: 540); “[social identity face involves] any group that a person is a member of and is concerned about. This can include small groups like one’s family, and larger groups like one’s ethnic group, religious group or nationality group” (2005:106).
Association rights: “We have a fundamental belief that we are entitled to social involvement with others, in keeping with the type of relationship that we have with them.” (2008: 16)
Politeness and impoliteness
421
Regarding interactional goals, Spencer-Oatey (2008: 17) writes: People often (although not always) have specific goals when they interact with others. These can be relational as well as transactional (i.e. task-focused) in nature. These ‘wants’ can significantly affect their perceptions of rapport because any failure to achieve them can cause frustration and annoyance.
Note that Spencer-Oatey separates goals/wants from face. Spencer-Oatey (e.g. 2008) devotes considerable space elaborating how these three components – face, rights and goals – are linked to pragmatic, linguistic and contextual features. This elaboration goes well beyond simple lexically and grammatically defined output strategies or simple social variables. The important point for the model is that Spencer-Oatey provides a detailed analytical framework which we can apply to language data. Threats to positive rapport or harmony between people can, given the three components of rapport management outlined above, be related to face, rights/obligations or interactional goals. However, rapport management is not confined, as in the case of Brown and Levinson (1987), to counterbalancing threats. SpencerOatey (2008: 32) suggests that there are four orientations: 1. Rapport enhancement orientation: a desire to strengthen or enhance harmonious relations between the interlocutors; 2. Rapport maintenance orientation: a desire to maintain or protect harmonious relations between the interlocutors; 3. Rapport neglect orientation: a lack of concern or interest in the quality of relations between the interlocutors (perhaps because of a focus on self); 4. Rapport challenge orientation: a desire to challenge or impair harmonious relations between the interlocutors.
Rapport enhancement tallies with Leech’s (1983a) accommodation of acts that simply enhance politeness, perhaps to strengthen friendship. Rapport maintenance could be simply a matter of performing politic behaviour, or a matter of restoring relations in the light of threatening behaviour. Rapport neglect accommodates Brown and Levinson’s (1987) observation for the bald on record strategy that in emergency situations politeness is not an issue, along with the many other reasons why somebody may neglect relations (e.g. weighting their own concerns above that of the others). Finally, rapport challenge accommodates impoliteness, to be discussed in section 3. In the last four sentences, I have made these broad connections with some aspects of politeness. Spencer-Oatey is not at all concerned with plotting notions such as “polite” or “impolite” in her scheme; she is simply proposing a second-order framework of interpersonal relations. Frame-based Although, as we have already seen, some researchers have referred to the notion of a frame, it is Marina Terkourafi (e.g. 2001, 2002, 2003, 2005a, 2005b) who goes
422
Jonathan Culpeper
the whole hog and creates the frame-based approach to politeness, anchoring it in pragmatic theory. She argues that we should analyse the concrete linguistic realisations (i.e. formulae) and particular contexts of use which co-constitute “frames”. This avoids problematic notions like directness. Moreover, “[i]t is the regular co-occurrence of particular types of context and particular linguistic expressions as the unchallenged realisations of particular acts that create the perception of politeness” (2005a: 248; see also 2005b: 213; my emphasis). The fact that the expressions are not only associated with a particular context but go unchallenged is an important point. This feature seems to be similar to Haugh’s point that evidence of politeness can be found in, amongst other things, “the reciprocation of concern evident in the adjacent placement of expressions of concern relevant to the norms in both in that particular interaction” (2007b: 312). Note that the fact that we are dealing with regularities means that we can deploy quantitative as well as qualitative methodologies (a simplistic quantitative methodology, such as counting up a particular form, is not possible, as we must count up forms in particular contexts that are unchallenged). There are also some echoes of politic behaviour relating as it does to social routines. Terkourafi suggests that it is through that regularity of co-occurrence that we acquire “a knowledge of which expressions to use in which situations” (2002: 197), that is, “experientially acquired structures of anticipated ‘default’ behaviour” (2002: 197). She also points out that formulae are more easily processed by both speaker and hearer, when juggling face concerns, goals, and so on, and also that using them demonstrates a knowledge of community norms (2002: 196). Thus, “formulaic speech carries the burden of polite discourse” (2002: 197; see also references given in 2005b: 213). The fact that this is so accounts for the observation that politeness often passes unnoticed (Kasper 1990: 193). Of course, it is not the case that such conventionalised formulae – the stuff of anticipated politeness – constitute the only way politeness is conveyed and understood. Terkourafi (e.g. 2001, 2005b) develops Neo-Gricean pragmatics (e.g. Levinson 2000) to account for more implicational/inferential modes. The background to this lies in Grice’s (1975) distinction between particularised and generalised implicatures, and its elaboration by Levinson (e.g. 1995, 2000). Particularised implicatures are worked out from scratch on the basis of the particular context the utterance appears in; generalised implicatures have a more stable association with particular linguistic forms (cf. Grice 1989: 37 (this is a reprint of his 1975 paper)). Levinson’s particular contribution was to characterise generalised implicatures as a level of meaning between particularised implicatures and fully conventionalised (non-defeasible) implicatures. Terkourafi’s further contribution was to split generalised implicatures into two, a division based on the relationship with context. The first captures situations where the implicature is weakly context-dependent, requiring a minimal amount of contextual information relating to the social context of use in which the utterance was routinised and thus conventionalised to some de-
Politeness and impoliteness
423
gree; the second, as described by Levinson, captures situations where the implicature is even more weakly context-dependent – its meaning is presumed in a variety of contexts. The resulting implicational/inferential cline is as follows (Terkourafi 2005b: 211–2): Particularised implicature (utterance-token meaning derived in nonce context)
A
A
Generalised implicature I (utterance-type meaning presumed in minimal context)
Generalised implicature II (utterance-type meaning presumed in all contexts ceteris paribus)
A
Coded meaning (sentence meaning)
Hitherto, standard, classical Gricean accounts of politeness (e.g. Leech 1983a) have made no explicit connection with generalised implicatures, instead discussing politeness in terms of the recovery of the speaker’s intentions in deviating from Gricean cooperativeness on a particular occasion (i.e. in terms of particularised implicatures) (see section 2.3.1). In the introduction to their second edition, Brown and Levinson (1987: 6–7) admit that they may have underplayed the role of generalised conversational implicatures. Terkourafi, in contrast, argues that, whilst politeness can involve full inferencing in a nonce context, what lies at its heart is a generalised implicature of the first type given above. Her argument is neatly summarised here (Terkourafi 2005a: 251, original emphasis): Politeness is achieved on the basis of a generalised implicature when an expression x is uttered in a context with which – based on the addressee’s previous experience of similar contexts – expression x regularly co-occurs. In this case, rather than engaging in fullblown inferencing about the speaker’s intention, the addressee draws on that previous experience (represented holistically as a frame) to derive the proposition that “in offering expression x the speaker is being polite” as a generalised implicature of the speaker’s utterance. On the basis of this generalised implicature, the addressee may then come to hold the further belief that the speaker is polite.
Terkourafi offers a coherent, rich, pragmatic account of politeness. In a number of her publications she investigates politeness in Cypriot Greek. It remains for future research to apply the frame-based approach to other contexts.
3.
Impoliteness
Although the study of impoliteness has a fairly long history (usually in the guise of the study of swearing, an example being Montagu 2001 [1968]) and although early scientific attempts to address the topic (e.g. Lachenicht’s 1980) did not galvanise scholars, momentum has been increasing, with the arrival of, for example, Lakoff (1989), Austin (1990), Beebe (1995), Culpeper 1996, Kienpointner (1997), Tracy and Tracy (1998), Culpeper, Bousfield and Wichmann (2003), Mills (2003) and
424
Jonathan Culpeper
Culpeper (2005). The year 2008 has been an especially important year, as the field saw the arrival of its first monograph, Bousfield (2008), its first volume of papers, Bousfield and Locher (2008), and first journal special issue devoted to impoliteness: “Impoliteness: Eclecticism and Diaspora” (Journal of Politeness Research 4 (2), edited by Bousfield and Culpeper).4 Towards the end of his book, Eelen (2001: 245–6) supplies a summary list of problems with traditional politeness models, problems which are a “good starting point for the research for a more adequate model of politeness” (2001: 245). The first listed is: “The inability to adequately account for impoliteness by the same concepts that explain politeness” (2001:245). This is important, because readers will have noted in the course of this chapter various places where politeness models accommodate or at least having a bearing on impoliteness. Earlier in his book Eelen (2001:98–104) elaborates on the specific problems, which are: x
x
treating impoliteness as failed or absent politeness (e.g. the failure to redress or redress adequately a FTA); treating, explicitly or otherwise, impoliteness as the opposite of politeness, yet only developing concepts for explaining politeness (e.g. Blum-Kulka [1992] 2005 develops cultural scripts to explain politeness but not impoliteness).
Of course, it should be stressed that these problems, especially the first one, do not apply equally to all politeness models (it does not apply, for example, to Leech 1983a). What is generally true is that we do not have a complete apparatus for accommodating or explaining the rich array of communicative action undertaken to achieve and/or perceived by participants as achieving impoliteness (see, for example, the descriptions in Lachenicht 1980; Beebe 1995; Culpeper 1996; Bousfield 2008). At this point we should pause to reflect on whether Eelen is correct. Should a model of politeness account for impoliteness using the same concepts, or is a completely different model required? Recollect that both of the relational frameworks discussed in the previous section accommodate, explicitly or implicitly, both politeness and impoliteness. But the important distinction is that relational frameworks are not models of politeness themselves; they are models of interpersonal relations which may accommodate at least some aspects of politeness, impoliteness, and so on. Still, it cannot be denied that impoliteness phenomena are intimately connected with politeness. One way in which the degree of impoliteness varies is according to the degree of politeness expected: if I told the Vice Chancellor of my University to be quiet, I suspect considerably more offence would be taken than if I told my young daughter to do the same. Moreover, the fact that sarcasm trades off politeness is further evidence of this relationship. For example, “thank you” (with exaggerated prosody) uttered by somebody to whom a great disfavour has been done, reminds hearers of the distance between favours that normally receive polite thanks and the disfavour in this instance. There are, however, also some important
Politeness and impoliteness
425
differences between politeness and impoliteness. Within a relational approach, impoliteness can be associated with a negative evaluation as opposed to a positive evaluation, which can be aligned with politeness. However, as I have pointed out on a number of occasions, the categories positive and negative are very broad. When we look at the specifics, the polite/impolite opposition runs into some difficulties. One possible characteristic of politeness, as noted above, is consideration. In a very broad sense any impoliteness involves being inconsiderate, but defining something in terms of a negative (i.e. what it is not) is not very informative about what it actually is. I conducted a study of 100 impoliteness events reported by British undergraduates. 133 of the total of 200 descriptive labels that informants supplied for those events fell into 6 groups (in order of predominance): patronising, inconsiderate, rude, aggressive, inappropriate and hurtful (see Culpeper 2011). Clearly, being inconsiderate is a descriptive label that strikes a chord with participants. However, patronising, by far the most frequent label, does not have a ready opposite concept in politeness theory. Presumably, it relates to an abuse of power, a lack of deference, but neither power nor deference are well described in the classic politeness theories. Similarly, aggressive, with overtones of violence and power, has only a very general opposite in politeness theory, that of harmony. And there are yet other differences or cases where there do not appear to be easy diametric opposites between impoliteness and politeness. Anger is one of the most frequent emotional reactions associated with impoliteness, particularly when a social norm or right is perceived to have been infringed (Culpeper 2011). But anger lacks a similarly specific emotional opposite associated with politeness. Furthermore, taboo lexical items appear relatively frequently in impolite language. Lexical euphemisms, however, whilst they are associated with politeness, play a minor role.5 One way in which there is an obvious difference between politeness and impoliteness is that impoliteness has its own set of conventionalised impolite formulae. I understand conventionalisation here in the same way as Terkourafi (e.g. 2003), namely, items conventionalised for a particular context of use. As we saw in section 2.3.2, for such items to count as polite they must go unchallenged (e.g. Terkourafi 2005a; see also Haugh 2007b, for a related point). Conversely, then, a characteristic of conventionally impolite formulae is that they are challenged. In my data, by far the most frequent impolite formulae type are insults. These fall into the following four groups (all examples are taken from naturally-occurring data; square brackets give an indication of structural slots, which are optional to varying degrees, and slashes separate attested examples within):
426
Jonathan Culpeper
1. Personalized negative vocatives – [you] [fucking/rotten/dirty/fat/little/etc.][moron/fuck/plonker/dickhead/ berk/pig/shit/bastard/loser/liar/minx/brat/slut/squirt/sod/bugger/ etc.] [you] 2. Personalized negative assertions – [you] [are] [so/such a] [shit/stink/thick/stupid/bitchy/bitch/hypocrite/ disappointment/gay/nuts/nuttier than a fruit cake/hopeless/pathetic/fussy/ terrible/fat/ugly/etc.] – [you] [can’t do] [anything right/basic arithmetic/etc.] – [you] [disgust me/make me] [sick/etc.] 3. Personalized negative references – [your] [stinking/little] [mouth/act/arse/body/corpse/hands/guts/trap/breath/etc.] 4. Personalized third-person negative references (in the hearing of the target) – [the] [daft] [bimbo] – [she] [’s] [nutzo] Other impolite formulae types that emerged are: Pointed criticisms/complaints, Challenging or unpalatable questions and/or presuppositions, Condescensions, Message enforcers, Dismissals, Silencers, Threats, Curses and ill-wishes, and Non-supportive intrusions (further detail and exemplification can be found in Culpeper 2011). Of course, whether or not impoliteness formulae result in impoliteness will depend on the hearer’s assessment of its usage in context. Consider this example: [Lawrence Dallaglio, former England Rugby captain, describing the very close family in which he grew up] As Francesca and John left the house, she came back to give Mum a kiss and they said goodbye in the way they often did. ‘Bye, you bitch,’ Francesca said. ‘Get out of here, go on, you bitch,’ replied Mum. It’s in the blood: My life (2007), from an extract given in The Week, 10/11/07
Here, in the direct speech, we see a conventionalised insulting vocative, “you bitch”, and also a conventionalised dismissal, “get out of here”. McEnery (2006: 39, 236) provides corpus evidence that there is a strong tendency in British English for bitch to be used between women, as here. Nevertheless, these items project contexts that are dramatically at odds with the situation within which they are uttered. Rather than antagonistic relationships, hate, coercion, and so on, we have a strong loving family unit (Francesca has just demonstrated her affection by giving her mother a kiss). The recontextualisation of impoliteness formulae in socially opposite contexts reinforces socially opposite effects, namely, affectionate, intimate bonds amongst individuals and the identity of that group. Here we have the opposite of genuine impoliteness, that is, mock impoliteness, which consists of impolite forms whose effects are (at least theoretically for the most part) cancelled by the context (the term “mock impoliteness” is used in Leech 1983a; also see Bernal 2008, for a discussion of genuine vs. non-genuine impoliteness). Banter is the key
Politeness and impoliteness
427
everyday label, though most types of teasing and some jokes also have in common the fact that they involve mock impoliteness. One of the lacunae in Brown and Levinson (1987) is that they do not treat banter at all; in contrast, Leech (1983a) describes mock impoliteness within his Banter Principle: In order to show solidarity with h, say something which is (i) obviously untrue, and (ii) obviously impolite to h [and this will give rise to an interpretation such that] what s says is impolite to h and is clearly untrue. Therefore what s really means is polite to h and true. (1983a: 144)
Banter, of course, also exists in a heavily ritualised form as a kind of language game – a specific activity type. In America some forms of this activity are known as “sounding”, “playing the dozens” or “signifying”, which takes place particularly amongst black adolescents (Abrahams 1962; Labov’s 1972; for a recent and particularly nuanced account of banter in a community in France, see Tetreault 2009; see also Murphy this volume) Needless to say, impoliteness is frequently achieved and understood without the use of formulae, in other words, through implicit means, as illustrated in the following example (a diary report from a British undergraduate): As I walked over to the table to collect the glasses, Sarah’s said to Joe ‘come on Joe let’s go outside’, implying she didn’t want me there. This was at the pub on Sunday night, and I just let the glasses go and walked away. I didn’t particularly feel bad, but angry at the way she had said that straight away when I got there. We aren’t particularly friends but she was really rude in front of others.
The interpretation of Sarah’s utterance partly rests on assumptions about for whom the message is intended. Clearly, the informant assumes that, whilst the addressee is Joe, the target is her, something which seems to be supported by the fact that it was said “straight away when I got there”. It is possible, of course, that the offender also used non-verbal means to clarify the target, such as looking at her whilst she spoke. Taking the informant as the target, the utterance “come on Joe let’s go outside” seems to have no relevance at all for her: it flouts the Maxim of relation (Grice 1975). The informant draws the implicature that going outside entails moving away from where she is, in other words, she is being excluded. In my data, the most frequent implicit strategy by which impoliteness is understood is sarcasm. Sarcasm can be understood as mock politeness, that is, the politeness is not understood to be genuine (cf. Culpeper 1996, who draws on Leech 1983a). The message conveyed is partially mixed: some aspects, such as the use of politeness formulae, suggest politeness; other aspects, typically contextual or co-textual, suggest impoliteness. For example, a member of staff at Lancaster University writing to complain about somebody backing into her car in the car park and then disappearing concludes her complaint: “Thank you SO VERY MUCH”. Note the capitalisation here. The parallel in spoken language is the prosody. Mixed sarcastic messages often involve multimodality: the verbal content conflicting with the prosody or vis-
428
Jonathan Culpeper
ual aspects. In all such cases, the overall assessment must be weighted towards aspects suggesting impoliteness, leaving the aspects suggesting politeness (typically the formulaically polite words) as a superficial veneer, reminding the target of the distance between the polite context and the impolite one in hand. To some extent, the jury is still out on whether a theory of politeness should encompass impoliteness and also how it should do so. One of the problems in moving this issue forward is that there are so many different politeness models.
4.
Concluding remarks on the definition of politeness
Although the discussion thus far has presented various definitions of politeness which at least capture some aspects of what it is and has conveyed a sense of what politeness involves, it does not quite fully pin down what it is. I propose a definition of politeness that might be said to be an attitudinal view: Politeness involves (a) an attitude comprised of particular positive evaluative beliefs about particular behaviours in particular social contexts, (b) the activation of that attitude by those particular in-context-behaviours, and (c) the actual or potential description of those in-context-behaviours and/or the person who produced them as polite, courteous, considerate, etc. Linguistic politeness refers to linguistic or behavioural material that is used to trigger politeness attitudes. Politeness strategies (plans of action for achieving politeness effects) and formulae (linguistic/behavioural forms for achieving politeness effects) are conventionally associated to some degree with contexts in which politeness attitudes are activated. Impoliteness, although its performance involves significant differences, can be defined along similar but contrary lines: it involves negative attitudes activated by in-context-behaviours which are associated, along with the person who gave rise to them, with impoliteness metalanguage (e.g. impolite, rude, discourteous, etc.).
Like Haugh (2007a: 91), I see politeness as an interpersonal attitude. Attitudes, of course, are well-established in social psychology, and especially language attitude research. An attitude involves a favourable or unfavourable reaction to stimuli, and has cognitive, affective and behavioural elements (see Bradac et al. 2001, and references therein). Note that conceiving of politeness as an attitude accommodates the frequently stated point that politeness is subjective and evaluative (e.g. Eelen 2001; Watts 2003; Spencer-Oatey 2005; Ruhi 2008). However, it should be noted that simply referring to “positive” evaluative beliefs is not sufficiently specific. It is unlikely that politeness involves any positive belief. For example, amusing somebody is an interpersonal activity that is generally viewed positively, but it is not at all clear that it would normally be considered a matter of politeness. A key objective for researchers is to understand the subset of positive evaluative beliefs that count as politeness on a particular occasion. The concept of face (Goffman 1967) is one mechanism for trying to doing this. However, I am not convinced that face easily accommodates all politeness-relevant positive beliefs. People also have
Politeness and impoliteness
429
such beliefs about social organisation and behaviours within social organisations, about how people should be treated, about what is fair and what is not, and some of these beliefs are associated with politeness. Such beliefs – which can include beliefs about facework itself – are part of a society’s moral order. For example, it was very clear from my diary report data that friends and lovers have clear expectations about behaviours in that role (including, for example, a particular amount of attention paid to each other). Spencer-Oatey’s (e.g. 2008) rapport management framework does a good job of accommodating this array of evaluative beliefs, incorporating as it does the notion of sociality rights. Still, even with Spencer-Oatey’s more sophisticated framework there is no guarantee that every positive evaluative belief necessarily correlates with politeness attitudes. The way I have tried to remedy this in my definition is not only to link politeness attitudes to the behaviours that trigger them but also to the actual or potential post-facto metalinguistic description of those behaviours and/or the person who produced them as polite (or a closely related synonym). Linguistic politeness refers to linguistic or behavioural forms that are (conventionally) associated with contexts in which politeness attitudes are activated (this view is consistent with, for example, Terkourafi e.g. 2001, outlined above). We acquire linguistic politeness from our experience of social interactions (e.g. ErvinTripp et al. 1990; Escandell-Vidal 1996; Snow et al. 1990). It involves the use of expressions that are both contextually appropriate and positively evaluated by the target (cf. Locher and Watts 2005). Remember the use of please, as discussed towards the beginning of this chapter. It is not used by anybody to anybody, or in any context, and when it is used, it is generally considered interpersonally positive. The point about politeness routines/markers is that knowledge of both their appropriate context and their positive social meaning has become conventionally associated with the form. Of course, this does not mean that simply using a politeness routine/ marker will result in politeness being achieved. Politeness always involves an overall contextual judgment; as is frequently pointed out, politeness is not solely determined by forms alone (cf. Fraser and Nolan 1981: 96; Watts 2003: 168; Locher and Watts 2008: 78). Thus, “Go to hell please”, said to get rid of somebody, might well be considered interpersonally negative, despite the fact that a conventional politeness marker has been used. In fact, this particular utterance achieves its power, because politeness is part of the conventionalised meaning of please.
Notes 1. The problem of the limitations of classic communicative theory, particularly with respect to how it copes with interaction, is in fact acknowledged by Brown and Levinson (1987: 48), although critics usually ignore this acknowledgement, as Arundale (2006: 194–5) points out.
430
Jonathan Culpeper
2. Whilst it is clear that this volume did much to set the agenda for the new discursive approach to politeness, it should be noted that not every paper within it could be described as discursive. 3. In Spencer-Oatey (2000, 2002) there were two types of face, “quality face” and “social identity face”. In Spencer-Oatey (2007, 2008) “relational face” is added. Three categories of similar definition can also be found in Domenici and Littlejohn (2006: 6, 13). 4. This year also saw the publication of another journal special issue, which, although it is not exclusively devoted to impoliteness, contains a significant number of papers focusing on impoliteness: “(Im)politeness in Spanish-speaking Socio-cultural Contexts” (Pragmatics 18 (4), edited by Diana Bravo). 5. Leech (2009) argued that taboo language is one of two impoliteness areas which politeness theory, specifically his own, cannot adequately account for. The other concerns the negative acts, threats and curses.
References Abrahams, Roger D. 1962 Playing the dozens. Journal of American Folklore 75: 209–220. Aijmer, Karin 1996 Conversational Routines in English: Convention and Creativity. London and New York: Longman. Arndt, Horst and Richard Wayne Janney 1987 Intergrammar: Toward an Integrative Model of Verbal, Prosodic and Kinesic Choices in Speech. Berlin: de Gruyter. Arundale, Robert B. 1999 An alternative model and ideology of communication for an alternative to politeness theory. Pragmatics 9(1): 119–153. 2006 Face as relational and interactional: a communication framework for research on face, facework, and politeness. Journal of Politeness Research: Language, Behaviour, Culture 2(2): 193–217. Austin, John L. 1962 How to Do Things with Words. Oxford: Oxford University Press. Austin, Paddy 1990 Politeness revisited – the dark side. In: Allen Bell and Janet Holmes (eds.), New Zealand ways of speaking English, 276–294. Philadelphia: Multilingual Matters. Bargiela-Chiappini, Francesca 2003 Face and politeness: New (insights) for old (concepts) Journal of Pragmatics 35(10–11): 1453–1469. Baxter, Leslie A. 1984 An investigation of compliance-gaining as politeness. Human Communication Research 10(3): 427–456. Beebe, Leslie, M. 1995 Polite fictions: Instrumental rudeness as pragmatic competence. Linguistics and the Education of Language Teachers: Ethnolinguistic, Psycholinguistics
Politeness and impoliteness
431
and Sociolinguistic Aspects. Georgetown University Round Table on Languages and Linguistics. Georgetown: Georgetown University Press. Bernal, Maria 2008 Do insults always insult? Genuine politeness versus non-genuine politeness in colloquial Spanish. [¿Insultan los insultos? Descortesía auténtica vs. descortesía no auténtica en el español coloquial]. Pragmatics 18(4): 775–802. Biber, Douglas, Stig Johansson, Geoffrey Leech, Susan Conrad and Edward Finegan 1999 Longman Grammar of Spoken and Written English. Harlow: Pearson Education. Blum-Kulka, Shoshana 1985 The Language of requesting in Israeli Society. In: Joseph Forgas (ed.), Language and Social Situations, 113–39. New York: Springer-Verlag. 1987 Indirectness and politeness in requests: Same or different?’ Journal of Pragmatics 11: 131–146. 2005[1992] The metapragmatics of politeness in Israeli society. In: Richard J. Watts, Sachiko Ide and Konrad Ehlich (eds.), Politeness in Language: Studies in its History, Theory and Practice, 255–280. Berlin and New York: Mouton de Gruyter. Bourdieu, Pierre 1991 Language and Symbolic Power. Cambridge: Polity Press. Bousfield, Derek 2008 Impoliteness in Interaction. Philadelphia and Amsterdam: John Benjamins. Bousfield, Derek and Jonathan Culpeper (eds.) 2008 Impoliteness: Eclecticism and Diaspora. Special issue of the Journal of Politeness Research 4(2). Bousfield, Derek and Miriam Locher (eds.) 2008 Impoliteness in Language: Studies on its Interplay with Power in Theory and Practice. Berlin and New York: Mouton de Gruyter. Bradac, James, J. Cargile, Aaron Castelan and Jennifer S. Hallett 2001 Language Attitudes: Retrospect, Conspect, and Prospect. In: W. Peter Robinson and Howard Giles (eds.), The New Handbook of Language and Social Psychology, 137–55. Chichester: John Wiley. Bravo, Diana (ed.) 2008 (Im)politeness in Spanish-speaking socio-cultural contexts. Special issue of Pragmatics 18(4). Brookins, Timothy 2010 A politeness analysis of Catullus’ polymetric poems: Can Leech’s GSP cross the Ancient-Modern divide. Journal of Pragmatics 42: 1283–1295. Brown, Bert R. 1970 Face-saving following experimentally induced embarrassment. Journal of Experimental Social Psychology 6: 255–271. Brown, Penelope 1995 Politeness strategies and the attribution of intentions: The case of Tzeltal irony. In: Esther Goody (ed.), Social intelligence and interaction: Expressions and Implications of the Social Bias in Human Intelligence, 153–174. Cambridge: Cambridge University Press.
432
Jonathan Culpeper
Brown, Roger and Albert Gilman 1989 Politeness theory and Shakespeare’s four major tragedies. Language in Society 18: 159–212. Brown, Penelope and Stephen C. Levinson 1987[1978] Politeness. Some universals in language usage. Cambridge: Cambridge University Press. Christie, Chris 2007 Relevance theory and politeness. Journal of Politeness Research: Language, Behaviour, Culture 3: 269–294. Cody, Michael, Margaret McLaughlin and Michael J. Schneider 1981 The impact of relational consequences and intimacy on the selection of interpersonal persuasion tactics: A reanalysis. Communication Quarterly, 29, 91–106. Coupland, Nikolas, Karen Grainger and Justine Coupland 1988 Politeness in context: Intergenerational issues (Review article). Language in Society, 17: 253–262. Craig, Robert T., Karen Tracy and Frances Spisak 1986 The discourse of requests: Assessment of a politeness approach. Human Communication Research 12(4): 437–468. Culpeper, Jonathan 1996 Towards an anatomy of impoliteness. Journal of Pragmatics 25: 349–67. 2005 Impoliteness and entertainment in the television quiz show: The Weakest Link. Journal of Politeness Research 1: 35–72. 2011 Impoliteness: Using Language to Cause Offence. Cambridge: Cambridge University Press. Culpeper, Jonathan, Derek Bousfield and Anne Wichmann 2003 Impoliteness reviewed: With special reference to dynamic and prosodic aspects. Journal of Pragmatics 35: 1545–1579. Domenici, Kathy and Stephen W. Littlejohn 2006 Facework. London: Sage. Duranti, Alessandro and Charles Goodwin (eds.) 1992 Rethinking Context: Language as an Interactive Phenomenon. Cambridge: Cambridge University Press. Edwards, Derek and Jonathan Potter 1992 Discursive Psychology. London: Sage. Eelen, Gino 2001 A Critique of Politeness Theories. Manchester: St. Jerome. Ervin-Tripp, Susan M., Nancy Bell, Martin Lampert and Amy Strage 1987 Understanding requests. Linguistics 25(1): 107–143. Ervin-Tripp, Susan M., Jiansheng Guo and Martin Lampert 1990 Politeness and persuasion in children’s control acts. Journal of Pragmatics 14: 307–331. Escandell-Vidal, Victoria 1996 Towards a cognitive approach to politeness. In: Katarzyna Jaszczolt and Ken Turner (eds.), Contrastive Semantics and Pragmatics, Vol. 2: Discourse Strategies, 621–650. Oxford: Pergamon. 1998 Intonation and Procedural Encoding: The Case of Spanish Interrogatives. In: Villy Rouchota and Andreas H. Jucker (eds.), Current Issues in Relevance Theory, 169–203. Amsterdam: John Benjamins.
Politeness and impoliteness
433
Fairclough, Norman 1989 Language and Power. 1st edition. London: Longman. Falbo, Toni and Letitia A. Peplau 1980 Power strategies in intimate relationships. Journal of Personality and Social Psychology 38(4): 618–628. Field, Margaret 1998 Politeness and indirection in Navajo directives. Southwest Journal of Linguistics 17(2): 23–34. Fraser, Bruce 1990 Perspectives on politeness. Journal of Pragmatics 14(2): 219–36. 1999 Wither politeness? Plenary lecture delivered at the International Symposium for Linguistic Politeness, Bangkok, 8th of December 1999. Fraser, Bruce and William Nolan 1981 The association of deference with linguistic form. International Journal of the Sociology of Language 27: 93–109. Geis, Michael 1995 Speech Acts and Conversational Action. Cambridge: Cambridge University Press. Gibbs, Raymond W. 1981 Your wish is my command: Convention and context in interpreting indirect requests. Journal of Verbal Learning and Verbal Behavior 20: 431–444. Goffman, Erving 1967 Interaction ritual. Chicago: Aldine Publishing. Grice, H. Paul 1957 Meaning. The Philosophical Review 66: 377–88. 1969 Utterer’s Meaning and Intentions. The Philosophical Review 68: 147–77. 1975 Logic and conversation. In: Peter Cole and Jerry Morgan (eds.), Syntax and Semantics 3: Speech Acts, 41–58. London and New York: Academic Press. 1989 Studies in the Way of Words. Cambridge, Mass: Harvard University Press. Gu, Yueguo 1990 Politeness phenomena in modern Chinese. Journal of Pragmatics 14(2): 237–257. Halliday, Michael A. K. 1978 Language as Social Semiotic: The Social Interpretation of Language and Meaning. London: Edward Arnold. Harris, Sandra 2001 Being politically impolite: Extending politeness theory to adversarial political discourse. Discourse and Society 12(4): 451–472. 2003 Politeness and power: Making and responding to requests in institutional settings. Text – Interdisciplinary Journal for the Study of Discourse 23(1): 27–52. Haugh, Michael 2003 Anticipated versus inferred politeness. Multilingua 22(4): 397–413. 2007a The co-constitution of politeness implicature in conversation. Journal of Pragmatics 39(1): 84–110. 2007b The discursive challenge to politeness research: An international alternative. Journal of Politeness Research 3(7): 295–317. 2008 Intention in pragmatics. Intercultural Pragmatics 5(2): 99–110.
434
Jonathan Culpeper
Ho, David Yau-fai 1976 On the concept of face. American Journal of Sociology 81: 867–84. Holmes, Janet 2005 Politeness and postmodernism – an appropriate approach to the analysis of language and gender? Journal of Sociolinguistics 9(1): 108–117. Holmes, Janet and Maria Stubbe 2003 Power and Politeness in the Workplace: A Sociolinguistic Analysis of Talk at Work. Pearson, Education, Edinburgh. Holmes, Janet and Stephanie Schnurr 2005 Politeness, humor and gender in the workplace: Negotiating norms and identifying contestation. Journal of Politeness Research: Language, Behaviour, Culture 1(1): 121–149. Holtgraves, Thomas 1986 Language structure in social interaction: Perceptions of direct and indirect speech acts and interactants who use them. Journal of Personality and Social Psychology 51(2): 305–314. 1994 Communication in context: The effects of speaker status on the comprehension of indirect requests. Journal of Experimental Psychology: Learning, Memory and Cognition 20: 1205–1218. Holtgraves, Thomas and Joong-Nam Yang 1990 Politeness as universal: Cross-cultural perceptions of request strategies and inferences based on their use. Journal of Personality and Social Psychology 59(4): 719–29. 1992 The interpersonal underpinnings of request strategies: General principles and individual differences. Journal of Personality and social Psychology 62(2): 246–256. Hu, Hsien Chin 1944 The Chinese concepts of “face”. American Anthropologist 46: 45–64. Ide, Sachiko 1989 Formal forms and discernment. Two neglected aspects of universals of linguistic politeness. Multilingua 8(2–3): 223–248. 1993 Preface: The search for integrated universals of linguistic politeness. Multilingua 12(1): 7–11. Jary, Mark 1998 Relevance theory and the communication of politeness. Journal of Pragmatics 30: 1–19. Kasper, Gabriele 1990 Linguistic politeness: current research issues. Journal of Pragmatics 14(2): 193–218. Kienpointner, Manfred 1997 Varieties of rudeness: Types and functions of impolite utterances. Functions of Language 4(2): 251–287. Labov, William 1972 Language in the Inner City: Studies in the Black English Vernacular. Oxford: Blackwell. Lachenicht, Lance G. 1980 Aggravating language: a study of abusive and insulting language. International Journal of Human Communication 13(4): 607–688.
Politeness and impoliteness
435
Lakoff, Robin T. 1973 The logic of politeness; or, minding your p’s and q’s’. Papers from the Ninth Regional Meeting of the Chicago Linguistic Society: 292–305. 1989 The limits of politeness: Therapeutic and courtroom discourse. Multilingua 8(2–3): 101–129. Leech, Geoffrey N. 1977 Language and Tact. Linguistic Agency University of Trier, Series A, Paper No.46. [Reprinted in Leech, G. (1980) Explorations in semantics and pragmatics. Pragmatics and Beyond I(5), 79–117. Amsterdam: John Benjamins.] 1983a Principles of Pragmatics. London: Longman. 1983b Pragmatics, discourse analysis, stylistics and “the celebrated letter”. Prose Studies 6(2): 141–57. 1992 Pragmatic principles in Shaw’s You Never Can Tell. In: Michael Toolan (ed.), Language, Text and Context, 259–278. London: Routledge. 2003 Towards an anatomy of politeness in communication. International Journal of Pragmatics 14: 101–123. 2007 Politeness: Is there an East-West divide? Journal of Politeness Research: Language, Behaviour, Culture 3(2): 167–206. 2009 How far can a theory of politeness be a theory of impoliteness? Illustrated from Edward Albee’s Who’s Afraid of Virginia Woolf. Plenary lecture delivered at the Linguistic Impoliteness and Rudeness Conference: LIAR II, Lancaster, 30 June 2009. Leichty, Greg and James Applegate 1991 Social cognitive and situational influences on the use of face-saving persuasive strategies. Human Communication Research 17: 451–484. Levinson, Stephen C. 1995 Interactional biases in human thinking. In: Esther N. Goody (ed.), Social Intelligence and Interaction: Expressions and Implications of the Social Bias in Human Intelligence, 221–260. Cambridge: Cambridge University Press. 2000 Presumptive Meanings: The Theory of Generalised Conversational Implicature. Cambridge, Massachusetts: M.I.T. Press. Lim, Tae-Seop and John W. Bowers 1991 Face-work: Solidarity, approbation, and tact. Human Communication Research 17: 415–50. Locher, Miriam A. 2004 Power and Politeness in Action: Disagreements in Oral Communication. Berlin and New York: Mouton de Gruyter. 2006 Polite behaviour within relational work. The discursive approach to politeness. Multilingua 25(3): 249–267. Locher, Miriam A. and Derek Bousfield 2008 Introduction: Impoliteness and power in language. In: Derek Bousfield and Miriam A. Locher (eds.), Impoliteness in Language: Studies on its Interplay with Power in Theory and Practice, 1–13. Berlin/New York: Mouton de Gruyter. Locher, Miriam A. and Richard J. Watts 2005 Politeness theory and relational work. Journal of Politeness Research: Language, Behaviour, Culture 1(1): 9–33.
436 2008
Jonathan Culpeper
Relational work and impoliteness: Negotiating norms of linguistic behaviour. In: Derek Bousfield and Miriam A. Locher (eds.), Impoliteness in Language: Studies on its Interplay with Power in Theory and Practice, 77–99. Berlin and New York: Mouton de Gruyter. Lustig, Myron W. and Stephen King 1980 The effect of communication apprehension and situation on communication strategy choices. Human Communication Research 7: 74–82. Mao, LuMing Robert 1994 Beyond politeness theory: ‘Face’ revisited and renewed. Journal of Pragmatics 21(5): 451–486. Matsumoto, Yoshiko 1988 Reexamination of the universality of face: politeness phenomena in Japanese. Journal of Pragmatics 12(4): 403–426. McEnery, Anthony M. 2006 Swearing in English: Bad Language, Purity and Power from 1586 to the Present. London: Routledge. Meier, Ardith J. 1995 Defining politeness: universality in appropriateness. Language Sciences 17(4): 345–356. Mills, Sara 2003 Gender and Politeness. Cambridge: Cambridge University Press. Montagu, Ashley 2001[1967] Anatomy of Swearing. Pennsylvania: University of Pennsylvania Press. Mullany, Louise 2004 Gender, politeness and institutional power roles: Humour as a tactic to gain compliance in workplace business meetings. Multilingua 23(1–2): 13–37. Myers, Greg 1991 Politeness and certainty: The language of collaboration in an AI project, Social Studies of Science, 21(1): 37–73. Nwoye, Onuigbo G. 1992 Linguistic politeness and socio-cultural variations of the notion of face. Journal of Pragmatics 18(4): 309–328. O’Driscoll, Jim 2007 Brown & Levinson’s face: How it can – and can’t – help us to understand interaction across cultures. Intercultural Pragmatics 4(4): 463–492. Penman, Robyn 1990 Facework and politeness: Multiple goals in courtroom discourse. Journal of Language and Social Psychology 9: 15–38. Piirainen-Marsh, Arja 2005 Managing adversarial questions in broadcast interviews. Journal of Politeness Research: Language, Behaviour, Culture 1(2): 193–217. Pike, Kenneth 1967[1954] Etic and emic standpoints for the description of behavior. In: Donald C. Hildum (ed.), Language and thought: An enduring problem in psychology, 32–39. Princeton, N.J.: D Van Norstrand Company. Ruhi, Sükriye ¸ 2008 Intentionality, communicative intentions and the implication of politeness. Intercultural Pragmatics 5(3): 287–314.
Politeness and impoliteness
437
Scollon, Ron and Suzanne W. Scollon 1981 Narrative, Literacy and Face in Inter-ethnic Communication. New Jersey: Ablex. Searle, John R. 1969 Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. 1975 Indirect speech acts. In: Paul Cole and Jerry L. Morgan (eds.), Syntax and Semantics 3, 59–82. New York: Academic Press. Shimanoff, Susan B. 1977 Investigating politeness. In: Elinor Ochs Keenan and Dina Bennett (eds.), Discourse across Time and Space, 213–41. Los Angeles: University of Southern California Press. Snow, Catherine E., Rivka Y. Perlmann, Jean Berko Gleason, and Nahid Hooshyar 1990 Developmental Perspectives on Politeness: Sources of Children’s Knowledge. Journal of Pragmatics 14: 289–305. Spencer-Oatey, Helen 1996 Reconsidering power and social distance. Journal of Pragmatics 26: 1–24. 2000 Rapport management: A framework for analysis. In: Helen D. M. SpencerOatey (ed.), Culturally Speaking: Managing Rapport Through Talk Across Cultures, 11–46. London and New York: Continuum. 2002 Managing rapport in talk: Using rapport sensitive incidents to explore the motivational concerns underlying the management of relations. Journal of Pragmatics 34(5): 529–545. 2005 (Im)Politeness, face and perceptions of rapport: Unpackaging their bases and interrelationships. Journal of Politeness Research: Language, Behaviour, Culture 1(1): 95–119. 2007 Theories of identity and the analysis of face. Journal of Pragmatics 39(4): 639–656. 2008 Culturally Speaking: Managing Rapport through Talk across Cultures. 2nd edition. London and New York: Continuum. Sperber, Dan and Deirdre Wilson 1995 Relevance: Communication and Cognition. 2nd edition. Oxford, U.K. and Cambridge, U.S.A.: Blackwell. Terkourafi, Marina 2001 Politeness in Cypriot Greek: A frame-based approach. Unpublished PhD dissertation. Cambridge: University of Cambridge. 2002 Politeness and formulaicity: Evidence from Cypriot Greek. Journal of Greek Linguistics 3: 179–201. 2003 Generalised and particularised implicatures of politeness. In: Peter Kühnlein, Hannes Rieser and Henk Zeevat (eds.), Perspectives on Dialogue in the New Millennium, 151–166. Amsterdam: John Benjamins. 2005a Beyond the micro-level in politeness research. Journal of Politeness Research: Language, Behaviour, Culture 1(2): 237–262. 2005b Pragmatic correlates of frequency of use: The case for a notion of “minimal context”. In: Sophia Marmaridou, Kiki Nikiforidou and Eleni Antonopoulou (eds.), Reviewing Linguistic Thought: Converging Trends for the 21st Century, 209–233. Berlin: Mouton de Gruyter.
438
Jonathan Culpeper
Tetreault, Chantal 2009 Reflecting respect: Transcultural communicative practices of Muslim French youth. Pragmatics 19(1): 65–84. Thomas, Jenny 1995 Meaning in Interaction: An Introduction to Pragmatics. London and New York: Longman. Tiisala, Seija 2004 Power and politeness: Languages and salutation formulas in correspondence between Sweden and the German Hanse. Journal of Historical Pragmatics 5(2): 193–206. Tracy, Karen 1990 The many faces of facework. In: Howard Giles and W. Peter Robinson (eds.), Handbook of Language and Social Psychology, 209–226. Oxford: John Wiley and Sons. Tracy, Karen and Sarah J. Tracy 1998 Rudeness at 911: Reconceptualizing Face and Face Attack. Human Communication Research 25(2): 225–251. Turner, Ken 2003 Wx = D(S, H) + P(H, S) + Rx: (Notes towards an investigation). Revue de Sémantique et Pragmatique 13: 47–67. Watts, Richard J. 1989 Relevance and relational work: linguistic politeness as politic behaviour. Multilingua 8: 131–166. 2003 Politeness. Cambridge: Cambridge University Press. 2005a[1992] Linguistic politeness and politic verbal behaviour: Reconsidering claims for universality. In: Richard J. Watts, Sachiko Ide and Konrad Ehlich (eds.), Politeness in Language: Studies in its History, Theory and Practice, 43–69. Berlin and New York: Mouton de Gruyter. 2005b Linguistic politeness research: Quo vadis? In: Richard J. Watts, Sachiko Ide, and Konrad Ehlich (eds.), Politeness in Language: Studies in its History, Theory and Practice, xi-xlvii. 2nd edition. Berlin and New York: Mouton de Gruyter. Watts, Richard J., Sachiko Ide and Konrad Ehlich (eds.) 2005a[1992] Introduction. In: Richard J. Watts, Sachiko Ide and Konrad Ehlich (eds.), Politeness in Language: Studies in its History, Theory and Practice, 1–17. Berlin and New York: Mouton de Gruyter. Watts, Richard J., Sachiko Ide and Konrad Ehlich (eds.) 2005b[1992] Politeness in Language: Studies in its History, Theory and Practice. Berlin and New York: Mouton de Gruyter. Wichmann, Anne 2004 The intonation of please-requests: a corpus-based study. Journal of Pragmatics 36(9): 1521–1549. Wierzbicka, Anna 2003[1991] Cross-Cultural Pragmatics: The Semantics of Human Interaction. 2nd edition. Berlin and London: Mouton de Gruyter. Wolfson, Nessa 1988 The bulge: A theory of speech behaviour and social distance. In: Jonathan Fine (ed.), Second Language Discourse: A Textbook of Current Research, 21–38. Norwood, N.J.: Ablex.
14.
Honorifics and address terms Sachiko Ide and Kishiko Ueno
1.
Introduction
1.1.
The hypotheses on the origins of meaning of language
Honorifics and address terms are concerned with linguistic forms and their pragmatics associated with the speaker’s attitude toward the participants of conversation and the nature of the setting. They are, therefore, not elements of language that convey propositional content. Hurford (2007: 173) presented two hypotheses as to the origins of meaning of language: (1) the Communicative Act Foundation hypothesis, and (2) the Independent Assumption hypothesis. In the former, utterances contain nothing descriptive nor logically compelling, but are primitive other-directed acts such as “Aha!”, which simply expresses surprise directed at an addressee. In the latter, utterances have descriptive content independent of any illocutionary expressions with intended effects on the addressees, such as “Today is Saturday”. The former hypothesis aptly applies to the Japanese language, while the latter is clear in the English language, the language which provided the basis for explorations carried out in the framework of Chomsky’s generative syntax. In light of Hurford’s Communicative Act Foundation hypothesis, it is conceivable that Japanese utterances used to be signals for communication, upon which grew complex utterance types. This hypothesis seems to be fruitful for the understanding of the importance of communicative signals over and above the propositional content in Japanese utterances. Among various communicative signals for communicative acts are honorifics, address terms, sentence final particles, and various other modal expressions. Even though a good deal of literature both in English and Japanese exists describing and discussing honorific forms based on objective perspectives, what is crucially missing is a description and explanation of what it means for native speakers as well as for their communities to speak with such linguistic devices as honorifics and other modal expressions. It is the purpose of this chapter, therefore, to discuss the logic of the pragmatics of honorifics, so that readers who are speakers of non-honorific languages will have a grasp of the native speakers’ sense of pragmatics that obligatorily requires the use of modal expressions, among which honorifics play a crucial role.
440 1.2.
Sachiko Ide and Kishiko Ueno
The inside perspective
To illustrate what it means to observe speech act from an inside perspective, the following figures are presented. Figures 1 and 2 illustrate how Japanese speakers must situate themselves in the context while speaking, whereas English speakers take an objective perspective on a speech event. The illustrations below should serve to show how modal agreement to the contextual construal is essential in Japanese pragmatics. The utterance in the following scene, taken from Kawabata’s novel The Izu Dancer and Seidenstecker’s translation of it into English should make the difference in speakers’ perspectives clear.
Figure 1. A scene from The Izu Dancer from the Japanese speaker’s perspective
Figure 2. A scene from The Izu Dancer from the English speaker’s perspective
Honorifics and address terms
441
The pictures show a man wearing a hat that signifies that the wearer is an Ichiko ‘high school student’ (University of Tokyo in the present system). This is the scene of the first encounter of the protagonist, the Izu dancer, and the novelist Kawabata, in what is a love story about the Izu dancer and the novelist. A party of entertainers passes by in the middle of the Izu Plateau and meets Kawabata. Recognizing the hat, the older sister of the protagonist whispers in her sister’s ear, “He is a high school boy (student)”. This is what the translator, Seidenstecker, put in the English sentence. This clear propositional sentence makes sense in English, but it would be odd if it were literally translated back into Japanese. Japanese pragmatics requires that the speaker speak by embedding herself in the context. A close look at the utterance in the balloon of Figure 1 shows how the utterance is indexed to the context. There is no subject, ‘he’, nor copula ‘is’, to relate the subject to the predicate. Instead, something else is stated in the original Japanese utterance. The first is -san, an honorific title, which is suffixed to gakusei (‘a student’, in Seidenstecker’s translation ‘a boy’). The other is yo, a sentence final particle. What is the meaning of the fact that there is no subject and no verb in the Japanese utterance? It would be inappropriate to say that they are deleted, since it is not customary to have either. As is obvious in Figure 1, the speaker does not need to indicate the subject, as both the speaker and the hearer are looking at the referent, the high school student. What is relevant in the context need not be verbalized. This simplicity of not verbalizing the obvious referent is the essence of the aesthetics of Japanese verbal behavior. It could be said that this is in accord with the simplicity that is prevalent in Japanese art forms such as haiku, flower arrangement, the tea ceremony, and the Noh play. As to the absence of the copula ‘to be’, as is shown in the balloon, desu (HON COP) could have been inserted, but if it had been inserted, the speaker would have made the scene more formal than would be appropriate. The informal copula da could have been used instead, but the effect of using this copula would be to make the statement a strong assertion, which would lead to a less friendly atmosphere than its absence. Not having a copula, a predicate verb, is the most appropriate phrasing in this kind of congenial relationship between the speaker and the hearer, who are sisters. In the case of English, however, omitting either or both the subject and the copula is not an option. Thus, in Figure 2, the English utterance in the balloon is “He is a high school boy”. It should be noted that the speaker’s perspective when making an utterance is outside the speech event as the speaker is looking at the scene from an objective stance. The investigation of the pragmatics of honorifics must begin by questioning why there is this difference in the speakers’ perspectives on speech events. What has been indicated so far is that it is essential for Japanese pragmatics to index the context and to show agreement with the context, while it is essential in speaking
442
Sachiko Ide and Kishiko Ueno
English to have a subject, a verb, and subject-verb agreement. In Japanese speech, modal elements such as an honorific morpheme and a final particle constitute pragmatic well-formedness. What is required of Japanese pragmatics, therefore, is to show the speaker’s discernment and sentiment about the contextual elements of the speech event. This is realized only if the speaker takes an inside perspective, as illustrated in Figure 1, not the outside perspective as illustrated in Figure 2. It is essential to observe and discuss the pragmatics of an honorific language from the inside perspective of the native speaker’s discourse, an approach seldom taken to date. 1.3.
The outline of this chapter
Using this understanding of the inside perspective as a point of departure, this chapter will describe and discuss the pragmatics of honorifics and address terms. Most of the literature on honorifics and address terms to date stems from constructs based on an objective perspective, as illustrated in Fig. 2. What is to follow in this chapter is a discussion on how the speakers of honorific languages speak in their daily lives, and why they speak the way they do. It is based on observation and argument grounded on the logic of ba, a semantic space that makes it possible to account for the practice of honorifics from the inside perspective of the context of a situation. The next section reviews the literature on honorifics and address terms, with special attention to some works from the considerable amount of literature written in Japanese, and therefore virtually unknown to the outside world. The third section presents an overview of honorific forms and address terms. Then honorifics are discussed in terms of linguistic politeness. The goal is to clarify how and why the pragmatics of honorifics is intrinsically distinct from Brown and Levinson’s (1978, 1987) framework of linguistic politeness. Finally, a discussion of the Japanese Language Planning Commission will be introduced to provide evidence for the considerable concern in contemporary Japanese society regarding the use of honorifics.
2.
Review of the literature on honorifics
Phenomena connected to linguistic honorifics have received some attention in the fields of anthropologically oriented studies of languages, sociolinguistics, and pragmatics in the last fifty years or so. In the following, several of the works written in English are discussed, as well as a few noteworthy examples from the extensive literature in Japanese by Japanese indigenous linguists.
Honorifics and address terms
2.1.
443
Literature on honorifics and address terms
Among the influential classics dating back to the early days of pragmatic studies are Geertz’s (1960) anthropological description of Javanese linguistic etiquette, Martin’s (1964) article on the speech levels of Japanese and Korean, Brown and Gilman’s (1960) seminal work on pronouns in Indo-European languages, and Brown and Ford’s (1964) analysis of address terms in American English. Based on the anthropological observation of the life of Javanese people, Geertz (1960) remarked that the entire Javanese etiquette system is symbolized in their language use. According to Geertz, the choice of linguistic forms and speech style is determined by the social status or familiarity of the speakers involved in a conversation. Martin (1964) presented a description of Japanese and Korean speech levels that are expressed by honorifics, and discussed them in terms of the social and interactional structure. Martin regarded the relationship between honorific and nonhonorific speech levels reflected in a sentence ending in Japanese and Korean as analogous to the use of pronouns of tu and vous in modern French. Martin aptly remarked, “we shall probably have speech levels in Japanese and Korean as long as we have plurals in English (1964: 412)”. This statement is convincing as it shows his deep insight into the Japanese and Korean languages. He was aware that the choice of speech levels with or without honorifics is pragmatically obligatory, just as marking plurals is obligatory in English. Brown and Gilman (1960) explained the use of the V form, honorific pronouns (e.g., French vous, German Sie) and the T form, non-honorific pronouns (e.g., French tu, German du) in terms of the two dimensions fundamental to the analysis of social life: power and solidarity. Brown and Ford (1964) demonstrated the system of address forms in American English based on empirical data, and explained that the progress from the nonreciprocal use of TLN (title plus last name) to FN (first name) occurs according to the increase of the sense of equality. Braun’s (1988) volume on address terms presented the fruits of a Kiel research project aimed at gathering information on the systems of address terms in a number of languages. It offered an exhaustive list of publications on address terms as well as the results of a large-scale empirical research project. The investigation of address systems in various languages showed that address is so differentiated and culture-specific that any universal theory of address does not fit the reality. It maintained that even a notion such as reciprocity/non-reciprocity of power and solidarity proposed by Brown and Gilman (1960) and Brown and Ford (1964) is not robust, but it discussed the fact that important factors concerning the choice of address terms are regional origin, age, social status/education/occupation, sex, group membership, political/religious views, and personality.
444
Sachiko Ide and Kishiko Ueno
Irvine (1992, 1995, 2009) and Agha (1994, 2002) are to be noted as landmarks among the works on honorific languages. Irvine observed a wide range of descriptions on honorific languages: Javanese, Nahuatl, ChiBemba, Japanese, Shilluk, and Zulu among others. Based on the structural understanding of the descriptions of these honorifics and their usage, Irvine boldly generalizes and defines honorifics: “Linguistic honorifics are forms of speech that signal understanding of some aspect(s) of the form-meaning relationship (2009:156)”. Agha, acknowledging various approaches from a number of perspectives, saw honorifics primarily functioning as registers. What is lacking in all these works is an explanation of how and why people live using honorifics in their lives in their individual speech communities. 2.2.
Literature on Japanese linguistics
Honorifics have always been a major concern in kokugo-gaku (the scholarship of the national language) that is independent of western scholarship. In fact, there are quite a number of studies of honorifics. For example, the compendium of studies on honorifics (Tahara et al. 1966) lists no less than 800 works. Despite the considerable amount of literature written in Japanese, little is known about it outside Japan. It would not be exaggerated to claim that honorifics are one of the major concerns in the academic community of scholars of Japanese linguistics. Studies on honorifics in kokugo-gaku consist of three areas: (1) theoretical studies of the honorific system, (2) historical studies of honorifics, and (3) sociolinguistic studies of honorifics. First, the theorization of honorific grammar has been the major subject of the discussions in Japanese linguistics. Among the leading authorities are Yamada Yoshio and Tokieda Motoki; each of them laid the groundwork for the grammar of honorifics from different perspectives. Yamada’s (1924) theory is noteworthy as it is the earliest comprehensive framework for the grammar of Japanese honorifics. Yamada divided honorifics into two groups, humble honorifics and respect honorifics, and argued that the rule of honorifics is governed by the subject of a sentence. Humble honorifics are used when the subject is the first person, and respect honorifics are used when the subject is the second or the third person. Yamada’s emphasis on the correspondence between the person and respect/humble honorifics has been developed by many scholars in succeeding generations. Tokieda (1941) presented the structure of Japanese from a holistic perspective and proposed that the Japanese language consists of shi ( ), that is, proposition and ji ( ), that is, modality. The choice of subject and object honorifics pertains to proposition and the choice of addressee honorifics pertains to modality. Tokieda’s idea that the structure of language can be regarded as a process of the speaker’s cognition and perception has laid a foundation for interpreting honorifics as a
Honorifics and address terms
445
marker of the speaker’s attitudes and feelings. Even though his idea has not been taken up by major schools of linguists, some Japanese linguists followed Tokieda’s tradition. Among them is Watanabe (1971), who claimed a sentence consists of a propositional element plus a modal element by which the speakers express their feelings toward the speech event. The classification of honorifics Watanabe proposed consists of three parts: honorifics expressing deference toward the person who is talked about, those expressing deference toward the addressee, and those in which the speaker expresses modesty. Tsujimura (1963), a student of Tokieda, advanced a taxonomy of honorifics which consisted of two groups, “referent honorifics”, which concern propositional elements, and “addressee honorifics”, which concern modality elements. Tsujimura (1991) focused on the functional aspect of honorific usage. He questioned the functions of honorifics and proposed four functions: (1) honorifics used for higher-status persons, (2) honorifics used for persons with little familiarity, (3) honorifics used for public/formal occasions, and, notably, (4) honorifics used for expressing a person’s elegance/grace. Yamaguchi (2004), analyzed historical documents in their contexts and claimed that the speakers’ attitudes toward and assessments of an event are expressed by modality, notably by honorifics. Second, it has been possible to research the historical use of honorifics, thanks to the richness of the available historical documents. Tsujimura (1968, 1992) discussed diachronic observations of honorifics. Kindaichi (1959) explored the origin of honorifics. Ishizaka (1944) discussed honorifics in such earlier documents as Manyoshu, the earliest extant anthology of Japanese verse and Kojiki, the oldest legendary stories. Manyoshu and Kojiki were both compiled in the 8th century. Historical examinations of honorifics show how the forms of honorifics have developed along with social and historical changes. As to the origins of honorifics, the most widely accepted assumption is that honorifics were derived from extolling the various kinds of gods (Kindaichi 1959). This idea is supported by the fact that most honorific terms come from euphemisms or praise (Tsujimura 1968). According to Tsujimura (1992) and Kasuga (1977), honorifics were predominantly used in referring to the gods and the Emperor in the Nara era (710~), the period of the Emperor’s predominance. It is notable that self-honorification, i.e., the gods’ or the Emperor’s use of honorifics in denoting themselves, is observed in the same way as the earliest use of nos by the Roman Emperor in speaking of himself as the recipient of the reverential vos (Brown and Gilman 1960:254). Thus, honorifics in the Nara era can be characterized as “absolute honorifics”, that is, honorifics exclusively used to refer to the absolute beings. In the Heian era (784~), when a feudal aristocracy was established, honorifics were to be used not only toward the Emperor but also toward the nobility. According to Tsujimura (1992), it was in this period that addressee honorifics came into existence. The expression which came to be used as an addressee honorific was haberi, originally a humble honorific verb meaning “serve the gods or the Em-
446
Sachiko Ide and Kishiko Ueno
peror”. In addition, honorifics began to shift from absolute honorifics to relative honorifics. The use came to be determined by the relative relationship among the speaker, the addressee, and the referent. The conception of addressee honorifics and the shift from “absolute honorifics” to “relative honorifics” is attributed to the complexity and the mobility of the society. The Kamakura era (1185~) and Muromachi era (1331~) are the time when warriors ruled, and honorifics were governed by the status difference among warrior families (Toyama 1977, Tsujimura 1992). As to addressee honorifics, sourou, originally an honorific verb meaning “serve the superior”, took the place of haberi. In the Edo era (1603~), society was divided into four classes, the top of which was samurai, followed by farmer, artisan, and tradesman, and the use of honorifics was strictly ruled by social positions. Most of the honorific forms used today, such as the addressee honorifics desu/ masu, or subject honorifics o (go) … ni naru, took on their contemporary forms in this period (Tsujimura 1992). From the Meiji era (1868–1911) to the present, along with the abolition of the four classes, the use of honorifics became less determined by social class differences. In 1945, after World War II, the new constitution was laid down in which democratization was introduced. Thus, the marking of social status became less frequent, and the use of honorifics became sensitive to the relationship between conversational participants and the formality of the setting. Third, sociolinguistic studies of honorifics in Japan began along with the founding of the National Language Research Institute in 1948. Since this institute aims at the investigation of the language practice of ordinary people all over Japan, it conducted a series of large-scale surveys on the use of honorifics. These surveys investigated how the use of honorifics is the reflection of such sociolinguistic variables as region, gender, age, and rank in the workplace. The other notable works are Sanada’s (1993) investigation into the use of honorifics in a rural district, and Ide et al.’s (1986) study on sex difference in the use of honorifics. Using data gathered in 1971, Sanada (1993) presented a case where the use of honorifics is strictly determined by the norms of the community. He made a close observation of the use of honorifics by every member of a hamlet where only six families lived in six households. Sanada shows that the choices of honorifics are determined by the addressee’s family status and age. What is astonishing is that the same honorific form is used to the same addressee regardless of differences in speakers. This shows how the choice of linguistic forms is made passively. This is the clearest evidence for the claim that the use of honorific forms occurs according to wakimae (see section 4). Unlike the Western tradition of speaking, where the speakers volitionally choose linguistic forms according to their intention, Japanese speakers choose from expressions shared in the community according to the status and age of the addressees. The speakers’ mind is geared toward fitting the social norm of the context, and toward asking themselves which form is supposed to be
Honorifics and address terms
447
used instead of which they want to use. This illustrates pragmatics according to wakimae, the sense of place in context (cf. section 4.3). Ide et al. (1986) conducted a large-scale survey on the status of honorifics employed by middle-aged subjects, including some 250 men and some 250 women living in and around Tokyo. The research data were analyzed, and the following results were obtained. (1) Women are observed to use higher levels of honorifics. (2) Women evaluate the same honorific form lower than men, thus, from the speaker’s point of view, it is not perceived that women are speaking with a higher level of politeness. (3) The reason for the women’s use of higher honorifics can be attributed to the fact that it is a role difference, not a gender difference itself that leads to the differences in the use of honorifics in Japanese societies, since both men and women engaging mainly in the work place tend to use linguistic forms with lower level honorifics.
3.
The fundamentals of honorifics and address terms in Japanese
Japanese polite expressions involve two kinds of honorifics, one expressed by means of changing the shape of nominal elements, and the other by predicative elements. The former type consists of the polite expressions in the category of address terms. The latter, on the other hand, is a rather complex system in which the sociological nature of the nominal referent and the interpersonal relation of the nominal referents need to be taken into consideration. In addition, the formality of the setting plays an important role in this latter system. 3.1.
Honorification of nominal elements: Address terms
Nouns undergo morphological indexing when their referents are considered worthy of the speaker’s deference or distance. 3.1.1.
Person referents
The complex forms of person referents in Japanese are roughly grouped into three categories: personal pronouns, names with titles, and professional ranks. They are used both as address terms and as the nominal elements of sentences as subjects or objects. Table 1 presents representative forms of person referents. The forms indexing deferential1 status are marked with either one or two asterisks, two asterisks indicating a higher degree of honorification than one asterisk. The bold type indicates the level of formality.
448
Sachiko Ide and Kishiko Ueno
(1)
Personal pronouns
Table 1.
Personal pronouns
First Person
Men’s Speech
Women’s Speech
Gloss
watakushi**
watakushi*
‘I’
watashi*
watashi
boku
atashi
ore Second Person
anata
anata
‘you’
ano kata*
ano kata*
‘that person’
ano hito
ano hito
‘that person’
kare
kare
‘he’
kanojo
kanojo
‘she’
kimi omae Third Person
In speaking, first person pronouns are likely to be omitted. The use of second person pronouns is normally avoided, as it is obvious from the context who ‘you’ is. When the second person pronoun is addressed from the inferior to the superior, it may even convey rudeness. (2) (a)
(b)
Names with titles (LN: last name, such as Yamada; Names with titles FN: first name, such as Keiko) LN/FN/kinship term-sama** (e.g. Yamada-sama**, Keiko-sama**, otoo-sama** ‘father’) LN/FN/kinship terms-san* (e.g. Yamada-san*, Keiko-san*, otoo-san*‘father’) (LN)-sensei** (e.g. Yamada-sensei**) (FN)-sense* (e.g. Keiko-sense*)
Sensei literally means ‘a teacher’. But it is also used as a title, not only for all kinds of teachers, from kindergarten teachers to university professors, but also for other respected professionals such as doctors, dentists, politicians, and writers. The use of sensei indexes the speaker’s perception of the other as an honorable professional. It can also be used as a personal referent even without last names. Whether -sensei can be used instead of -sama or -san as an honorable title depends on the conventions of the group.
Honorifics and address terms
(3)
Professional ranks (LN) kaichoo (LN) shachoo (LN) daijin (LN) gakuchoo (LN) buchoo
449
‘president (of an organization)’ ‘president (of a company)’ ‘minister’ ‘president (of a university)’ ‘manager’
These professional ranks are either used independently or with last names. They co-occur with sama and san (e.g., shachoo-san) with added deference. The use of these professional ranks instead of personal names indexes the speaker’s perception of the status of the addressee or the referent. Note that the lowest professional ranks are never used for address or for reference. Moreover, sama and san can be attached to some professional names and social roles. For example, isha ‘doctor’ and kyaku ‘customer’ can be addressed or referred to as o-isha-sama and o-kyaku-sama, respectively, being prefixed by the honorific prefix o- and suffixed by -sama. The professional name or social role plus -san is pervasively used, as seen in yaoya-san ‘greengrocery-san’, daiku-san ‘carpenter-san’, ekiin-san ‘station employee-san’, kanja-san ‘patient-san’. These are not necessarily honorifics which express deference toward a person of higher status, but rather beautifying language or expressions conveying affectionate feelings. 3.1.2.
Nouns with honorific prefixes
Nouns referring to objects may take the honorific prefix go- or o-. These prefixes are attached to nouns that refer to objects that are associated with persons towards whom deference should be shown. (4a)
(4b)
3.2.
Sino-Japanese nouns Yamada-sensei no go- ryokoo Yamada-Prof. GEN HON- travel ‘Prof. Yamada’s travel’ Other nouns Yamada-sensei no otayori Yamada- Prof. GEN HON- letter ‘Prof. Yamada’s letter’ Honorification of predicative elements
Honorification of predicative elements can be divided into two types: referent honorifics and addressee honorifics.
450 3.2.1.
Sachiko Ide and Kishiko Ueno
Referent honorifics
3.2.1.1. Subject honorifics Subject honorifics involve the prefix o- or go- and the ending ni naru which is attached to the infinitive form of a verb, as in (5b). Only the prefix is attached when the predicate is an adjective or a nominal adjective, as in (6b). (5a)
(5b)
(6a)
(6b)
(7a)
(7b)
Keiko wa eki made aruku. Keiko TOP station to walk ‘Keiko walks to the station.’ Yamada-sensei wa eki made o-aruki ni naru. Yamada-Prof. TOP station to HON-walk ‘Prof. Yamada walks to the station.’ Keiko wa utsukushii. Keiko TOP beautiful ‘Keiko is beautiful.’ Yamada-sensei wa outsukushii. Yamada-Prof. TOP HON- beautiful ‘Prof. Yamada is beautiful.’ Keiko wa byooki da. Keiko TOP sick COP ‘Keiko is sick.’ Yamada- sensei wa go- byooki da. Yamada- Prof. TOP HON- sick COP ‘Prof. Yamada is sick.’
Besides these o … ni naru forms, there is a productive subject honorific verb ending. The suffix (r)are (homophonous with the passive suffix) may be attached to a verb. (8a)
(8b)
Keiko ga hon wo kai- ta. Keiko NOM book ACC write- PAST ‘Keiko wrote a book.’ Yamada- sensei ga hon wo kak- are- ta. Yamada- Prof. NOM book ACC write- HON- PAST ‘Prof. Yamada wrote a book.’
There are also idiosyncratic suppletive forms of subject honorifics. (9)
iku iru kuru kureru iu
‘go’ ‘exist’ ‘come’ ‘give’ ‘say’
irassharu kudasaru ossharu
Honorifics and address terms
451
3.2.1.2. Object honorifics Object honorifics involve the prefix o- or go- and the ending suru attached to the infinitive form of a verb. (10a) Watashi wa Keiko ni chikoku no wake wo tazune -ta. I TOP Keiko DAT late arrival GEN reason ACC ask -PAST ‘I asked Keiko the reason for her late arrival.’ (10b) Watashi wa Yamada- sensei ni chikoku no wake wo I TOP Yamada- Prof. DAT late arrival GEN reason ACC otazune shi- ta. HON- askPAST ‘I asked Prof. Yamada the reason for his late arrival.’ There are some idiosyncratic suppletive forms. (11)
3.2.2.
iku kiku au
‘go’ ‘hear’ ‘meet’
morau
‘receive’
shiru
‘know’
ukagau omeni-kakaru itadaku choodai-suru zonji-ageru
Addressee honorifics
The addressee honorific is what is called teineigo ‘polite language’. It indexes the deference or distance toward the addressee or the formality of the setting. As illustrated below, it can be applied independently of the referent honorific. (12a) Keiko ga kita. Keiko NOM come- PAST ‘Keiko came.’ (12b) Keiko ga ki mashita. Keiko NOM come ADD HON- PAST ‘Keiko came.’ (12c) Yamada-sensei ga kita. Yamada-Prof. NOM come- PAST ‘Prof. Yamada came.’ (12d) Yamada-sensei ga irasshatta. Yamada-Prof. NOM come-REF HON- PAST ‘Prof. Yamada came.’ (12e) Yamada-sensei ga ki mashita. Yamada-Prof. NOM come ADD HON- PAST ‘Prof. Yamada came.’
452
Sachiko Ide and Kishiko Ueno
(12f) Yamada-sensei ga o-ide ni narimashita. Yamada-Prof. NOM come-REF HON- ADD HON- PAST ‘Prof. Yamada came.’ (12g) Yamada-sensei ga irasshaimashita. Yamada-Prof. NOM come-REF HON- ADD HON- PAST ‘Prof. Yamada came.’ As shown in the examples in (12), addressee honorifics are realized with the masu ending. The addressee honorific form of the copula da is desu, or de-gozaimasu, a super-honorific form. (13a) Keiko wa gakusei da. Keiko TOP student COP ‘Keiko is a student.’ (13b) Keiko wa gakusei desu. Keiko TOP student ADD HON COP ‘Keiko is a student.’ (13c) Keiko wa gakusei de-gozaimasu. Keiko TOP student SUP ADD HON COP ‘Keiko is a student.’ 3.3.
Humble forms
There are forms which neither raise the referent nor express politeness toward the addressee, but humble the speaker. (14a) Watashi I ‘I go.’ (14b) Watashi I ‘I go.’
ga iku. NOM go ga mairu. NOM go-HUM HON
Humble forms are idiosyncratic suppletive forms. (15)
iku kuru omou shiru iru iu suru
‘go’ ‘come’ ‘think’ ‘know’ ‘exist’ ‘say’ ‘do’
mairu zonzuru oru moosu itasu
Honorifics and address terms
453
These expressions occur generally with the addressee honorific ending masu, e.g. watashi ga mairi masu, as an expression of a higher degree of politeness toward the addressee. The humble forms are distinguished from what is called kenjoogo ‘humble language’2, expressing the speaker’s humble attitude not toward a specific referent but toward no one in particular, except in some cases when the addressee is the target. When these humble forms are used in referring to the speaker’s behavior or belongings, the status of the other participants is relatively raised. By lowering one’s own status, the speaker shows his or her modesty. Thus, humble forms have the same function as the honorifics used for politeness.
4.
Honorifics and linguistic politeness
4.1.
Honorifics and neglected aspects of the theories of linguistic politeness
Honorifics are the central element in the linguistic politeness of Japanese language practice. What is the relation between the practice of using honorifics and the practice of linguistic politeness? It seems honorifics have not been given a proper explanation in the frameworks of current linguistic politeness theories. Over the last few decades, the topic of linguistic politeness has been the subject of repeated discussion in the field of pragmatics (Culpeper this volume). The principles of linguistic politeness created from Western perspectives have been challenged by empirical evidence from a non-Western language, Japanese, where honorifics constitute an integral part of the language practice (Ide 1989; Matsumoto 1989). This section deals with the neglected aspects of linguistic politeness from the perspective of Japanese. Before discussing honorifics and linguistic politeness, it would be useful to define what the term linguistic politeness means. Linguistic politeness refers to the language usage associated with smooth communication, realized (1) through speakers’ use of intentional strategies to allow their messages to be received by the addressees without threatening their faces, and (2) through speakers’ choices of expressions and linguistic forms to index their sense of place, that is where speakers place themselves in relation to the addressees in daily practice. The first type is realized by the speakers’ volitional strategies, while the second type is realized by the speakers’ indexing of their sense of place in the context. It is the second type of linguistic politeness that this chapter on honorifics and address terms describes. During the 1970s and the early 1980s, major universal principles of linguistic politeness were proposed, notably by Lakoff (1973, 1975), Brown and Levinson (1978, 1987), and Leech (1983). In discussing the problems of judging the grammaticality of a sentence, which had been the major concern of transformational grammar, Lakoff (1970) argued for
454
Sachiko Ide and Kishiko Ueno
the need to consider the context of a sentence in order to judge its grammaticality. This was the beginning of the field of pragmatics in the United States. The context has to be analyzed, she claimed, in terms of rules people follow in speaking, i.e., rules of pragmatic competence, which consist of (1) the rule of clarity, and (2) the rule of politeness. The rule of politeness, in turn, is further refined into three rules. They are (1) “keep aloof”, (2) “give options”, and (3) “show sympathy” (Lakoff 1973). This seminal work triggered interest in what is now established as “linguistic politeness”. Assuming “face” and “rationality” of individuals as common properties of human beings, Brown and Levinson (1978, 1987) posited the universals for one aspect of language use, i.e., linguistic politeness on the basis of the target languages of their field works: Tzeltal spoken in South Mexico and Tamil spoken in South India, as well as English, their native language. They assumed that making any speech act is a face threatening act to the addressee (and the speaker). They presented a framework of strategies for politeness. This consisted of five major clusters of strategies by means of which most polite, deferential, or tactful verbal expressions in different cultures and languages can be explained. These clusters are (1) “without redressive action, baldly”, (2) “positive politeness”, (3) “negative politeness”, (4) “off record”, and (5) “don’t do the Face Threatening Act”. Leech (1983), in attempting to present the overall principles of pragmatics based on Gricean Maxims, treated the politeness principle as one of the three principles in interpersonal rhetoric. This politeness principle consisted of six maxims: “tact”, “generosity”, “approbation”, “modesty”, “agreement”, and “sympathy”. What is common for these pioneering works on linguistic politeness is that they claim, whether explicitly or not, the universal applicability of their principles of linguistic politeness, assuming the practice of linguistic politeness is performed by using strategies. However, Ide (1989) argued that the universality of the principles is questionable from the perspective of languages with honorifics, in particular Japanese. According to Ide, Brown and Levinson’s (1978, 1987) framework, in which honorifics are subsumed under their negative politeness strategy No. 5, “Give deference”, is counter-intuitive to speakers of the Japanese language. Their framework is counter-intuitive because they neglect two aspects relevant to the use of honorifics: one relates to the linguistic level and the other to the use level. The neglected linguistic aspect is the choice of “formal linguistic forms” among varieties with different degrees of formality. The other is the wakimae use, that is, the use of polite forms, not by volitional choice, but by indexing one’s sense of place in context. The closest equivalent English term for wakimae is ‘discernment’ (Hill et al. 1986: 347–348). 4.2.
Linguistic aspect: Formal forms
The point here concerning the linguistic forms arises from the fact that Brown and Levinson’s (1978, 1987) framework fails to give a proper account of formal lin-
Honorifics and address terms
455
guistic forms such as honorifics, which are among the major means of expressing linguistic politeness in honorific languages and address terms. In Japanese, a request can be expressed politely, even using imperative forms, if honorific verb forms are used. (16)
(17)
(18)
(19)
Hon wo yome. book ACC read ‘Read books.’ (non-polite) Hon wo o-yomi-nasai. book ACC read-REF HON ‘Read books.’ (polite) Hon wo yoma nai ka. book ACC read NEG Q ‘Won’t you read books?’ (polite) Hon wo o-yomi-ni-nari mase n ka. book ACC read-REF HON ADD HON NEG Q ‘Won’t you read books?’ (very polite)
(16) is a simple imperative without honorifics, and thus is not polite, just as it would not be in Western languages. (17) employs the imperative, but a referent honorific is used. Therefore, it is polite. (18) is made polite by the use of specific strategies; it has been made less imposing by the strategy of its transformation into a negative and interrogative form. Thus, this is polite according to Lakoff’s (1973) Rule 2 of politeness, “give options”, and Brown and Levinson’s (1978, 1987) negative politeness strategies. Example (19) is the combination of (17) and (18), and therefore the most polite of the examples. These examples illustrate the fact that there are two types of devices to make an utterance polite: One is the choice of formal forms as in (17), and the other is the use of strategies, as in (18). It is the former device, the choice of formal forms that is neglected in the framework proposed by Brown and Levinson (1978, 1987). Lakoff (1973) proposes a rule of politeness that employs the use of formal forms. In her three rules for politeness, Rule 1 requires the use of formal linguistic forms. It is Lakoff who incorporated this aspect of formal linguistic forms into the rules of linguistic politeness. The use of formal forms is not unique to languages with grammatically developed honorific systems. Well known examples would be the choice of the pronoun V (Vous) in contrast to T (Tu) and the choice of the address terms TLN in contrast to FN to mark politeness. The contrast of formal vs. non-formal forms is observed in the choice between forms such as hello vs. hi, and purchase vs. buy or dine vs. eat. Brown and Levinson (1978, 1987) treated formal forms as expressions of negative politeness strategies. However, they should not be categorized as strategies, since there are some fundamental differences between the choice of formal
456
Sachiko Ide and Kishiko Ueno
forms and the use of strategies. Formal forms are (1) limited in choice, (2) sociopragmatically obligatory, (3) grammatically obligatory, and (4) made in accordance not only with a person who is the addressee, but also with a person who is the referent or the speaker. First, while the use of strategies allows a potentially unlimited number of linguistic expressions, the use of formal forms is a matter of choosing among a limited set of forms, a typical example being the choice between address terms of TLN and FN in English. Choosing a formal form or expression out of a limited number of choices makes an utterance polite for the following reasons. According to Levinson (1983: 129), formal forms should be explained as conventional implicature. When vous is used to a singular addressee, conventionally but non-truthconditionally indicates that the addressee is socially distant from, or socially superior to, the speaker. Furthermore Ide (1982: 382) stated, “When formal forms are used, they create a formal atmosphere where participants are kept away from each other, avoiding imposition. Non-imposition is the essence of polite behavior. Thus, to create a formal atmosphere by the use of formal forms is to be polite.” Second, the choice of formal linguistic forms is obligatory in the context of social conventions. (The # marks that the utterance is socio-pragmatically inappropriate.) (20)
(21)
#“Sensei wa hon wo yon- da. #professor TOP books ACC read- PAST #‘The professor read books.’ Sensei wa hon wo o-yomi-ni-nat- ta. professor TOP books ACC read-REF HON- PAST ‘The professor read books.’
In (21), an honorific form is used in referring to the action of a person of higher status than the speaker, in this case a professor being spoken of by a student. This is because the social rules of Japanese society require one to be polite to a higher status person like a professor. This use of an honorific verb form is the socio-pragmatic equivalent of grammatical agreement and may thus be termed socio-pragmatic agreement. Subject-predicate pragmatic agreement is determined by the social rule of the society in which the language is used. In Japanese society (21) is appropriate, but (20) is not, as it does not express due respect to the referent, the professor. Thus, the subject-predicate agreement of honorifics is socio-pragmatically obligatory. Levinson, in discussing honorifics as the linguistic form in which socially deictic information is encoded, distinguished between two honorifics, i.e., relational and absolute (1983: 90–91). He further stated that the relational variety is the most important. However, it must be remembered that this can only be said with reference to egalitarian societies. In societies where an elaborate honorific system has been developed, it is the absolute variety that is basic. One finds evidence for the absolute variety in a diachronic study (Brown and Gilman 1960) and in the descrip-
Honorifics and address terms
457
tion of honorific systems in stratified societies (Geertz 1960, Koshal 1987). In Japan, too, the absolute variety of honorifics can be found in such findings as Shibata (1988: 6). Shibata showed that in the Syuri area of Okinawa Prefecture the address terms for parents and grandparents and the response forms are determined according to the speakers’ and recipients’ social classes. Third, there are no neutral predicate forms. Levinson states, “In general, in such languages (South East Asian), it is almost impossible to say anything at all which is not sociolinguistically marked as appropriate for certain kinds of addressees only (1983: 90)”. Therefore, the choice of honorific or plain form is linguistically obligatory. The choices of pronouns (V or T) and address terms (TLN or FN) in some Western languages can be explained in the same way as the choice of honorifics. The speaker is bound to make a choice between a formal form V or TLN and a non-formal form T or FN. This is the primary use of pronouns and address terms. It is only in manipulative use of this primary usage that a speaker has the liberty of choosing FN instead of TLN. Matsumoto (1987) discussed the obligatory choice of honorifics or plain forms of copulas in Japanese, illustrating three variants of “Today is Saturday”, a nonimposing propositional statement in which there is no danger that a speech act threatens the face of the addressee. One can be expressed in a plain form (da), the second can be in the addressee honorific (desu), and the third can be in the super polite addressee honorific (de-gozaimasu). Thus, one is supposed to choose at least among the following three: (21a) Kyoo wa doyoobi da. Today TOP Saturday COP ‘Today is Saturday.’ (21b) Kyoo wa doyoobi desu. Today TOP Saturday ADD HON COP (21c) Kyoo wa doyoobi de-gozaimasu. Today TOP Saturday SUP ADD HON COP Matsumoto (1987) stated that even in such cases of non-FTA utterances, the speaker is required to make an obligatory choice among the variants, with or without honorifics, according to the formality of the setting and the relationship among the participants the speakers perceive. Fourth, the choice of formal forms is made in accordance with the referent or the speaker, which makes the use of formal forms distinct from verbal strategies oriented only toward the addressee. It is because of these fundamental differences between verbal strategies that are performed volitionally and formal linguistic forms that linguistic politeness is here categorized into two basic types. The use of formal linguistic forms is controlled by a different behavioral principle than that underlying the verbal strategies treated by Brown and Levinson (1978, 1987), as described below.
458 4.3.
Sachiko Ide and Kishiko Ueno
Pragmatic aspect: wakimae
The use of formal forms is inherently dependent upon the speakers’ observation of the social conventions of the society of which they are members. In a society, one behaves according to social conventions, one set of which may be called the social rules of politeness. Ide (1982: 366–377) stated the social rules of politeness for Japanese as: (1) be polite to a person of a higher social position, (2) be polite to a person with power, (3) be polite to an older person, and (4) be polite in a formal setting determined by the factors of participants, occasions, and topics. Except for (2), which could be relative, since a person can have power depending on the role in which one is involved, for example, a mother in the home has more power than the children, but as a secretary at work she has less than the boss, and (4) which is inherently relative, these social rules are essentially absolute in quality. Honorifics, in which the relative rank of the speaker, the referent, and the addressee are morphologically or lexically encoded, are used so as to comply with such rules of politeness. The practice of polite behavior according to social conventions is known as wakimae. To behave according to wakimae is to show verbally and non-verbally one’s sense of place or role in a given situation according to the common sense idea of the people in their lives. In society, an individual is expected to behave according to the sense of wakimae, which is often perceived on the basis of the status and the role of various levels ascribed to or acquired. To perceive and acknowledge the delicate status and/or role differences of the speaker, the addressee, and the referent are considered to be basic to keeping communication smooth. Thus, to observe wakimae by means of choices of expressions is an integral part of linguistic politeness in Japanese. In contrast to the wakimae aspect, the aspect of politeness that realizes the speaker’s intention by means of a wide range of possible expressions is called the ‘volitional’ aspect of use of linguistic politeness. This aspect of linguistic politeness is what Brown and Levinson (1978, 1987) assumed in their framework. Both aspects aim to achieve smooth communication, but they are different in that the speakers’ focus is placed primarily on their position in the context of speaking in the former and on their own intention in the latter.
5.
Honorifics in practice and the logic of ba (‘field’)
5.1.
Honorific practice viewed from an inside perspective
As was indicated in the title of this section, when the speaker speaks in Japanese, the speaker’s perspective lies inside the context. Thus, when speaking, the speakers embed themselves in the context. What is it like to speak when one is embedded in the context? When one constitutes a part in the context, one’s worldview cannot
Honorifics and address terms
459
be objective, but instead is likely to be intuitive. By intuitive is meant that the speaker observes the reality in the close vicinity and therefore perceives this reality directly. Intuition is the ability to know something by using perception rather than recognition. The context should be more aptly termed as ba, a semantic space, as this term provides a basis for explaining the complex mechanism working in the context. Hanks (2005: 207) explicated the elements of context in communication as follows: At any moment in interaction, multiple dimensions of access (among participants, objects, and settings) are simultaneously available for interactants. The selection and understanding of deictics [linguistically encoded signs to mark the context-SI and KU] on the simultaneous articulation of space, perception, discourse, commonsense and mutual knowledge, anticipation, and the framework of participation in which Sprs [speakers-SI and KU] and Adrs [addressees-SI and KU] orient to one another.
This is a superb analytic explanation of the complex reality of the deictic context and its function as viewed from an objective stance. With this knowledge in mind, how can the speaker possibly make the appropriate deictic selection in an actual setting of conversation? It looks extremely difficult to discern which element in the context is relevant at each moment of speaking. While it looks very complicated, the reality is that people in the real context intuitively make selections among linguistic codes and index them as though this task is executed automatically. How in the world is the speaker able to make such instant selections? It seems to be a question of complexity that must be faced if there is to be an understanding of language use according to the context. To approach this intricate question, it may be useful to review the literature about intuitive language use in context. Hill et al. (1986) and Ide (1989) postulated the wakimae aspect of linguistic politeness as the aspect of linguistic politeness that Brown and Levinson (1978) neglected. While the wakimae use has been postulated as a pragmatic framework, it has been expected that further explanations on how the mechanism of wakimae works would be forthcoming. This problem has awaited clarification for the last twenty years. 5.2.
The wakimae practice of honorifics in conversation
According to the wakimae rule of linguistic politeness, the speaker is supposed to use honorifics to those who are categorized as out-group members, and not use them with in-group members. Whether others are categorized as in-group members or not depends on the difference of status, age, familiarity, intimacy, and also the formality of the setting. However, data from natural discourse show that this rule does not always apply. The following illustrates a case where the speaker and the addressee do not follow this rule. How can this pragmatic violation of the rule be explained?
460
Sachiko Ide and Kishiko Ueno
The following excerpt is from the Mr. O corpus.3 T (a teacher) and S (a student) are involved in the joint task of making a story out of 15 picture cards. (22) 1 T: e-tto, arui te it te, ehtto, desu ne well walk CON go CON well HON COP FP ‘Well, (he) is walking, and well isn’t he?’ 2 S: a, nanka ah something ‘Ah, something. ’ 3 T: un yes ‘Yeah.’ 4 S: chotto omoitsui ta n desu kedo a-little hit-on PAST NOM HON COP well ‘(I) just hit on an idea, well.’ =============================================== 5 T: un yes ‘Yeah.’ 6 S: saisho boo wo mitsuke te at-first a-stick ACC find-out CON ‘At first, (he) found a stick.’ 7 T: un yes ‘Yeah.’ 8 S: nan da kore tsukae nai na tte omo t tara what COP this use-can not FP QT think PAST when ‘When he was thinking, “What is this? I can’t use it”.’ 9 T: un un yes yes ‘Yeah, yeah.’ 10 S: koo gake ni sashikakat te like this a-cliff to come-near CON ‘He came near to a cliff like this.’ 11 T: un yes ‘Yeah.’ 12 S: a, ano boo tsuka-erut te hiramei ta toka ah that stick use-can QT hit-upon PAST or-something ‘Ah, he thought that I can use that stick, or something.’ ===============================================
Honorifics and address terms
461
13 T: a, soo desu ne ah so HON COP FP ‘Ah, I think so too.’ 14 S: soo iu no wa doo desu ka so say NOM TOP how HON COP Q ‘What do you think about this (story)?’ Since this pair, a teacher and a student, displays a status difference and they have no previous acquaintance, the wakimae principle of linguistic politeness would call for the indexing of the out-group relationship by the use of the addressee honorific form of the copula desu. Therefore, T uses it in lines 1 and 13, and S uses it in lines 4 and 14. However, no honorifics were used from lines 5 to 12. What is happening here? It has been the understanding that the selection of expressions is determined by relevant sociolinguistic variables, such as status differences and the degree of familiarity. In this case, both the relationship of the speaker and the addressee and the setting stay the same. There is no element whatsoever that would allow the pair to make a shift from honorific use to non-honorific use in the middle of the conversation. How can these phenomena in the use of honorifics be explained? 5.3.
Shift of honorific use and dual mode thinking as the mechanism of ba
It was not until the logic of the ba, developed by Hiroshi Shimizu, was encountered that a way out of this dead end in the use of honorifics was found. Ba is the semantic space where the speech event takes place. The closest equivalent term for ba in English is ‘field’. Ba is introduced here, since it supplies the working mechanism of what is happening in the context. Hiroshi Shimizu, a biophysicist whose aim was to discover the complex (not complicated) system of life in its living state, uses the model of an improvised drama to explain the logic of the ba. In an improvised drama, there are actors, the audience, the stage, the theater, and a rough scenario. The actors offer an impromptu presentation on the stage, interacting with other actors and the audience, making adjustments to the situation as it develops. The drama’s scenario could change every time an actor acts. The story line of the drama resembles what is happening in real life, where what happens next is not predetermined. At every moment, actors recognize others’ positions in the changing ba. Then, the actors reorient themselves so as to make the new position suitable in the new phase of the ba. This is how an improvised drama is performed. Likewise, everyone lives in their ordinary lives, making adjustments to the changing ba that is complex and not analyzable into elements. The metaphor of the improvised drama is Shimizu’s model to discover the complex system of the state of living in terms of the logic of the ba (Shimizu 1987, 1990, 1996, 2000).
462
Sachiko Ide and Kishiko Ueno
The logic of the ba is introduced to compensate for the limits of the scientific way of thinking by reductionism. The logic of ba is characterized by a dual mode thinking to compensate for the elements lacking in the traditional approach of linear thinking in science. Among the advantages of this idea of the logic of ba, the features that might be relevant to an explanation of the use of honorifics and other modal expressions are the following. The state of living is realized in the ba, where dual functions are at work: (1) the function as a local being in the domain of the self-centered ego where one behaves as an individual, and (2) the function as a whole in the domain of place where one interacts with others in order to make a coherent whole. This nature of the dual function of the individual and the whole in the ba can be illustrated by the example of the skin cells in the body that know how to cure a cut in the skin. This is possible because the cells have not only the function of local being, but also the function of knowing their place in the whole. Just as the cells with the same genes grow to organize the various parts of the body, knowing the role of the local and the whole, the cells around the cut skin heal the cut area. This is not done by some supreme being giving orders to do so, but is self-organized by the nature of living cells that have the dual mode of functioning as individuals and as part of a coherent whole (Shimizu 2000:90). This illustration may appear oversimplified, but it can be inferred that every human being has this dual functioning capacity, just as the individual cells that constitute our body can function in a dual capacity. With this idea of a dual mode function in mind, the question of the shift between the use and non-use of honorifics in the example above can be approached. At the outset, both T and S, knowing the status difference, use addressee honorifics. The question is why they allowed themselves not to use honorifics in the middle of the dialogue (from lines 5 to 12). There was no indication of a change in the contextual element. Both speakers simultaneously dropped the use of honorifics in the middle of the dialogue and resumed their use later in lines 13 and 14. An interpretation is possible if the logic of ba with the dual mode function is applied. In the beginning, the speakers recognize each other in the domain of the self-centered ego. As they are out-group members, they index their sense of place with the use of the addressee honorific form desu. However, the conversational partners simultaneously recognize the other domain, the domain of place. It was in line 4 when S (the student) said ‘(I) hit on an idea …’ that the consciousness of the domain of place took precedence over the domain of self-centered ego and they began speaking without honorifics. Speaking without honorifics between the teacher and the student is against wakimae. It is at this point that creative indexing by not using honorifics was performed. This performance of a shift of style to nonuse of honorifics performed in the domain of place results in creating the shared ba between the speakers. This makes the scene dramatically friendly, and thus the joint task of making a coherent story was successfully performed. When the dra-
Honorifics and address terms
463
matic phase was over, however, the conversation partners revert to the perception of the domain of the self-centered ego, and return to the presuppositional indexing of the teacher-student relationship. They resume speaking with honorifics. It is in this way that dual mode functioning makes it possible to account for the shift in the use of honorifics within the same sociolinguistic setting. The point here is that human beings are equipped with the function of dual mode thinking, and the shift from one function to the other is performed not by conscious analytical recognition but rather by the subconscious or by intuition, and thus it appears to happen automatically. The concept of dual mode thinking is able to shed light on various pragmatic phenomena that have been considered contradictory. First, the use of honorifics indexes the relative place between the speaker and the hearer, while it indexes the speaker’s own place situated in the society. Why is it that the speaker can make two indexings in one utterance? Second, the use of honorifics can co-occur with the use of sentence final particles. The former indexes the distance between the speaker and the addressee, while the latter shortens the distance. Why can the speaker utter an utterance that simultaneously establishes the distance and shortens it? How can one explain the seemingly contradictory pragmatic phenomena in these cases? It is when dual mode thinking is introduced that these phenomena can be understood as not constituting contradictions at all. The foregoing is an illustration of the fact that the traditional analytical approach based on reductionism is not sufficient to explain the use of honorifics in natural discourse. The next question is: Where does this verbal practice come from? Why is it that dual mode thinking must be understood and applied to make sense of the use of honorifics? The root of this way of thinking, which may appear to be foreign to the Western scientific tradition, must be sought in the tradition of Eastern philosophy, but that is beyond the scope of this paper.
6.
The Japanese Language Commission and the use of honorifics
It is no exaggeration to say that the use of honorifics in Japanese is a matter of considerable discussion by the government. In December 2000, the Japanese Language Planning Commission, set up by the Agency of Culture, issued a report entitled “Polite Expressions in Contemporary Society”. Even though the phrase “polite expressions” was used, the intention was to provide guidelines for the use of honorifics suitable for the new society in the 21st century. It was the second time in history that the Japanese government had called for a commission on the use of honorifics. The first guideline was introduced in 1952 to reflect the newly established society after World War II. That was when Japanese society became democratic, with a new constitution for a democratic country that was introduced by the United States during the period of occupation.
464
Sachiko Ide and Kishiko Ueno
At the turn of the current century, with globalization going on, a reexamination of the use of honorifics was in order. The use of honorifics, which is at the core of Japanese pragmatics, has the potential to work as an obstruction in a modernized 21st century society. This could be because the use of honorifics appears to presuppose non-equality in the status of the conversational participants, or because it leads to an inefficient transmission of information, or because it is unfair to the growing number of immigrants from foreign countries, who have trouble learning the contextual information relevant to the use of honorifics. In view of the fact that the report would influence the future of the country, since it was supposed to lay the foundation for making textbooks authorized by the education ministry for all the schools under college level, serious discussions were carried out, taking in the opinions of forty-five representatives from various sectors of society concerned with language and speech. The Language Planning Commission did not aim to produce prescriptive guidelines for the use of honorifics, but was deliberately geared toward describing the principles governing the use of honorifics as they are used in present-day Japan. Acknowledging the differences in geography, age, and gender, the commission carefully dealt with the common core of the use of honorifics in Japanese. When the question was raised, for example, as to what should be done about the humbling use, which is used conventionally as a ritual practice, the commission had a long discussion about whether to abolish it or not. They finally decided that the humbling use, either using humbling honorifics or other humbling phrases, is an essence of Japanese pragmatics. When one speaks with humbling expressions, the meta-pragmatic function of humbling the speaker is to raise the addressee. When the raised addressee in return chooses a humble expression in response, it lowers the current speaker and raises the current addressee. Thus, the Japanese principle of pragmatics is to swing like a seesaw. Speaking by humbling themselves with reference to their conversational partners results in establishing the status of the two as equal. This can be interpreted as a ritualized verbal art, geared toward maintaining and strengthening the bonds of relation in accordance with the value of modesty Japanese people share. This verbal art can be explained in terms of the linguistic ideology that can be traced back to the philosophy of Mahayana Buddhism, where one of the principal teachings is muga ( ), meaning ‘no self’. A series of discussions over 20 months at the Language Planning Commission led to the proposition of a new concept called keii hyoogen (‘polite expression’), which is defined as expressions that are concerned with the participants and the setting. It is to be noted that the status of honorifics was discussed as follows: Honorifics play the central role, but are not the only expressions included in this concept. In other words, the concept includes other expressions that are defined as polite when examined in context. For example, when borrowing a book, one can make the utterance polite by either using honorifics as in (23), or by phrasing the request as a question as in (24).
Honorifics and address terms
(23)
(24)
465
Go hon wo o kashi kudasai. HON book ACC HON lend give-HON ‘Lend (me) (your) book.’ Hon wo kashite kureru ka. book ACC lend give Q ‘Will you lend (me) (your) book?’
The Language Planning Commission thus determined that polite expressions consist of at least two types, i.e., the one with honorifics and the one without using honorifics, even though in reality both types are mixed together as in (25). (25)
Go-hon wo o kashi kudasai masu ka. HON-book ACC HON lend give-HON ADD HON Q ‘Would you do me the favor of lending (me) (your) book?’
The report described the functions of honorifics and other polite expressions as follows. Their function is twofold: first, to index the differences of status or role, the familiarity of the interlocutors, and the formality of the setting, and second to index how speakers want to present themselves in the context of speaking. The practice of appropriate indexing of the relationship of the speaker, the addressee, the referent, and the formality of the setting, as well as the speaker’s position in context is the essence of the non-volitional aspect of linguistic politeness, which was introduced in Section 4 as wakimae. It is to be noted that the appreciation of verbal performance is not determined by the speaker’s own judgment. There is little room for speakers to use their own will, unless it is used for the purpose of producing creative meanings by transgression. Examples of such transgressions include irony, sarcasm, or contemptuous expressions created by being overly polite. As long as speakers properly subsume themselves in the ba, a semantic space, their perception of elements of ba occurs automatically, just as actors act in an improvised drama, not by conscious analysis of the contextual elements and a calculation to match the context. This instant agreement of the language selection in alignment with elements of ba conforming with generally accepted belief should be explained in terms of ritualized practice in everyday life. It is in this way that a congenial and harmonious atmosphere is created by appropriate indexing of the participants and the formality of the setting in the shared space and time of the ba common to the participants involved. Since the criteria of appropriateness are shared in terms of shared values in the minds of the participants, appropriate verbal performance results in synergizing the perception and cognition of every participant in each ba, and thus results in making participants relaxed and comfortable in a harmonious atmosphere because of the alignment of the form in ba. In a society where harmony in ba is traditionally sought, as described in the first article of the first constitution founded by Prince Shoutoku in the 8th century, verbal perform-
466
Sachiko Ide and Kishiko Ueno
ance creating harmony above and beyond the referential content has been maintained, even in contemporary society. Whether it is appropriate in the global community is the question to be posed.
7.
Conclusion
The academic tradition in linguistics has taken the objective perspective when looking at the linguistic and pragmatic phenomena in question, and examined them from an objective standpoint. This approach has resulted in detailed analyses and structural descriptions of intricate linguistic and pragmatic phenomena, leading to the discussion of theories of universals as well as highlighting the diversities of these phenomena. However, it puts the speaker of the language outside the conversational context. Most scholars working on honorific languages and address terms have dealt with these phenomena as though they were language devices to manipulate speech styles and registers. However, what has been lacking in this approach is the inside perspective of the native speakers. Comprehensive explanations of how and why the honorifics and address terms are used in the everyday lives of the native speakers and what it means for the speakers to speak with honorifics would be possible if these phenomena were examined from the speakers’ perspectives in the speech event. It is from this stance that this chapter has focused on the discussion of the pragmatics of only one language, Japanese, the language of the authors. Furthermore, the logic of ba has been introduced to account for the mechanism of the inseparable relation of the linguistic signs the speakers use and the context of speaking, which is one of the problems that up to now has proved difficult to explain. Investigations of further languages that employ honorifics and other politeness systems will show the extent to which the concept of ba has more general applicability and explanatory power.
Acknowledgements The authors would like to express our sincere acknowledgement to Dr. Karin Aijmer and Dr. Gisle Andersen, who carefully read the manuscript and gave invaluable comments to the earlier drafts of the manuscript.
Honorifics and address terms
467
Abbreviations used in glosses ACC ADD HON CON COP DAT FP GEN HON HON COP HUM HON NEG NOM PAST Q QT REF HON SUB SUP HON TOP
accusative addressee honorific connective particle copula dative final particle genitive honorific honorific copula humble honorific negative nominative past question quotative referent honorific subject super honorific topic
Notes 1. The term “deference” instead of “respect” is used throughout in view of the nature of the honorific use in Japanese society. However, when the literal meaning of the Japanese word sonkeigo is mentioned, “respect/deference” is used. 2. Most of the literature in Japanese linguistics categorizes this kind of humbling form as kenjoogo. In 2007, the language section of the Cultural Commission issued a report concerning the new categorization of kenjoogo: kenjoogo I ‘object honorifics’ and kenjoogo II ‘humble forms’. 3. The “Mr. O Corpus” is a cross-linguistic video corpus, collected under a Grant-in-Aid for Scientific Research from the Japanese Society for the Promotion of Science for the project on “Empirical and theoretical studies on culture, interaction, and language in Asia” (No. 15320054).
468
Sachiko Ide and Kishiko Ueno
References Agha, Asif 1994 Honorification. Annual Review of Anthropology 23: 277–302. Agha, Asif 2002 Honorific registers. In: Kuniyoshi Kataoka and Sachiko Ide (eds.), Culture, Interaction, and Language, 21–63. Tokyo: Hituzi Shobou. Braun, Friederike 1988 Terms of Address: Problems of Patterns and Usage in Various Languages and Cultures. Berlin/New York: Mouton de Gruyter. Brown, Roger and Albert Gilman 1960 The pronouns of power and solidarity. In: Thomas A. Sebeok (ed.), Style in language, 253–276. Cambridge, MA: MIT Press. Brown, Penelope and Stephen Levinson 1978 Universals in language use: Politeness phenomena. In: Esther N. Goody (ed.), Questions and Politeness, 56–289. Cambridge, U.K.: Cambridge University Press. Brown, Penelope and Stephen Levinson 1987 Politeness: Some Universals in Language Usage. Cambridge, U.K.: Cambridge University Press. Brown, Roger and Marguerite Ford 1964 Address in American English. In: Dell Hymes (ed.), Language in Culture and Society, 234–244. New York: Harper and Row. Geertz, Clifford 1960 The Religion of Java. Chicago: The University of Chicago Press. Hanks, William F. 2005 Explorations in the deictic field. Current Anthropology 46(2): 191–220. Hill, Beverly, Sachiko Ide, Shoko Ikuta, Akiko Kawasaki and Tsunao Ogino 1986 Universals of linguistic politeness: Quantitative evidence from Japanese and American English. Journal of Pragmatics 10: 347–371. Hurford, James R. 2007 The Origins of Meaning. Oxford/New York: Oxford University Press. Ide, Sachiko 1982 Japanese sociolinguistics: Politeness and women’s language. Lingua 57(2–4): 357–385. Ide, Sachiko 1989 Formal forms and discernment: Two neglected aspects of linguistic politeness. Multilingua 8(2–3): 223–248. Ide, Sachiko, Motoko Hori, Akiko Kawasaki, Shoko Ikuta, and Hitomi Haga 1986 Sex difference and politeness in Japanese. International Journal of Sociology of Language 58(2): 25–36. Irvine, Judith T. 1992 Ideologies of honorific language. Pragmatics (3): 251–262. Irvine, Judith T. 1995 Honorifics. In: Verschueren, Jef. and Jan-Ola Östman (eds.), Handbook of Pragmatics, 1–22. Amsterdam/Philadelphia: John Benjamins.
Honorifics and address terms
469
Irvine, Judith T. 2009 Honorifics. In: Senft, Gunter, Jan-Ola Östman and Jef Verschueren (eds.), Culture and Language Use, 156–172. Amsterdam/Philadelphia: John Benjamins. Ishizaka, Shozo 1944 Keigo-shi Ronkou [The History of Honorifics]. Tokyo: Oyashima Shuppan. Kasuga, Kazuo 1977 Keigo no hensen 1 [The development of honorifics 1]. In: Susumu Ono, and Takeshi Shibata (eds.), Keigo [Honorifics], 93–134. (Iwanami Kouza Nihongo 4 [The Iwanami Handbook of the Japanese Language 4].) Tokyo: Iwanami Shoten. Kindaichi, Kyosuke 1959 Nihon no Keigo [Japanese Honorifics]. Tokyo: Kadokawa Shoten. Koshal, Sanyukta 1987 Honorific systems of the Ladakhi language. Multilingua 6: 357–385. Lakoff, Robin 1970 Tense and its relation to participants. Language 46(4): 838–849. Lakoff, Robin 1973 The logic of politeness; or minding your p’s and q’s. Papers from the Ninth Regional Meeting of Chicago Linguistic Society, 292–305. Chicago. Lakoff, Robin 1975 Language and Woman’s Place. New York: Harper and Row. Leech, Geoffrey 1983 Principles of Pragmatics. London/New York: Longman. Levinson, Stephen 1983 Pragmatics. Cambridge, U.K.: Cambridge University Press. Martin, Samuel 1964 Speech levels in Japan and Korea. In: Dell Hymes (ed.), Language in Culture and Society, 407–415. New York: Harper and Row. Matsumoto, Yoshiko 1987 Politeness and conversational universals: Observations from Japanese. Paper presented at the 1987 International Pragmatics Conference. Antwerp, Belgium. Matsumoto, Yoshiko 1989 Politeness and conversational universals: Observations from Japanese. Multilingua 8(2–3): 207–222. Sanada, Shinji 1993 The dynamics of honorific behavior in a rural community in Japan. Multilingua 12(1): 81–94. Shibata, Takeshi 1988 Nihonjin no keigo [Honorifics of the Japanese people]. Kokubungaku [Japanese Literature] 33–15, 6–10. Shimizu, Hiroshi 1987 A general approach to complex systems in bioholonics. In: R. Graham and A. Wunderlin (eds.), Lasers and Synergetics, 204–423. Berlin: Springer-Verlag. Shimizu, Hiroshi 1990 A dynamical approach to semantic communication: The significance of biological complexity in the creation of semantic formation. In: Hiroshi Shimizu
470
Sachiko Ide and Kishiko Ueno
(ed.), Biological Complexity and Information, 145–185. Singapore: World Scientific. Shimizu, Hiroshi 1996 Seimeichi toshiteno Ba no Ronri [The Logic of Ba as the Concept for Understanding the Living State]. Tokyo: Chuuoukouronsha. Shimizu, Hiroshi 2000 Kyousou to basho [The Synergetic creation and place]. In Hiroshi Shimizu (ed.), Ba to Kyousou [Ba and the Synergetic Creation], 3–177. Tokyo: NTT Shuppan. Tahara, Keiko, Kikuko Tsukada and Hitoshi Nakasone (eds.) 1966 Keigohou Kenkyuu Bunken Souran [The Compendium of Literatures on Honorifics]. Kokubungaku [Japanese Literature] 11–8. Tokieda, Motoki 1941 Kokugogaku Genron [The Foundation of Japanese Linguistics]. Tokyo: Iwanami Shoten. Toyama, Eiji 1977 Keigo no hensen 2 [The development of honorifics 2]. In: Susumu Ono, and Takeshi Shibata (eds.), Keigo [Honorifics], 135–167. (Iwanami Kouza Nihongo 4 [The Iwanami Handbook of the Japanese Language 4].) Tokyo: Iwanami Shoten. Tsujimura, Toshiki 1963 Keigo no bunrui ni tsuite [On the taxonomy of honorifics]. Gengo to Bungei [Language and the Art of Literature] 5(2): 8–13. Tsujimura, Toshiki 1968 Keigo no Shiteki Kenkyuu [The Historical Study of Honorifics]. Tokyo: Tokyodou Shuppan. Tsujimura, Toshiki 1992 Keigo Ronkou [The Study of Honorifics]. Tokyo: Meiji Shoin. Tsujimura, Toshiki (ed.) 1991 Keigo no Youhou [The Usage of Honorifics]. Tokyo: Kadokawa Shoten. Watanabe, Minoru 1971 Kokugo Koubun-ron [The Study of Japanese Sentence Structure]. Tokyo: Hanawa Shobou. Yamada, Yoshio 1924 Keigohou no Kenkyuu [Study on the Grammar of Honorifics]. Tokyo: Houbunkan. Yamaguchi, Akiho 2004 Nihongo no Ronri: Kotoba ni Arawareru Shisou [The Logic of Japanese: Japanese Thought Reflected in the Language]. Tokyo: Taishuukan Shoten.
V. Sequential patterns and activities
15.
Social and pragmatic variation in the sequential organization of talk Jan Svennevig and Ronny Johansen
1.
Introduction
Conversation and other forms of verbal interaction are characterized by the fact that utterances are organized in sequences and are produced and understood according to the sequential context they appear in. This chapter investigates how sequential organization varies according to social and pragmatic parameters, such as speech community, medium of communication and activity type. After an illustrative example of what types of variation may affect sequential organization, the chapter presents and discusses Conversation Analysis (CA) as an approach to socio-pragmatic variation. Some examples of variation in sequential structure are then reviewed in order to show how sequence organization is linked to aspects of the social setting and the activity at hand. The cases in point are third position receipts of answers and the organization of preferred and dispreferred responses. Finally, we present an empirical analysis of gender variation in the use of displays of affection (hugs and declarations of affection) in closing sequences of chat interaction among youths.
2.
Variation in opening sequences: how are you?
In this section we will address the question of variation in relation to a single type of sequence, namely opening sequences, and more specifically, adjacency pairs introduced by the first pair part how are you?. This type of sequence has received considerable analytic attention, and may serve as a good example of what sorts of socio-pragmatic variation may relate to the sequential structure of talk. A place to start looking at variation in sequential organization is the internal context of the talk. A significant point made in Conversation Analytic research is that sequential position is crucial to understanding a linguistic or other semiotic token. So, for instance, Sacks in one of his lectures shows how the question how are you? may be used in two different sequential positions (Sacks 1992, I: 549ff). It may be used as a proper beginning of a conversation, that is, as a greeting substitute and thus not be answered as a question but responded to by a greeting in return. But more typically, it may be used after an initial greeting, in which cases it will be answered by a report such as fine or all right. Consequently, it is not only that the question is understood differently in the two environments, but also that what counts as a relevant response is different.
474
Jan Svennevig and Ronny Johansen
Variation may also be addressed according to the medium of communication being used. Sacks probably referred to face-to-face interaction in his lecture (although he does not state it explicitly). Schegloff (1986), however, notes that how are you-sequences in telephone openings systematically occur after initial greetings, and are thus not used there as conversational openers. Responses to the question come in three types, positive (terrific), negative (awful) and neutral (fine). Neutral responses are closure relevant and indicate that the speaker’s well-being is not relevant at that point and thus that the conversation may move on to the next sequence, such as the initiation of a “first topic”: (1) (from Schegloff 1986:114) Hyla: Hwaryuhh= Nancy: = Fi.’ne how’r you, Hyla: Okay:[y, Nancy: [Goo :d, (0.4) Hyla: mkhhh [hhh Nancy: [What’s doin, Negative and positive responses, on the other hand, engender sequence expansion and serve to topicalize the state of well-being (for instance by the interlocutor asking for an account of the situation). That the medium of communication is consequential for opening sequences has been additionally demonstrated by Arminen and Leinonen (2006). They show that in Finnish, the standard opening of landline phone calls is self-identification by the receiver (such as Tiina?). Receivers of mobile phone calls, by contrast, only use this opening when the caller is unknown. When the caller is known (and the name appears on the display) they instead answer by greeting the caller (such as no moi – ‘hi’). A main interest in Conversation Analytic research is variation according to the activity engaged in by the participants. In a special type of informal conversation we find yet another pattern. Drew and Chilton (2000) have studied telephone calls where relatives call each other on a regular basis just in order to “keep in touch”. Such conversations are generally made at a weekly scheduled time and the routine character of the call is amongst other things indexed by the fact that it is the called, and not the caller, who initiates the how are you-sequences. These calls contrast with ones that are not meant to be recognized as routine keeping in touch-conversations – but instead unscheduled or otherwise extraordinary telephone calls – where the caller will instead keep to the “standard” pattern of initiating the how are you-sequence. The question of who enquires about the other thus becomes an index of what sort of activity the speakers see themselves as engaged in. A major focus on sequential variation is found in the research on institutional interaction. The thrust of findings in this vast area of research concerns the differ-
Social and pragmatic variation in the sequential organization of talk
475
ences in organization between different institutional activities and relationships. Openings in institutional encounters between experts or service providers and clients or customers mainly differ from openings in informal conversation in that the exchange of how are yous is absent. Variation may also be studied according to national culture. As noted, the standard telephone opening in Finnish is self-identification rather than Hello. Lindström (1994) shows that this is also the pattern in Swedish (landline) telephone conversation (generally a combination of greeting and self-identification, such as hej det er Henrik – ‘Hi this is Henrik’). Furthermore she notes that how are yousequences are infrequent in telephone openings. When they occur, they are seldom reciprocated, as is the norm in American English openings (Schegloff 1986). In Greek telephone openings, by contrast, Sifianou (2002) finds such sequences to be “almost invariably present”. Furthermore, they may occur as what Sacks referred to as a greeting substitute, that is, directly after the receiver’s Hello and without any other greeting. She also notes that, in contrast with American openings, Greek how are you-questions are often answered “literally”. The wording of the question may be translated as both ‘how are you doing’ and ‘what are you doing’, thus opening up for an update on the receiver’s activities. The sequential organization of openings is thus different in Swedish, Finnish, Greek and American English telephone conversation. However, the differences do not seem to be total or absolute. Certain basic actions identified for telephone openings are the same, such as summons-answer, speaker identification, and greetings. This may alert us to an important point about socio-pragmatic distribution and variation, namely that it is crucial to notice also similarities between cultures and communities. The identification of similarities may in the long run lead to hypotheses about potentially universal or generic aspects of talk in interaction (Schegloff 2002, 2006). CA researchers have been less willing to study variation in conversational organization according to macro-sociological variables such as gender, ethnicity, age and socio-economic status. A study on how are you in an age perspective can illustrate some of the problems a CA position has with such approaches. Coupland, Coupland and Robinson (1992), working within the framework of interactional sociolinguistics, report on responses to the question how are you in interviews with elderly people in a day center. The interviews were presented to the elderly as a survey of experiences of health care. The interviewers were instructed to say how are you as the first verbal move – “said smiling and uniformly without any clear primary stress” (Coupland, Coupland and Robinson 1992: 220). The researchers found that most interviewees responded with either positive appraisals (such as alright thank you) or with hedged or qualified negative appraisals (such as not too bad or well (.) could be better). They claim that the respondents display an interpretation of the question as phatic and an orientation to “a preference for non- or at least hedged negativity” (Coupland, Coupland and Robinson 1992: 225). How-
476
Jan Svennevig and Ronny Johansen
ever, there were also some instances of bluntly negative self-appraisals (such as “I’m a long-standing asthmatic”). These responses are interpreted as related to life stage in that it was consistent with previous findings of the researchers that elderly people had a “propensity to disclose relatively more and more intimate “painful” experiences to first acquaintances” (Coupland, Coupland and Robinson 1992: 220). However, the negative responses do not seem to be produced in a dispreferred format (although the transcription is too rudimentary to ascertain this). These responses thus seem to invalidate the claim about both preference for nonnegativity and the orientation to phaticity. An alternative explanation in terms of sequential organization rather than external social characteristics of the participants would explain the response pattern in a non-contradictory way. A CA approach would rather interpret the responses in relation to the types of activity being pursued by the speakers, either as a part of an opening phase – where small talk is routinely exchanged and a phatic response would be appropriate – or as part of the interview proper – in which case a health report would be the relevant response. Crucial to whether the question should be understood as the one or the other would be how and when the question was produced, for instance whether it was asked while the interlocutors were entering the room or when they were sitting down with a questionnaire in front of them. But information about such aspects of the situation is not presented in the article. It is also worth noting that, in addition to these positively or negatively oriented responses, the researchers found responses that were ambiguous and indeterminate between a phatic and a factual interpretation (such as “I’m alright (.) I do suffer with my nerves though I get injections every month (.) but I (.) I’m going on fine”). This type of response would also be consistent with the hypothesis that the question was ambiguous to many interviewees and that they thus chose answers that would be relevant to both situations. Incidentally, this ambiguity may have been caused by the artificial and constructed character of the talk and by the scripted opening line of the talk. It seems likely that this would disorient many interviewees, who might rather expect some sort of greeting or welcome as the opening of the interview.
3.
CA as an approach to variation
The case discussed above illustrates two different perspectives on socio-pragmatic distribution and variation. In contrast to traditional variationist sociolinguistics, which defines social variables beforehand and independently of the discourse being analyzed, CA seeks primarily to describe identities and social relations displayed by the participants in the interaction as manifestly relevant to them. CA thereby disregards external characteristics that cannot be shown to be attended to by the participants in their mode of speaking and interacting. CA does not claim that other contextual factors do not influence talk, but merely that they are not rel-
Social and pragmatic variation in the sequential organization of talk
477
evant to the ethnomethodological project of CA, which is to explicate the interpretive resources of the participants. If there are such influences they are beyond the participants’ orientation in conversational practices, and beyond what can be studied in the emic perspective of CA. Another important point this case illustrates, is the reflexivity of talk and context. It is not the perspective of CA to look at how aspects of the context “influence” talk in a unilateral sense (cf. Introduction to the volume). Instead, CA treats talk as a major resource through which context is established and modified. Thus, the context cannot be defined in advance as a stable structure, but is partly constructed by the participants through their talk. For instance, by producing a report on their health, the respondents enact an institutional identity (as research interviewees) and not primarily an identity as elderly. So, instead of investigating how pre-defined phatic questions are responded to or how the a priori defined group of elderly people respond in interviews, CA sees responses as partly constitutive of the context. This means that a phatic response will display an understanding of the question as an introduction to an opening phase of small talk, whereas a factual response will display an understanding of it as an interview question. CA has often been criticized for not attending to the socio-cultural context of conversations. For instance, Coupland, Coupland and Robinson (1992) accuse this tradition of analyzing “decontextualized conversation with its inherent “mechanisms”” (Coupland, Coupland and Robinson 1992: 225). And to a certain extent it seems true that many CA researchers are more interested in generic organizations of practices than in socio-cultural variation. Schegloff (2006: 83), for instance, makes the point that the fundamental practices of turn-taking, sequence organization and repair are indeed generic to talk in interaction, and thus should be considered as candidate universals. But he makes an important qualification on just what it is that may be generic: The import of the claim that these organizations are generic is not that the way talk in interaction is done in the United States, or modern industrialized societies, is generic; it is that the organizational issues to which these organizations of practice are addressed are generic. Conversation analysis is not averse to finding, indeed celebrating, what appear to be differences in interaction in other cultures, societies, and languages. (Schegloff 2006: 84)
CA may thus be considered an approach which has as its program to identify both general, maybe even universal, principles of organization of talk in interaction and variation in the realization of these principles in different cultures or languages. Also Heritage (1997) emphasizes the difference between “the social institution of interaction” – meaning the generic organizational features of talk in interaction – and “social institutions in interaction” – meaning the different patterns of realization of talk in different institutional settings. He furthermore stresses the independence of conversational practices vis-à-vis the sociological or psychological characteristics of the participants:
478
Jan Svennevig and Ronny Johansen
[…] social interaction is informed by institutionalized structural organizations of practices to which participants are normatively oriented. […] these organizations of practices […] are fundamentally independent of the motivational, psychological or sociological characteristics of the participants. Rather than being dependant on these characteristics, conversational practices are the medium through which these sociological and psychological characteristics manifest themselves. (Heritage 1995: 396)
This being said, it is also important to underline that the CA account of sequence structure – in Schegloff’s words – “should not be understood as an inflexible template which mechanically generates “parts” assigned to various participants” (Schegloff 2007: 220). Rather it should be understood as an organizational resource which is applied in various ways relative to the activity engaged in by the interlocutors. The bulk of CA research on institutional interaction is just such a description of the special forms and functions such sequential patterns take relative to the specialized activities speakers engage in. Some of these findings will be reviewed in the next section. Before moving to an account of CA’s contribution to the study of sociopragmatic variation, another question of principle needs to be mentioned. A problem for making claims about variation is that CA studies seldom discuss explicitly the scope of their claims and the potential applicability to different situations or communities. Thereby, it is not always clear whether the mechanisms described in a specific study are meant to be understood as generic rules of conversation, a practice associated with the specific activity engaged in, or a feature associated with the socio-cultural identities of the speakers. It may even be discussed whether some studies commit the ethnocentric fallacy of implicitly claiming universal validity of rules and norms found in one’s own culture, as pointed out by Moerman (1988). The question of generalizability of results in CA research has been discussed by Peräkylä (1997) in the following terms: In terms of the traditional “distributional” understanding of generalizability, case studies on institutional interaction cannot offer much. […] However, the question of generalizability can also be approached from a different direction. The concept of possibility is a key to this. Social practices that are possible, that is, possibilities of language use, are the central objects of all conversation analytic case studies on interaction in particular institutional settings (Peräkylä 1997: 215, emphases as in the original).
The citations from Heritage and Peräkylä point to CA as a study of the linguistic and sequential resources or tools available to participants, but seem to deny the possibility of linking them in any systematic way to specific social institutions or identities. The procedures are independent of the institutional context and are merely possibilities of action in various settings. An approach to the question of generalization and distributional variation is suggested by Drew (2003), who recommends more explicitly contrastive studies between different types of activities or “genres” of interaction. His own analysis concerns the deployment of formulations in various institutional activities. Formu-
Social and pragmatic variation in the sequential organization of talk
479
lations are utterances that summarize the gist or the upshot of a preceding stretch of talk, often introduced by the discourse marker so (Heritage and Watson 1979). He compares the results of four previous studies that have identified different functions of formulations related to the institutional activity at hand. In psycho-therapy, formulations are mainly used by the therapist, who by formulating the gist of the patient’s utterance offers an interpretation of some implicit or allusive “message” in it. In radio phone-in programs, hosts have been found to use formulations to make a (tendentious) rephrasing of the caller’s claims, and subsequently use this to challenge and defeat the speaker’s position by a rebuttal. In news interviews, interviewers use formulations of upshot in order to invite the interviewee to assent to a strong or dramatic version of what he or she said. Finally, in industrial negotiations between management and labor unions, formulations are used to summarize the position each side is taking and propose a settlement. This approach shows that CA is not just interested in “decontextualized conversation” but may address both generic features and variation in a highly precise manner. It also illustrates the difference between the generic organization of practices and the context-dependent realizations of them. Certain conversational practices – such as formulations – have an independent structural organization that is identifiable as common traits across situations and communities. But on the other hand, they may be deployed differently and acquire different acitivity-specific functions in different situation types. Contrastive analyses of this type have the potential of addressing the problem raised by Peräkylä (1997) as well, namely of ascertaining the distributional generalizability of Conversation Analytic findings. In order to illustrate what sorts of problems may be involved in ascertaining the distributional validity of a claim, we will give an example from a study of repetition in consultations at a job training centre for immigrants in Norway (Svennevig 2004). Here it was observed that answers by the clients, who were non-native speakers of Norwegian, were massively repeated by the clerk. The extract below includes two such instances: (2) IFF 1 S: hva jobba mannen din med før dufør han ble pensjonist a? (.) A: S: jobba på Freia ja. A: ja he (h)lage sjokolade. S: jaja. A: He he he S: det er jo godt det. A: mja he he (S SMILES) (.)
S: what did you husband do before youbefore he retired? (.) A: S: worked at Freia (yes). A: yeah he (h)make chocolate. S: right. A: He he he S: that sure is good. A: myeah he he (S SMILES) (.)
480
Jan Svennevig and Ronny Johansen
S: men det er kanskje ikke så bra for kroppen. A: he (h)nei det er færlig.(h) S: det er farlig for kroppen ja. A: °ja.°
S: but it’s maybe not so good for the body. A: he (h)no it’s dangerous.(h) S: it’s dangerous for the body (yes). A: °yeah.°
In the first instance, the client (A) gives a short answer that is somewhat incongruent with the question in that she supplies just a company name rather than a job designation. The clerk (S) repeats this answer but simultaneously expands it by adding a verb phrase that makes the answer more congruent with the question. In the second instance, the non-native speaker’s answer is also slightly deviant in that it contains the unidiomatic adjective farlig (‘dangerous’) rather than usunt (‘unhealthy’) or the like.1 Here again, the native speaker repeats the answer rather than corrects it, but in addition expands it into a more explicit and unambiguous phrase. After the identification of certain such patterns of repetition in receipts of information (mainly performed by the native clerk), the question arose what types of practices these were and the range of their applicability. At least four different possibilities presented themselves as plausible: 1 Were they general receipt practices that could be found in any type of conversation? 2 Were they related to the specific institutional setting and activity at hand – the clerk gathering information about the client’s past experiences? 3 Were they related to the medium involved – the clerk frequently taking notes of the answers that were given? 4 Were they related to the non-native character of the answers given – often involving accent, unidiomatic formulations etc.? In order to answer the question about distributional range definitively one would need to compare different sets of data. Repetition has been widely studied by other researchers, so a comparison with previous studies is possible. Different studies give some support to all the possibilities. For instance, several studies have noted that repeats are common as a receipt token in informal conversation, such as Schegloff (1997) and Sorjonen (1996), and that they may in these cases perform specific communicative functions over and above marking receipt, such as confirming allusions (Schegloff 1996). Other studies have noted the presence of such receipts in institutional interviews. Mazeland and ten Have (1996) found that receipts were especially frequent in information gathering phases after information that was important to retain in an accurate form. Finally, studies of interaction between competent and not-yet-competent speakers (such as children and non-native learners) have shown that adults use repeats as scaffolding – partly to display a construal of the previous utterance, partly to expand and correct potentially deviant
Social and pragmatic variation in the sequential organization of talk
481
utterances (cf. Long 1996; Clark 2008). This may speak in favor of seeing these repeats as a practice associated with differences in linguistic competence. Although these different studies may provide points of comparison and contrast, they do not give sufficient information to explain fully why these receipts are used – and used so massively – in just these conversations. Once again, the approach CA recommends for answering this question is to analyze the instances themselves in depth in order to see what the participants seem to orient to as relevant in using such repeats. In the case above, the expanded form of the repeats shows an orientation by the native speaker towards disambiguating and “normalizing” the non-native’s utterance. Thereby, an orientation to the non-fluent character of the answer seems to be involved. In addition, the repeats were seen to be especially frequent as receipts of precise factual information in the form of names and numbers (such as Freia above). This pointed towards interpreting them as displays of what was being recorded as the official answer to the question, and thus as a typically institutional practice. In conclusion, then, questions of distribution and variation cannot be answered merely by noting the presence of various forms in different sets of data. The relevance to the activity at hand as displayed by the normative orientations of the participants has to be the basis for claims about distribution.
4.
Variation in types of sequential organization
In this section we will review some of the types of contrastive and variationist CA studies that have been performed until now. We have already considered some examples of how sequential organization differs in different activity types. There is not room here to give an extensive overview of all the types of variation in sequence organization, but we will consider a few more examples which have been shown to involve rather extensive variation and which have been given some attention in the literature. The first is the use of third position receipts and acknowledgements, and the other is preference organization. 4.1.
Third position receipts
Adjacency pairs are often expanded with a third move (a post-expansion) in the form of a receipt or acknowledgement. Three central types of minimal expansion of adjacency pairs are receipts of information (oh), acknowledgements (okay) and assessments (Schegloff 2007). These forms of expansion vary a great deal from setting to setting, and reveal interesting insights about the nature of the activities carried out. An early study of classroom discourse carried out by Sinclair and Coulthard (1975) (not within the CA framework) showed that teacher eliciting exchanges had the three part structure illustrated in Table 1:
482
Jan Svennevig and Ronny Johansen
Table 1.
Structure of classroom interaction
Initiative:
elicit
Teacher:
How do you use your muscles
Response:
reply
Pupil:
By working
Feedback:
evaluate
Teacher:
by Fworking yes
Coulthard and Brazil (1981) further specified the form of the third position evaluation by noting that it most often took the form of a yes produced with high pitch or a high pitched repetition (or reformulation) of the answer plus an agreeing item (as in the example above). This pattern is clearly related to the activity at hand, namely instruction. It appears in cases where teachers ask questions to which they already know the answer (so-called test questions). The evaluating receipt is thus functionally related to the didactic activity of testing the pupils’ knowledge. When people ask questions they do not know the answer to, they may mark the answer as news to them by producing the receipt token oh. According to Heritage (1984) this token indicates a change of state in the speaker from non-knowing to knowing. However, in certain institutional settings this use of oh is systematically avoided, such as in news interviews (Heritage 1985). Here, interviewers do not signal their reactions to answers, and the explanation is again revealing about the activity and the setting. Heritage (1985) explains this sequential pattern as being motivated by (and co-constitutive of) the institutional roles and relationships between the participants. By not responding to the interviewee’s answers interviewers treat them as being directed not to them personally, but to the broadcast audience. Thereby they take on a role of mediator between interviewee and audience. Furthermore, the lack of response is also functional in embodying a neutralistic stance in relation to the content of the answer. In this way, the lack of receipts to answers (third position receipts) is a key characteristic of the institutional role of an objective journalist and of carrying out the activity of speaking for an overhearing audience. A final example will show yet another activity-specific use of third position receipts. Button (1992) shows that in job interviews, interviewers systematically use third position expansions that do not provide the interviewees the opportunity to elaborate or clarify their answers. One of the practices is to give a receipt in the form of a simple thank you and go on to a new and unrelated question. This treats the answer as complete but does not reveal anything about how the interviewer has understood or evaluated the answer. The second practice is to provide an assessment, but move directly on to a new question without providing the candidate the opportunity to return to the answer. What these practices have in common is that they abort any further interactional negotiation or elaboration of the content of the answer. The answers are left as the interviewees’ sole responsibility, and the interviewers are thereby distanced from the answers received. This contributes to es-
Social and pragmatic variation in the sequential organization of talk
483
tablishing the interviewer’s neutrality and objectivity, not leading the interviewee in one or the other direction. 4.2.
Preference organization
Another type of organization which has been shown to vary according to situations and activities is preference organization. The original studies by Sacks (published in 1987) described a general preference for agreement, and Pomerantz (1984) substantiated this claim by showing how second assessments were produced in different formats according to whether they aligned with the first assessment or not. The concept of preference involves two different aspects. The first is the type of action that is preferred or dispreferred as a response, such as agreeing assessment versus disagreeing assessment. The second is the turn design of preferred and dispreferred responses. A preferred response format involves responding directly, quickly (“contiguously”) and in a short (minimal) form. A dispreferred response format is characterized by such things as initial pauses and hesitation, delay of the response proper (the declining or disagreeing action) by various types of prefaces, mitigation of the dispreferred action, and expansion of the response by the addition of explanations, accounts and the like (Pomerantz 1984). Several researchers have investigated situations that involve variation in the applicability of this claim, especially in institutional activity types. Atkinson and Drew (1979) have shown that in court proceedings, accusations are followed by denials in a preferred format. In fact, initial pausing and hesitation can give rise to the inference that the allegations are admitted to a certain extent. Greatbatch (1992) has shown how the turn-taking system in news interviews, with an interviewer acting as a moderator of the discussion between the parties, renders mechanisms such as pausing and delay ineffective as markers of dispreference. But also the other aspects of dispreference differ from ordinary conversation in that disagreements are not systematically delayed or mitigated. Greatbatch explains this as a consequence of the fact that the disagreements are formulated as answers to questions from the interviewer rather than as direct responses to the co-interviewee, and by being addressed to a third party the disagreements are “automatically” mitigated (Greatbatch 1992: 279). Sometimes the interviewees violate this turn-taking system and respond directly to each other, and even in these cases the preference features associated with disagreement in ordinary conversation are not observed. Also Clayman and Heritage (2002) show that interviewers may formulate questions and reformulate answers in ways that invite disagreement as the preferred type of response. For instance, reformulations may recast the prior answer in more extreme and provocative terms, and thereby invite the interviewee not just to respond to it, but to dispute it (Clayman and Heritage 2002: 309). All these observations seem to portray news interviews and panel debates as conversational activities which seek and encourage opposition and polarization be-
484
Jan Svennevig and Ronny Johansen
tween parties of a discussion. This points to an important principle, namely that preference organization should be considered as associated with the goals of the activity at hand. Consequently, it may be a too strong claim that there is a universal or generic “preference for agreement” in conversation. Rather, what is generic is the availability (and recognizability) of preferred and dispreferred turn designs. What needs to be further analyzed and specified is which types of responses are preferred and dispreferred in which types of activity. Also in ordinary conversation there is extensive variation in preference organization related to the types of actions being carried out. Already in Pomerantz’ study (1984), a notable exception was formulated: after self-deprecations the preferred response is disagreement rather than agreement. A type of activity that has been widely discussed concerning the status of preference is disputes. Kotthoff (1993) claims that in disputes the preference for agreement is changed to a preference for unmitigated disagreement. In such oppositional contexts, the speakers are expected to defend their position, and they do so without formatting it as a dispreferred action. In this activity type concessions to the opponent are the dispreferred actions and are typically produced in a dispreferred format with extensive hesitation and mitigation. There are some instances of direct or upgraded formulations of agreement (that is, in a preferred format), but they only occur as prefaces to further disagreement (sometimes taking the form of conventional idioms such as das ist ja schön und gut or ‘that’s all very well’). The conclusions drawn by Kotthoff are not consistently confirmed by later studies, however. Kangasharju (2009), for instance, finds that disagreementoriented phases (in which disagreements are produced in a preferred format) are rather short and alternate with “smoother” phases where the preference for agreement seems to be in operation. She thus concludes that the evidence does not support claiming that a preference for disagreement is a normative principle in disputes as such. A related type of sequence is responses to complaints. Dersley and Wootton (2000) describe two sorts of responses to complaints. One is the denial of any involvement in the complained-of action (“didn’t do it” denials) and the other – the most frequent of them – is a more partial denial, which admits some involvement in the action but denies culpability (“not at fault” denials). They show that the first type is produced in a preferred format while the other is produced in a dispreferred format. This study shows the importance of being specific in delimiting the actions and sequence types involved when making claims about preference. Maybe the different results concerning preference in disputes are an indication that the category dispute is too wide and heterogeneous, and that more specific types of disputing actions or activities have to be studied in order to discern the norms of preference.
Social and pragmatic variation in the sequential organization of talk
5.
485
Displays of affection in chat interaction
A sequence type which shows clear differences according to the relationship between the speakers is closing sequences of chat interaction. In this section we present an empirical study of gender variation in such sequences and argue that this shows that the participants orient to gender as a feature of the context with recognizable consequences for the interaction (so-called procedural consequentiality, Schegloff 1992). In addition we make a comparison between the sequential organization of closings in chat and face-to-face conversation. The data comes from a collection of 195 chat exchanges between Norwegian youths aged 14 to 15, using MSN as their chat program. The interactions are social chats which occurred naturally between friends and acquaintances while at home in their free time.2 The exchanges are repartitioned according to gender in table 2.3 Table 2.
Number of chat exchanges in the corpus
Participants
Total
Girl-girl
79
Girl-boy
86
Boy-boy
30
Total
5.1.
195
Closing sequences in chat interaction
Closing sequences in chat interaction are generally initiated by one of the parties announcing some reason for ending the conversation (most often presented as some external requirement obliging them to quit, such as having to go to bed or to do homework). After this the participants may move directly into a final closing exchange in four different ways. 1 2 3 4
Goodbye tokens (equivalent to Bye-bye etc.) Reassurance of future contact (such as See you tomorrow, Let’s keep in touch etc.) Hugs and kisses: (Hug, Kisses etc.) Declarations of affection (Love you, You and me forever etc.)
These possibilities are exemplified in the next four excerpts below: (3) Sander|| xD says: Sander|| xD says: men stikker jeg, skal “nyte” det fine but I’m off, gonna “savour” the været xD nice weather xD
486
Jan Svennevig and Ronny Johansen
Sander|| xD says: Hadebra Ken Rune says: hade
Sander|| xD says: Goodbye Ken Rune says: bye
In (3) the reason for ending is announced in the first line, followed immediately by a goodbye token by the same speaker (writer). The goodbye is reciprocated by the interlocutor and the exchange ends. (4) Heidi – Hans Arne