Discourse Markers and Beyond: Descriptive and Critical Perspectives on Discourse-Pragmatic Devices across Genres and Languages [1st ed. 2020] 978-3-030-37762-5, 978-3-030-37763-2

Table of contents :
Front Matter ....Pages i-xvii
Preliminary Issues: Category Membership, Methodology, Alternative Perspectives on Discourse Markers (Péter B. Furkó)....Pages 1-35
Front Matter ....Pages 37-37
Discourse Markers in Natural Conversations, Scripted Conversations and Political Interviews: Core and Peripheral Uses (Péter B. Furkó)....Pages 39-63
Discourse Markers from a Critical Perspective: Some Theoretical Issues (Péter B. Furkó)....Pages 65-90
Discourse Markers from a Critical Perspective: A Case Study of Discourse Markers in Parliamentary Speeches (Péter B. Furkó)....Pages 91-116
Front Matter ....Pages 117-117
The Use of Discourse Markers in Business English Textbooks: Issues in L2 Communicative Competence and Learners’ Input (Péter B. Furkó)....Pages 119-140
Discourse Markers in Scripted Discourse I: Issues of (Under)Specification in the Translation of Reformulation Markers (Péter B. Furkó)....Pages 141-164
Discourse Markers in Scripted Discourse II: The Representation and Translation of Irish English Stereotypes in Contemporary Cinematography (Péter B. Furkó)....Pages 165-186
Discourse Markers and Their Translation in Literary Discourse: A Case Study of Discourse-Pragmatic Devices in The Hobbit (Péter B. Furkó)....Pages 187-218
The Use of Discourse Markers in Bible Translations (Péter B. Furkó)....Pages 219-245
Back Matter ....Pages 247-285

Citation preview

POSTDISCIPLINARY STUDIES IN DISCOURSE SERIES EDITOR: JOHANNES ANGERMULLER

Discourse Markers and Beyond Descriptive and Critical Perspectives on Discourse-Pragmatic Devices across Genres and Languages

Péter B. Furkó

Postdisciplinary Studies in Discourse

Series Editor Johannes Angermuller Centre for Applied Linguistics University of Warwick Coventry, UK

Postdisciplinary Studies in Discourse engages in the exchange between discourse theory and analysis while putting emphasis on the intellectual challenges in discourse research. Moving beyond disciplinary divisions in today’s social sciences, the contributions deal with critical issues at the intersections between language and society. Edited by Johannes Angermuller together with members of DiscourseNet, the series welcomes high-quality manuscripts in discourse research from all disciplinary and geographical backgrounds. DiscourseNet is an international and interdisciplinary network of researchers which is open to discourse analysts and theorists from all backgrounds. Editorial Board Cristina Arancibia Aurora Fragonara Péter Furkó Tian Hailong Jens Maesse Eduardo Chávez Herrera Michael Kranert Jan Krasni María Laura Pardo Yannik Porsché Kaushalya Perera Luciana Radut-Gaghi Marco Antonio Ruiz Jan Zienkowski More information about this series at http://www.palgrave.com/gp/series/14534

Péter B. Furkó

Discourse Markers and Beyond Descriptive and Critical Perspectives on Discourse-Pragmatic Devices across Genres and Languages

Péter B. Furkó Department of English Linguistics Károli Gáspár University of the Reformed Church in Hungary Budapest, Hungary

Postdisciplinary Studies in Discourse ISBN 978-3-030-37762-5 ISBN 978-3-030-37763-2  (eBook) https://doi.org/10.1007/978-3-030-37763-2 © The Editor(s) (if applicable) and The Author(s) 2020 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Cover illustration: RooM the Agency/Alamy Stock Photo This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Contents

1 Preliminary Issues: Category Membership, Methodology, Alternative Perspectives on Discourse Markers 1 1 Theoretical and Empirical Approaches to Discourse Markers 1 2 Characteristics of Discourse Markers, Criteria for Discourse Marker Status 6 2.1 Non-propositionality and Optionality 6 2.2 Context-Dependence 8 2.3 Multifunctionality 8 2.4 Weak Clause Association and Variable Scope 9 2.5 Procedural Meaning/Non-compositionality 12 2.6 High Frequency, Orality, Stigmatization 13 3 Automatic Semantic Annotation: Testing Its Methods and Precision 15 4 Corpus and Methodology 16 5 Findings 18 6 Conclusions, Utility of USAS as a Heuristic Tool 27 7 Alternative Perspectives on Discourse-Pragmatic Devices: Outline of the Volume 28 References 32 v

vi      Contents

Part I  Discourse Markers Across Genres 2 Discourse Markers in Natural Conversations, Scripted Conversations and Political Interviews: Core and Peripheral Uses 39 1 Introduction 39 1.1 Types of Reports: Direct Reports, Indirect Reports and Voicing 40 1.2 Research Questions 42 2 Research Material 43 3 Automated Semantic Annotation and Keyness of Reporting Verbs and Expressions 46 4 The Use of Discourse Markers in Reporting Based on the Findings of Manual Annotation 50 5 Prototypical and Peripheral Uses of Discourse Markers in Reporting Across Four Genres 52 5.1 Core and Peripheral Examples from the MPI Sub-corpus 52 5.2 Core and Peripheral Examples from Scripted Discourse 56 5.3 Core and Peripheral Examples from Celebrity Interviews and Natural Conversations 57 6 Summary, Conclusions 59 References 61 3 Discourse Markers from a Critical Perspective: Some Theoretical Issues 65 1 Introduction 65 1.1 Discourse Marker Research and Its Relevance to (Critical) Discourse Theory 66 1.2 Perspectives on the Manipulative Potential of Discourse Markers 67 2 Previous Micro-Analyses of Manipulative Intent in Political Discourse 69

Contents     vii

3 The Political News Interview as a Genre 71 4 Data and Methodology 72 5 Discourse Markers Marking Manipulative Intent in Political Interviews 74 5.1 Evidential Markers: Suppression and Backgrounding 74 5.2 General Extenders: Playing Down the Importance of Alternative Viewpoints 77 5.3 The Role of Quotation Markers in Decontextualizing and Recontextualizing Texts, Legitimizing Opinions and Polarizing the Audience 79 5.4 Other Manifestations of Manipulation: Conversationalization and the Exploitation of Ambiguity 81 6 Conclusions, Directions for Further Research 83 References 85 4 Discourse Markers from a Critical Perspective: A Case Study of Discourse Markers in Parliamentary Speeches 91 1 Introduction: Populism and Populist Discursive Strategies 91 2 Contextual Background 93 3 Research Questions, Corpus and Methodology 95 4 Characteristics of Parliamentary Speeches 96 5 Propositional Lexical Items and Indexicals Used as Manifestations of Populist Discursive Strategies 98 6 Discourse Markers and Modal Adverbs Used as Manifestations of Populist Discursive Strategies 102 7 Conclusions, Directions for Further Research 113 References 114

viii      Contents

Part II  Discourse Markers Across Languages 5 The Use of Discourse Markers in Business English Textbooks: Issues in L2 Communicative Competence and Learners’ Input 119 1 Introduction 119 2 Components of Communicative Competence 120 3 The Role of Discourse Markers in Shaping Learners’ Communicative Competence 122 4 Characteristics of Discourse Markers as Sources of Learners’ Difficulties 125 5 Mapping the Functional Spectrum of Discourse Markers in a Corpus of Business English Textbooks 126 5.1 Research Process 126 5.2 The Functional Spectrum of Well in Naturally Occurring Discourse 127 5.3 The Functional Spectrum of Of Course in Naturally Occurring Discourse 129 5.4 The Functional Spectrum of Well and Of Course in a Corpus of Business English Textbooks 130 6 Conclusion 136 References 138 6 Discourse Markers in Scripted Discourse I: Issues of (Under)Specification in the Translation of Reformulation Markers 141 1 Introduction 141 1.1 Contrastive Approaches to Discourse Markers 142 2 Preliminary Considerations 143 2.1 The Translation of Discourse Markers—Theory and Practice 143 2.2 From Scripted Discourse to the Language of Subtitles 146 3 Previous Accounts of Reformulation 148 3.1 Definitions and Lists of Reformulation Markers 148

Contents     ix

3.2 The Functional Spectrum of I Mean 151 3.3 The Functional Spectrum of Actually 152 4 The Study 153 4.1 Data and Methodology 153 4.2 Findings 155 4.3 Discussion, Translation Strategies and Implications 157 5 Conclusions, Limitations and Directions for Further Research 159 References 160 7 Discourse Markers in Scripted Discourse II: The Representation and Translation of Irish English Stereotypes in Contemporary Cinematography 165 1 Introduction 165 1.1 The Concepts of “Stereotype” and “Stereotyping” 166 1.2 “Stereotype” and “Stereotyping” in Interactional Sociolinguistics and Sociopragmatics 168 2 Examples of Discourse-Pragmatic Stereotyping 169 2.1 Request Strategies 169 2.2 Compliment Responses (CRs) 171 2.3 Responses to Thanks (RTs) 173 3 Stereotypes Incurred by the Use of Discourse Markers 174 4 Quantitative Perspectives on IrE Stereotypes 179 4.1 Keyness Analysis and Automated Semantic Tagging of IrE Scripted Discourse with Reference to AmE Scripted Dialogues 179 4.2 Keyness Analysis and Automated Semantic Tagging of IrE Scripted Discourse with Reference to BrE Scripted Dialogues 181 5 Conclusions, Fuzzy Boundaries Between Pragmalinguistic and Sociopragmatic Features as Well as Sociocultural Norms 183 References 184

x      Contents

8 Discourse Markers and Their Translation in Literary Discourse: A Case Study of Discourse-Pragmatic Devices in The Hobbit 187 1 Introduction 187 1.1 Literary Pragmatics—Perspectives and Approaches 188 1.2 Perspectives on Tolkien’s Linguistic Beliefs and Style 189 2 Metapragmatic Awareness and Metacommunicative Reflections in Tolkien’s Novels 190 3 The Functional Spectrum of Discourse Markers in The Hobbit 195 3.1 The Use of Evidential Markers 195 3.2 Reformulation Markers—Self-Correction and Mock Technicality 198 3.3 “… the Gold and the Dragon, and All That”—General Extenders in The Hobbit 200 4 Quantitative Perspectives on Tolkien’s Literary Style and Authentication Strategies 202 5 Some Issues Regarding the Translation of Authentication Strategies 206 5.1 Strategies and Creativity in the Translation of Discourse Markers in The Hobbit 208 5.2 The Translation of Speech Community Creating Devices in the LOR Trilogy 212 6 Conclusions, Tolkien’s Philosophy of Language Reconsidered 214 References 215 9 The Use of Discourse Markers in Bible Translations 219 1 Introduction, the Contribution of Discourse Marker Research to the Study of New Testament Sentence Conjunctions 219 1.1 Categorization and Typologies 220 1.2 Optionality and Stigmatization 222

Contents     xi

2 Greek Sentence Conjunctions in John and Galatians and Their Corresponding Discourse Markers in Various Bible Translations 225 2.1 The Use of καί in Narrative and Expository Bible Texts 226 2.2 The Functional Spectrum of δέ 228 2.3 The Functional Spectrum of γάρ 231 2.4 ἀλλά as a Global Marker of Contrast 234 3 Quantitative Perspectives on the Formal and Functional Equivalence of New Testament Discourse Markers 235 4 Conclusions, Directions for Further Research 241 References 242 Concluding Remarks 247 Software and Online Tools 251 Data Availability 253 Appendix—List of Abbreviations 255 References 257 Index 283

List of Figures

Chapter 2 Fig. 1 Concordance plots of Q2.1 and Q2.2 tags across the four sub-corpora (Source Adopted from Furkó et al. [2019: 253]) Fig. 2 Presence and absence of discourse markers in reporting across the four corpora (Source Adopted from Furkó et al. [2019: 253])

48 51

Chapter 3 Fig. 1 Types of reporting across four genres: scripted discourse (SD), mediatized political interviews (MPI), celebrity interviews (CI) and natural conversation (NC) (Source Adopted from Furkó et al. [2019: 263])

81

Chapter 8 Fig. 1 Translations of well in two Hungarian editions of The Hobbit 210 Fig. 2 Translations of of course in two Hungarian editions of The Hobbit 210 xiii

xiv      List of Figures

Chapter 9 Fig. 1 Specification and underspecification in 150 verses of Galatians 237 Fig. 2 Specification and underspecification in 150 verses of John 237

List of Tables

Chapter 1 Table 1 Alternative terms for English discourse-pragmatic devices Table 2 Individual authors’ focus on categorial properties discourse markers display Table 3 Summary of discourse marker and non-discourse marker-related semantic tags assigned to the most frequent discourse marker types in the MPI and CI sub-corpora Table 4 Inter-annotator agreement between automated and manual tagging of discourse marker/non-discourse marker tokens

3 5 19 21

Chapter 2 Table 1 Semantic fields in USAS Table 2 Normalized frequencies of the USAS categories relevant to reporting

46 47

xv

xvi      List of Tables

Chapter 4 Table 1 Frequency and keyness of potentially populist uses of lexical items and suffixes in speeches given by governing and opposition parties in the period between 8 May 2018 and 18 September 2019 99 Table 2 Frequency and keyness of potentially populist uses of lexical items and suffixes in speeches given by governing and opposition parties surrounding the immigration quota referendum of 2016 100 Table 3 Frequency and keyness of potentially manipulative discourse markers in speeches given by members of governing and opposition parties in the period of 8 May 2018 and 18 September 2019 104 Table 4 Frequency and keyness of potentially manipulative discourse markers in speeches given by members of governing and opposition parties 105

Chapter 5 Table 1 The use and functional spectrum of well and of course in BE textbooks adopted from Furkó and Mónos (2013: 142–143) 132 Table 2 The use and functional spectrum of well and of course in BE textbooks analysed in this chapter 134

Chapter 6 Table 1 RMs listed by author and year of publication 150

Chapter 7 Table 1 Keyness analysis of lexical items in the IEC and SD sorted by test corpus and keyness 180 Table 2 Keyness analysis of lexical items in the IEC and BSD sorted by test corpus and keyness 182

List of Tables     xvii

Chapter 8 Table 1 Lexical items associated with authentication and the pragmatic mode, sorted by test corpus and keyness 204 Table 2 Formal indicators of the use of the pragmatic mode in the THC and the CSLC 205 Table 3 Frequency of USAS tags associated with textual, interactional and subjectivity markers in the THC and the CSLC sorted by log-likelihood 207 Table 4 Hungarian discourse markers associated with spontaneous conversations in two Hungarian editions of The Hobbit 211 Table 5 Pragmatic routines and their translation based on Bayona (2003: 81) 213

Chapter 9 Table 1 Translation Equivalents of kαί (TEs in each row are listed in order of frequency) 227 Table 2 Translation equivalents of δε (TT discourse markers in each row are listed in order of frequency) 232 Table 3 Frequency of specification and underspecification underspecification strategies in the various Bible translations 236 Table 4 The frequency of and, but and for in the KJV and the ASV 238 Table 5 Frequency of USAS tags associated with textual, interactional and subjectivity markers across the KJV and the ASV of the New Testament sorted by log-likelihood 239

1 Preliminary Issues: Category Membership, Methodology, Alternative Perspectives on Discourse Markers

1

Theoretical and Empirical Approaches to Discourse Markers

The present volume is informed by research in a sub-field of pragmatics, discourse marker research, often considered a “growth industry” (Fraser 1999: 931) and, at the same time, a “testing ground” (Bordería 2008: 1354) for pragmatic theories. Discourse markers comprise a functional class of linguistic items that do not typically change the propositional meaning of an utterance but are essential for the organization and structuring of discourse, for marking the speaker’s attitudes to the proposition being expressed as well as for facilitating processes of pragmatic inferences. A variety of approaches and definitions have been offered, each one informed by a particular theoretical framework (Conversation Analysis, Interactional Sociolinguistics, Rhetorical Structure Theory, Relevance Theory, etc., for an overview cf. Fischer 2006; Furkó 2007; Vaskó 2016; Dér 2017). Despite the rapidly growing body of discourse marker research, experts in the field observe over and over again that there are still a number of fundamental questions that need to be answered (cf. e.g. Schourup 1999; Fraser 1999; Dér 2010, 2017; Heine 2013; Vaskó 2016). Some of the © The Author(s) 2020 P. B. Furk´o, Discourse Markers and Beyond, Postdisciplinary Studies in Discourse, https://doi.org/10.1007/978-3-030-37763-2_1

1

2

P. B. Furkó

issues include the lack of generally accepted terminology and classifications and uncertainty regarding essential formal, semantic and pragmatic characteristics. The resulting terminological turmoil is illustrated in Table 1, which provides a summary of some of the terms in classical studies and more recent analyses referring to the respective discourse uses of well, of course, oh, etc.: The issue of terminology, alternative terms and overlapping categories such as pragmatic markers, connectives, contextualization cues and enunciative markers will be the focus of Chapter 3 of the volume, while a number of monographs (cf. Furkó 2007; Beeching 2016; Dér 2017) also provide detailed discussions of the theoretical assumptions behind the use of competing terms. Furkó (2007), in particular, takes the pragmatic marker-discourse marker dichotomy as its major theme. Dér (2017: 9) observes that the term discourse marker appears to be the most inclusive and frequently used in the English literature, while its Hungarian mirror translation “diskurzusjelöl˝o” also appears to be the most widespread in the respective academic community (Dér 2017: 10). Accordingly, I will be using this as an umbrella term in the present introduction and make the distinction between textual (prototypically discourse) functions and (inter)subjective (prototypically pragmatic) functions prominent in subsequent chapters where relevant. The present volume will provide a series of empirical case studies on the use of discourse markers across genres and languages as well as critical analyses of the manipulative potential of discourse markers. Therefore, the present introduction will focus on the second problem area mentioned above, i.e. the issue of categorization and category membership as well as the methods by which one can identify lexical items that are discourse markers and distinguish them from non-discourse marker uses of the source categories. Describing the characteristics of the functional class of discourse markers and developing criteria for deciding for every given instance whether or not it is a discourse marker have been major preoccupations in recent discourse marker research. Authors usually provide exhaustive lists of the formal, functional and stylistic features that are associated with discourse

int int

– – – – PP – PFM DM booster –

cue

DM/IS*

PP DP PFM DM PM DM

– – – DM – DM

IS

–

DP DM – int/pause marker

oh

– – – DM

of course

DP DM PE DM/pause marker* int –

well

– – – – – int

–

–

int int

DP – – pause marker

ah

Legend DP—discourse particle DM—discourse marker PE—pragmatic expression PM—pragmatic marker int—interjection cue—cue word/cue phrase IS—interactional signal PP—pragmatic particle PFM—pragmatic force modifier *—categorized according to the position/slot they take in the utterance

Schourup (1985) Schiffrin (1987) Erman (1987) Fraser (1990) and (1999) James (1974) Wierzbicka (1991) Hirschberg and Litman (1993) Stenström (1994) Holmes (1995) Kroon (1995) Nikula (1996) Fuller (2003) Beeching (2016) Crible (2017)

Table 1 Alternative terms for English discourse-pragmatic devices

– – – – – DM

DM/IS*

cue

– –

DP DM – DM

now

PP DP PFM DM PM DM

IS

–

DP DM PE ‘90-PM ‘99-DM – –

I mean

– DP – – – DM

–

cue

– –

– DM – DM

but

PP DP PFM DM PM DM

DM/IS*

–

– –

DP DM PE PM

you know 1 Preliminary Issues: Category Membership, Methodology …

3

4

P. B. Furkó

markers as a functional class (cf. e.g. Schourup 1999; Fraser 1999; Beeching 2016; Brinton 2017), but empirical studies rely on different subsets of such criterial features when identifying particular instances of discourse markers in a given corpus (for a detailed discussion, cf. Crible 2017). Naturally, this makes it difficult to compare the results of empirical research even if similar datasets are involved. Table 2 illustrates this problem. An even more challenging task is to develop annotation software that can automatically identify discourse markers in oral discourse and filter out non-discourse marker tokens of lexical items that are frequently used as discourse marker types (e.g. adverbial uses of well or now, prepositional uses of like, etc.). Moreover, to date, few attempts have been made to use automated means of identification involving semantic criteria and semantic fields, since one of the very criterial features of discourse markers is their semantic underspecification (cf. Crible et al. 2019), which is a result of the diachronic process of semantic bleaching (cf. Brinton 2017: 31). Accordingly, the present introduction will explore the utility of using an automated semantic tagging software, USAS as a preannotation tool for the identification of oral discourse markers, including (inter)subjective as well as textual markers. After an overview of the formal and functional features that can be used for manual annotation, and after comparing the results of manual and automatic annotation of selected discourse markers, I will argue that despite the semantic underspecification of most discourse markers, automatic semantic annotation (ASA) can be an effective tool for the disambiguation between discourse marker and non-discourse marker uses with regard to certain items, but needs to be complemented by extensive manual error correction and filtering.

(x)

x

(x)

x (x)

x

x x x

x

oral

x x x

x x x

x x x

context

x

x

x x

synt.

Legend seq.—sequentiality-coherence-connectivity context—context-dependence—context-coordination oral.—orality synt.—syntactic criteria (diversity, non-integration) proced.—procedural meaning poly-funct.—poly-functionality attitude—marking attitudes scope—variable scope, functional scope non-prop.—non-propositional content inv.—invariable form

Schiffrin (1987) Fraser (1990, 1999) Redeker (1990, 1991) Stenström (1994) Kroon (1995) Knott and Sanders (1998) Andersen (1998) Hansen (1998) Risselada and Spooren (1998) Romaine and Lange (1998) Blakemore (1987, 2002) González (2004) Crible (2017)

seq.

x

x

x

proced

x x

poly-func.

x

(x)

x (x)

(x)

(x)

attitude

Table 2 Individual authors’ focus on categorial properties discourse markers display

x

x

scope

x

x

non-prop.

x

inv. 1 Preliminary Issues: Category Membership, Methodology …

5

6

P. B. Furkó

2

Characteristics of Discourse Markers, Criteria for Discourse Marker Status

2.1

Non-propositionality and Optionality

Many scholars (cf. Schourup 1999) consider non-propositionality (nontruth-conditionality) as a sine qua non for discourse marker status, yet others include propositional items such as then and after that. While it is generally agreed that certain discourse markers (e.g. well, however, etc.) contribute nothing to the truth-conditions of the proposition expressed by an utterance, the non-truth-conditionality of others (frankly, I think) have generated a great deal of controversy (cf. Infantidou-Trouki 1992; Brinton 2017: 127ff ). Blakemore (1987: 106) argues that a distinction has to be made between truth-conditional and non-truth-conditional meaning, on the one hand, and conceptual vs. procedural meaning, on the other. Thus, many of the controversies stem from the fact that certain scholars confuse the two distinctions and use them interchangeably. Schourup (1999), for example, uses the compositionality test to argue in favour of the truthconditionality of in addition: (1a) Owens is a respected drama critic. I tell you in addition that she has written … (1b) Owens is a respected drama critic. In addition, she has written …

While in addition is indeed truth-conditional, the above test would predict that frankly is also truth-conditional. However, as Blakemore (2002) would argue, discourse marker uses of frankly are non-truth-conditional, but conceptual. It is, therefore, important to point out that the compositionality test will be a useful tool in deciding whether individual discourse markers have conceptual or procedural meaning, the truthfunctionality of discourse markers is tested more efficiently in terms of whether they retain their original meaning when embedded in if-clauses or under the scope of factive connectives such as because:

1 Preliminary Issues: Category Membership, Methodology …

7

(2a) Allegedly / Obviously / Frankly, the cook has poisoned the soup. (2b) If the cook has allegedly / ?obviously / *frankly poisoned the soup, we can eat the meal without worrying. (2c) We shouldn’t eat the soup, because the cook has allegedly/?obviously/*frankly poisoned it.

The uncertainty with regard to whether or not obviously retains its original meaning in (2c) suggests to many that the truth-functionality–nontruth-functionality distinction should be viewed as a continuum, rather than a dichotomy, which is consistent with the finding in grammaticalization theory that due to the diachronic grammaticalization processes that are synchronically manifested in the use of discourse markers, there is a fuzzy boundary between uses that are non-truth-conditional and (omissible) and those that are not (for a detailed discussion, cf. Andersen 2001; Blakemore 2002; Dér 2017). Optionality as a distinguishing feature is in many respects derivative of the previously discussed criterion of non-propositionality, and discourse markers are considered optional from the perspective of sentence meaning because their absence does not change the conditions under which the sentence is true. There are, however, two further senses in which discourse markers are claimed to be optional. Firstly, they may be seen as syntactically optional in the sense that removal of a discourse marker does not alter the grammaticality of its host sentence. Secondly, they are optional in the sense that if a discourse marker is omitted, the relationship it signals is still available to the hearer, though no longer explicitly cued (cf. Schourup 1999: 231). The above statement does not entail that discourse markers are useless; rather, it reflects the view according to which discourse markers guide the hearer towards a particular interpretation of the connection between a sequence of utterances and at the same time rule out unintended interpretations.

8

P. B. Furkó

2.2

Context-Dependence

Discourse markers’ extreme context-dependence is frequently identified with their inherent indexicality. Aijmer, for example, considers indexicality as the most important property of discourse markers, a property whereby discourse markers are linked to attitudes, evaluation, types of speakers and other features of the communicative situation (cf. Aijmer 2002: 5). In this respect, discourse markers can be compared to deictics, i.e. another borderline phenomenon can be observed if we look at some of the definitions of deictic expressions, which often overlap with those of discourse markers. Both categories are usually defined in terms of context-dependence, i.e. in terms of having meaning only by virtue of an indexical connection to some aspect of the speech event (cf. e.g. Sidnell 1998). Levinson (2004), in fact, considers discourse markers as discourse deictics, other subgroups including spatial, temporal and social deictics. Similarities between indexicals and discourse markers are also recognized by proponents of Relevance Theory. Carston, for example, notes that the two seemingly disparate phenomena are brought together by the fact that both encode a procedure rather than a concept, and both play a role in guiding the hearer in the pragmatic inferential phase of understanding an utterance (Carston 1998: 24). The difference between the two sets of phenomena, according to Carston, is that indexicals constrain the inferential construction of explicatures and discourse markers (discourse connectives in RT terms) constrain the derivation of implicatures (in other words, intended contextual assumptions and contextual effects).

2.3

Multifunctionality

In addition to playing a role in pragmatic inferencing, individual discourse markers are also associated with a plethora of functions including hedging and politeness functions. What is more, they can also be salient in conversational exchanges as openers, turn-taking devices, hesitational devices, backchannels, markers of topic shift and of receipt of

1 Preliminary Issues: Category Membership, Methodology …

9

information, and so on (cf. e.g. Beeching 2016: 4ff ). Discourse markers are inherently multifunctional and ambiguous, since there is a lot of interpersonal and discourse burden on their signalling capacity. Discourse markers signal interpersonal and discourse functions simultaneously; thus, they are ambiguous between the two levels, and on the other hand, they are vague with regard to signalling particular relations on a given level as well (ibid.). The multifunctionality of discourse markers also brings up the question of whether different uses of a given marker are to be considered incidental and unrelated (maximalist approach) or motivated and related (minimalist approach) and whether there is an invariant “core meaning” of discourse markers that is context-independent and preserves some component of the lexeme’s original semantic meaning. Since the focus of the present introduction is on differentiating between discourse marker and non-discourse marker uses of a given item, further discussion will not ensue on the multifunctionality of discourse marker uses. The theoretical issue will be picked up in Sect. 1.1 of Chapter 9 with regard to connectives, often considered the most underspecified subgroup of discourse markers (cf. e.g. Crible 2018: 211). Multifunctionality will also be an important consideration from an empirical perspective when developing coding schemes for manual annotation, as we will see in Sect. 4 of Chapter 2 as well as most of the chapters in the volume.

2.4

Weak Clause Association and Variable Scope

It is frequently observed in the literature that discourse markers usually occur either outside the syntactic structure or loosely attached to it (cf. e.g. Crible 2017: 111). Quirk et al. classify many linguistic items that are elsewhere included among discourse markers as conjuncts (e.g. nonetheless) which are considered to be clause elements but to have a detached role relative to other, more closely interrelated clause elements such as subject, complement and object:

10

P. B. Furkó

Conjuncts are more like disjuncts than adjuncts in having a relatively detached and ‘superordinate’ role as compared with other clause elements (Quirk et al. 1985: 631).

In addition, some of the items that Quirk et al. refer to as “disjuncts” (e.g. obviously, sentence initial surprisingly and frankly) also display a whole range of properties associated with the functional class of discourse markers. It is important to note that the property of weak clause association is relative to elements external to the discourse marker’s lexical form, since several discourse markers clearly have their own internal syntactic structure (e.g. on the other hand ) and others (e.g. y’know, I mean) constitute clauses from a syntactic point of view despite the fact that they are no longer considered to be compositional, but procedural (cf. e.g. Furkó 2014). Weak clause association is frequently discussed in relation to phonological independence: discourse markers often constitute independent tone units or are set off from the main clause by “comma intonation” (cf. Hansen 1997: 156). Adding weak clause association and a corresponding lack of intonational integration to our list of criteria could also be justified from the perspective of grammaticalization theory. An important clause of the definition of grammaticalization states that it takes place in special morpho-syntactic environments. In the case of discourse markers, this environment can be associated with sentence-initial position, and hence, many scholars regard quasi-initiality as yet another distinguishing feature of discourse markers (cf. e.g. Schourup 1999). However, Dér (2017: 15) questions the validity of these features on grounds of empirical evidence and distinguishes between different degrees of initiality. Moreover, once discourse markers enter an advanced stage of grammaticalization, they become syntactically independent and can appear at various parts of the sentence, with an accompanying “comma intonation”, resulting in fuzzy boundaries and borderline cases in terms of syntactic dependence /independence of individual tokens. Discourse markers’ position in an utterance can also influence their scope, which is variable, as is illustrated by (3a) and (3b):

1 Preliminary Issues: Category Membership, Methodology …

11

(3a) Interviewer: I know how close you are to your mom. How old is she? Interviewee: Well, she probably doesn’t want me to say… (3b) You’re not going to have quality if you can’t sleep and you itch and you bitch and you weep and you cry and you bloat and you can’t remember anything and you don’t have a, well, sex drive. (examples taken from Furkó 2014)

As the examples above show, the size of the linguistic unit well can take in its scope ranges from a whole sentence to a single word. Waltereit (2006) observes that this variability is a remarkable property, but it is not an exclusive feature of discourse markers, since conjunctions as a word-class (and even some individual conjunctions as a lexical item) can also have variable scope, giving the following sentences as examples: (4a) Ed and Doris loved each other. (4b) Ed worked at the barber’s, and Doris worked in a department store.

In (4a), and has scope over two NPs, and in (4b), it has scope over two clauses. However, the difference between and used as a conjunction and its discourse marker use lies in the fact that the scope of the conjunction and can always be determined in grammatical terms. It could be defined as ranging over two constituents of the same type adjacent to and, which, in turn, make up a constituent of again the same type. The scope of discourse markers, in contrast, cannot be determined in grammatical terms, as is clear from (5) below: (5) My husband got a notice t’go into the service and we moved it up. And my father died the week … after we got married. And I just felt, that move was meant to be. (Schiffrin 1987: 53, emphasis in the original)

Schiffrin (1987) concludes that and has “freedom of scope”, rather than “variable scope”, since “we can no more use and to identify the

12

P. B. Furkó

interactional unit that is being continued than we can use and to identify the idea that is being coordinated” (Schiffrin 1987:150). Traugott (1995) relates the feature of variable scope to grammaticalization and argues that in addition to nominal clines (nominal adposition > case) and verbal clines (main verb > tense, aspect, mood marker), which are “staples of grammaticalization theory”, a further cline: Clause internal Adverbial > Sentence Adverbial > Discourse Particle should be added to the inventory (Traugott 1995: 1). According to Traugott, this cline involves increased syntactic freedom and scope. Brinton (2017: 24) further refines Traugott’s (1995) clines and adds scope within the proposition > scope over the proposition > scope over discourse as a separate cline in the evolution from propositional to textual and interpersonal meaning.

2.5

Procedural Meaning/Non-compositionality

Although most scholars treat non-compositionality as a property of discourse markers per se (for a detailed account, cf. Brinton 2017), Blakemore (2002) associates discourse markers with procedural meaning and uses non-compositionality as a test to decide whether individual items are conceptual or procedural. Blakemore also claims that if discourse markers are synonymous with their non-discourse marker counterparts, they encode conceptual meaning. Thus, seriously and in other words in (6a) and (7a) encode a concept parallel to (6b) and (7b), respectively. On the other hand, well (as in 8a) encodes a procedure, since it is not synonymous with well in (8b): (6a) Seriously, you will have to leave. (6b)He looked at me very seriously. (7a)In other words, you’re banned. (7b)She asked me to try and put it in other words. (8a)A: What time should we leave? B: Well, the train leaves at 11.23. (8b)You haven’t ironed this very well.

1 Preliminary Issues: Category Membership, Methodology …

13

A second test Blakemore uses is to see if a given item can combine with linguistic items encoding conceptual meaning to produce complex expressions. As far as the question of synonymity is concerned, it is important to note that the fact that, on the basis of native intuitions, no correspondence can be found between the adverbial well and its discourse marker counterpart, does not mean that such a relationship is absent (cf. e.g. Furkó 2013). Native intuitions, naturally, disregard diachronic aspects of individual lexical and grammatical items, and it is exactly these aspects that account for the fuzziness of the category of discourse markers and the resulting borderline cases.

2.6

High Frequency, Orality, Stigmatization

Last but not least, some of the stylistic features core members of the functional class of discourse markers display need to be considered. While semantic-functional properties are more important in determining class membership than formal and stylistic ones, stylistic criteria can also be helpful in determining discourse marker status and differentiating between discourse marker and non-discourse marker tokens. It is important to note that high frequency of use is the backbone of various processes of grammaticalization as well as pragmaticalization (cf. e.g. Furkó 2014; Dér 2017). In other words, the more frequently an item is used, the more likely it is that its formal-functional properties are going to change, and once it has entered the process of grammaticalization, the faster it is going to go through the substages of that process. A number of studies on discourse markers observe that the frequency of discourse markers can be primarily observed in speech (e.g. Beeching 2016); what is more, one of the most salient features of oral style is the use of items such as well, right, ok and you know. For example, in their classical study, Brown and Yule (1983: 17) label well, erm, I think, you know, if you see what I mean, I mean, of course “prefabricated fillers”, when drawing up a list of contrasting characteristics of spoken and written language. They also point out that these items’ overuse is often stigmatized by prescriptivists (ibid.).

14

P. B. Furkó

However, it is easy to illustrate the meaningfulness and distinctive (as opposed to random) use of even the two most used discourse markers, you know and I mean. As Fox Tree and Schrock (2002: 731) illustrate, it matters where you know or I mean appears in an utterance and they are not interchangeable: (9a) Original: me and the Edinburgh girl got together after dinner late in the evening and decided they’d really got us along to make it look right, you know they had after all had candidates from other universities. Alternative: me and the Edinburgh girl got together after dinner you know late in the evening and decided they’d really got us along to make it look right, they had after all had candidates from other universities. (9b) Original: but I don’t think it’s feasible. I mean I know this is the first time I’ve done it, and I’m not in a main line paper, but I’m sure it’ll take me all my time to do it in three weeks. Alternative: but I don’t think it’s feasible. I know I mean this is the first time I’ve done it, and I’m not in a main line paper, but I’m sure it’ll take me all my time to do it in three weeks. (example taken from Fox Tree and Schrock 2002: 731)

In (9a) Original, you know comments on what is meant by “look right”, whereas in (9a) Alternative it comments on what “after dinner” means (in other words, they differ in what they take within their scope, see Sect. 2.5 above). In (9b) Original, I mean comments on why the speaker says “I don’t think it’s feasible”, without overwriting the statement, but in (9b) Alternative, I mean comments on “I know”, retrospectively treating it as a false start. Moreover, as both manual and automatic annotation will illustrate, there is no principled basis on which one could exclude from the functional class of discourse markers connectives such as however, after all, consequently and a range of other items characteristic of formal style, some of which (e.g. besides, however, moreover ) are in fact included in Brown and Yule’s (ibid.) list of the characteristics of written language. The issue of overuse and stigmatization will be taken up in Sect. 3 of Chapter 7.

1 Preliminary Issues: Category Membership, Methodology …

3

15

Automatic Semantic Annotation: Testing Its Methods and Precision

After the above discussion of the categorial features, discourse markers are usually associated with I cannot but agree with Crible (2017), who argues that “any categorical definition is only useful insofar as it is endorsed by an empirical model of identification and annotation” (2017: 99). She also provides an overview of most of the above features and comes to the conclusion that they can be reduced to three criteria that are helpful when identifying discourse marker types (1) and setting apart discourse marker tokens from non-discourse marker tokens (2&3): (1) multifunctionality, (2) syntactic (non-)integration and (3) functional scope, respectively (Crible 2017: 105). As for the issue of automatic annotation, there are a variety of computerized semantic tagging (CST) systems, including artificial intelligencebased, knowledge-based, corpus-based and semantic taxonomy-based systems (for an overview, cf. e.g. Prentice 2010). By way of introducing the methodology used in the following chapters, the present analysis draws on the results gained from the UCREL Semantic Analysis System (USAS), which has the major advantage of combining these approaches. Furthermore, USAS groups lexical items in terms of a taxonomy of semantic fields and assigns semantic categories to all words, including grammatical and other procedural (non-propositional) items, which is relevant for the present dissertation in view of the fact that the lexical items under scrutiny are highly procedural and semantically bleached (cf. Sect. 2 above). USAS system uses an automatic coding scheme of 21 semantic fields (for details, cf. Table 1 in this Chapter), subdivided into 232 subcategories. For reasons of brevity, only the tags that have been associated with the discourse marker types under analysis will be discussed in the present section, and the complete coding scheme can be found at http://ucrel.lancs.ac.uk/usas/. USAS uses disambiguation methods including part-of-speech tagging, general likelihood ranking, multiword-expression extraction, domain of discourse identification and contextual rules (for a detailed discussion, cf. Rayson et al. 2004). Previous

16

P. B. Furkó

evaluations of the accuracy of the system reported a precision value of 91% (ibid.), i.e. a 9% margin of error applying to lexical items across the board (including propositional and non-propositional items). The research questions in the introductory study were as follows: 1. Are the disambiguation methods USAS uses sufficient for filtering out non-discourse marker tokens of the most frequent discourse marker types? 2. Does the margin of error reported to apply in general apply to the identification of discourse markers as well? 3. Are individual discourse markers identified/tagged with a similar margin of error? 4. If individual discourse markers are tagged with varying precisions by USAS, what formal-functional properties of the relevant discourse markers might explain the differences?

4

Corpus and Methodology

In the course of the research, two sub-corpora of the same size (100,000 words each) were used: • a corpus of the official transcripts of 39 confrontational type of mediatized political interviews (henceforth MPI sub-corpus) downsampled from BBC’s Hard Talk and Newsnight (available at http://bbc.co.uk), the transcripts comprise a total of 99,225 words ± 2%, allowing for technical/transcript-specific information such as the indication of participants’ names; • a corpus of the official transcripts 50 celebrity interviews (henceforth CI sub-corpus) downsampled from CNN’s Larry King Live (available at http://www.cnn.com), each interview lasts approximately 50 minutes, the total word count (100,436 ± 2%) of the CNN sub-corpus thus makes it comparable to the BBC sub-corpus.

1 Preliminary Issues: Category Membership, Methodology …

17

To increase comparability, the downsampling technique (cf. Khosravinik 2010) involved selecting the same time periods, similar topics and themes of the interviews. Further genre-specific details about the two sub-corpora will be provided in this Chapter of the volume. In the present section, the results of automatic tagging will be compared to findings based on manual annotation with respect to the D-values as well as the functional distribution of individual lexical items (see Sect. 5 below). The research process has been as follows: in order to identify and compare the USAS tags of oral discourse markers in the two sub-corpora, the semantic tags assigned to frequent discourse markers (e.g. I mean, you know, in other words, so, well ) were considered, and then these semantic tags were used to identify further types and tokens relevant to discourse marking. It was found that 95.1% of the instances of discourse markers trawled from the two sub-corpora through this method are either tagged with Z4, described in the USAS manual as the “discourse bin” (including items such as oh, I mean, you know, basically, obviously, right, yeah, yes) or with A5.x, described as “evaluative terms depicting quality” (including discourse markers such as well, OK, okay, good, right, alright ). The frequency of the relevant tags across the two sub-corpora was compared, as well as the ratio between discourse marker-relevant tags (i.e. Z4 and A5.x) and non-discourse marker-relevant tags (e.g. B2, I1.1, T1.3, etc., see below for details). In the second stage, a representative sample of 400 tokens in the MPI sub-corpus was manually annotated using a numeric code of 1 for discourse marker and 2 for non-discourse marker tokens with a view to comparing the results of automatic and manual tagging. When deciding if an individual token is a discourse marker or not, Crible’s (2017) criterial features (2017: 99, see also Sect. 2) were applied by a single expert annotator. The tokens that were selected for the sample were weighted for their frequency in the corpus, while discourse marker and non-discourse marker tokens were included in equal proportions. For example, the 429 tokens of well comprise 19.6% of all automatically tagged items; thus, 78 tokens (39 A5.1-tagged and 39 non-A5.1 tagged by USAS) were included in the sample.

18

5

P. B. Furkó

Findings

Table 3 summarizes the raw frequency of the relevant lexical items’ discourse marker and non-discourse marker-related USAS tags. Since both sub-corpora were compiled in a way that they are of the same size of 100,000 words, the raw frequencies can also be compared as normalized frequencies. As a first step, the ratio of discourse marker and non-discourse marker tokens of individual items was compared with the results of previous research in the course of which discourse markers in the same subcorpora were manually annotated (cf. Furkó and Abuczki 2014 and Sect. 3 in this Chapter). In order to gauge the categorial multifunctionality of discourse markers, the measure of D-function ratio or D-value (a term proposed by Stenström 1990) was used. An individual item’s D-value is calculated as a quotient of the number of tokens that fulfil discourse-pragmatic functions and the total number of occurrences in a given corpus. The D-value of oh, for example, is 1 (100%) in the London-Lund Corpus, since it is used exclusively as a discourse marker, whereas well showed a D-value of 0.86 (86%), as 14% of its tokens serve non-discourse marker (adverbial, nominal, etc.) functions (ibid.). If we calculate the D-values of individual discourse markers based on the above values and compare them to the findings of previous research, we see that the results of automatic annotation and manual annotation converge to a great extent. Mean, for example, has a D-value of 0.808 in the MPI corpus based on automatic annotation (calculated as the number of Z4 tags divided by all tokens of mean, i.e. 141), while manual annotation yielded a D-value of 0.797 (cf. Furkó and Abuczki 2014: 50). Similarly, manual annotation yielded a D-value of 0.82 for well in the MPI corpus (Furkó and Abuczki 2014: 54), while Table 1 yields a D-value of 0.839 for this lexical item (360 Z4 tags divided by the total number of tokens, i.e. 429). The table also correctly predicts that most of the lexical items under scrutiny have higher D-values in the CI sub-corpus than in the MPI sub-corpus, which is explained by the fact that there is a higher degree of conversationalization in celebrity interviews, i.e. they are more similar to spontaneous, informal, face-to-face conversations (cf. this chapter and

frequency of DM-related tag in the MPI

360xA5.1 14xZ4 4xZ4 205xZ4 6xZ4 114xZ4 4xZ4 165xA5.4 126xZ4 55xZ4, 53xA5.3

lexical item

well (429) sort (38) now (299) (you) know (346) like (97) (I) mean (141) (in other) words (11) actually (165) (I) think (549) right (114)

312xA5.1 25xZ4 1xZ4 455xZ4 17xZ4 201xZ4 13xZ4 72xA5.4 121xZ4 211xZ4, 98xA5.3

frequency of DM-related tag in the CI 14xI1.1, 55xN5 21xA4.1, 3xA1.1.1 288xT1.1.2, 7xZ5 140xX2.2, 1xZ6 51xZ5, 40xE2+ 27xQ1.1 7xQ.3 0 423xX2.1 6xT1.1.2

frequency of non-DM-related tag in the MPI

1xA7, 2xB2, 24xN5 10xA4.1 229xT1.1.2, 6xZ5 307xX2.2 238xZ5, 139xE2+ 30xQ1.1, 5xS2.2.2 7xQ.3 0 319xX2.1 12xN3.8, 16xS7.4, 15xT1.1.2

frequency of non-DM-related tag in the CI

Table 3 Summary of discourse marker and non-discourse marker-related semantic tags assigned to the most frequent discourse marker types in the MPI and CI sub-corpora

1 Preliminary Issues: Category Membership, Methodology …

19

20

P. B. Furkó

Chapter 2). For example, the D-value of well is 0.92, and the D-value of mean is 0.851 in the CI sub-corpus based on automatic annotation (312 A5.1 tags divided by a total of 339 tokens, 201 Z4 tags divided by a total of 236 tokens, respectively). In the second stage of the research, a representative sample of tokens in the MPI was manually annotated using numeric 1 for discourse marker tokens and 2 for non-discourse marker uses. With a view to comparing the results of automatic and manual annotation, all discourse markerrelated tags (Z4 and A5.x) yielded by USAS were re-coded as numeric 1, while non-discourse marker tags (B2, I1.1, T1.3, etc.) were re-coded as 2. Consequently, the extracted list of the corresponding manual and automated tags was entered into a reliability calculator (Freelon’s ReCal 2 for 2 coders) in order to calculate inter-annotator agreement statistics. Table 4 shows the result. Although the above inter-coder agreement values appear high (cf. Spooren and Degand 2010), it is important to note that there is a great degree of variation in the precision with which individual discourse markers are tagged by USAS. On the one hand, there are discourse markers such as I mean and you know whose discourse marker and non-discourse marker uses are disambiguated with surprising precision (resulting in a kappa score of 0.05); between the CNN sub-corpus and a random sample of 100.000 words of the Santa Barbara Corpus of American English (cf. Sect. 2 in Chapter 1) as a reference corpus (χ 2 = 0.4286, p > 0.05); and between the BBC corpus and the London-Lund Corpus, used as a reference corpus of British English (χ 2 = 0.5167, p > 0.05).

3 Discourse Markers from a Critical Perspective …

83

Yet another global manipulative intent is manifested in the exploitation of the ambiguity of discourse markers in political discourse. It is commonly observed in the literature that I think is the most notoriously ambiguous discourse marker (cf. Aijmer 1997), and, as it happens, it is the most frequently used discourse marker in the BBC corpus (50 tpttw as opposed to the 26 tpttw in the non-surreptitious conversations of the LLC, Juilland’s D = 0.68, CV = 0.32/31.58%). Holmes, for example, identified “two distinct and contrastive [core] functions of I think” (Holmes 1990: 199), expressing either uncertainty or certainty. Thus, I think can function as a booster or a downgrader, it can make a point specific/emphatic or backgrounded, and it can express involvement or detachment. In the BBC sub-corpus, there are several cases where the ambiguity cannot be resolved with reference to either suprasegmental features (intonation, stress and pitch) or co-occurrence patterns with modal verbs and adverbs (cf. Fetzer 2014). As a result, the data suggests that I think is used strategically and the ambiguity between emphasis and backgrounding, certainty and uncertainty, subjectivity and stance taking is often intentional. The probability of manipulative intent is further increased if one considers that there are social class differences in terms of the use of particular discourse markers. Huspek (1989) suggests that I think has different functions and interpretations for working-class speakers and powerholding speakers: while it marks certainty for middle-class speakers, it marks solidarity among members of the working class (Huspek, 1989 cited in Aijmer and Simon-Vandenbergen 2009: 15). By using I think, therefore, politicians can appeal (or so they might believe) to the rationality of the middle-class listeners and the solidarity of the working-class audience, at the same time.

6

Conclusions, Directions for Further Research

Because of its inherently semasiological nature, discourse marker research has been characterized by micro-analyses and bottom-up approaches to a variety of discourse-pragmatic phenomena. However, if discourse marker

84

P. B. Furkó

research is informed by CDA, an inherently top-down approach preoccupied with ethical and societal concerns (cf. Angermuller et al. 2014: 11), one can gain new insights into distributional patterns that were previously unnoticed. The genre-based analysis of political news interviews and celebrity interviews (cf. Chapters 1 and 2) revealed that the functional spectrum of discourse markers in celebrity interviews is in many ways similar to that in spontaneous informal conversations, while discourse markers fulfil fewer interactional and more heteroglossic functions in more prototypical political news interviews. However, on the basis of the analysis of manipulative uses, further refinements to the distributional patterns of discourse markers can be made. As was seen above, celebrity interviews are markedly different from spontaneous face-to-face conversations in that manipulative uses are as salient in celebrity interviews as in close-to-the-core political news interviews. These tendencies were revealed after taking a look at co-occurrence patterns with pronominal choices and vocatives, both of which have been extensively studied as manipulative in the relevant CDA literature. Conversely, discourse marker research has much to offer in informing CDA analyses and can either reinforce observations made on the basis of other lexical and morpho-syntactic choices or reveal new discursive strategies of control and dominance through quantitative analyses of functional distributions, D-values and co-occurrence patterns. Wodak and Meyer (2009) observe that “CDA researchers very rarely work with interactional texts such as dialogues” (Wodak and Meyer 2009: 10), a tendency which seems to have changed in recent years (cf. e.g. Mölder 2010; Kantara 2012). The analysis of interactional texts, therefore, requires a new toolbox, to which discourse marker research can provide a useful addition. A promising direction for further research is the analysis of discourse markers’ use from a multimodal perspective. A pilot study (cf. Abuczki and Furkó 2015) suggests that there are correlations between particular discourse marker functions and the non-verbal signs (facial expressions, eyebrow movement, hand gestures, etc.) communicators make while or immediately prior to uttering discourse markers. Multimodality has long been a focus of CDA research; thus, a critically informed, multimodal analysis of discourse marker used in political interviews might reveal even

3 Discourse Markers from a Critical Perspective …

85

more “traces of the punctuation of the unconscious” (Angermuller et al. 2014: 158) as well as influencing or manipulative intent.

References Abuczki, Ágnes, and B. Péter Furkó. 2015. Towards the Construction of a Decision Tree for the Functional Disambiguation of Hungarian DSDs. Poster presented at TextLink: Structuring Discourse in Multilingual Europe, First Action Conference, 25–28 January, Louvain-la-Neuve, Belgium. Aijmer, Karin. 1997. I Think—An English Modal particle. In Modality in Germanic Languages: Historical and Comparative Perspective, ed. T. Swan and O.J. Westvik, 1–47. Berlin and New York: Mouton de Gruyter. Aijmer, Karin, and Anne-Marie Simon-Vandenbergen. 2009. Pragmatic Markers. In Handbook of Pragmatics: 2009 Installment, ed. Jan-Ola Östman and Jef Verschuere, 1–29. Amsterdam and Philadephia: John Benjamins Publishing Company. Allott, Nicholas. 2005. The Role of Misused Concepts in Manufacturing Consent: A Cognitive Account. In Manipulation and Ideologies in the Twentieth Century, ed. L. de Saussure and P. Schulz, 147–168. Amsterdam: John Benjamins Publishing Company. Angermuller, Johannes. 2014. Post-Structuralist Discourse Analysis—Subjectivity in Enunciative Pragmatics. Basingstoke, Houndmills: Palgrave Macmillan. Angermuller, Johannes, Dominique Maingueneau, and Ruth Wodak (eds.). 2014. The Discourse Studies Reader—Main Currents in Theory and Analysis. Amsterdam and Philadelphia: John Benjamins Publishing Company. Armasu, Veronica-Diana. 2013. The Concept of Politeness in Everyday and Professional Discourse. PhD dissertation, Babes-Bolyai University. http:// 193.231.20.119/doctorat/teza/fisier/1427. Accessed 15 January 2016. Authier-Revuz, Jacqueline. 1984/2014. Hétérogénéité(s) énonciative(s). (Enunciative heterogeneities) Langages 73: 98–111, In The Discourse Studies Reader—Main Currents in Theory and Analysis, ed. Johannes Angermuller, Dominique Maingueneau, and Ruth Wodak, 155–165. Amsterdam and Philadelphia: John Benjamins Publishing Company. Bakhtin, Mikhail. 1987. The Dialogic Imagination: 4 Essays. Austin: University of Texas Press. Becker, Annette. 2007. Are You Saying…? A Cross-Cultural Analysis of Interviewing Practices in TV Election Night Coverages. In Political Discourse in the Media, ed. Anita Fetzer and Gerda Lauerbach, 139–162. Amsterdam: John Benjamins Publishing Company.

86

P. B. Furkó

Beeching, Kate. 2016. Pragmatic Markers in British English—Meaning in Social Interaction. Cambridge: Cambridge University Press. Billig, Michael. 2008. The Language of Critical Discourse Analysis: The Case of Nominalization. Discourse & Society 19 (6): 783–800. Blakemore, Diane, and Fabrizio Gallai. 2014. Discourse Markers in Free Indirect Style and Interpreting. Journal of Pragmatics 60: 106–120. Blommaert, Jan. 2005. Discourse—A Critical Introduction. Cambridge: Cambridge University Press. Bramley, Nicolette Ruth. 2001. Pronouns of Politics: The Use of Pronouns in the Construction of ‘Self ’ and ‘Other’ in Political Interviews. PhD dissertation, Australian National University. https://digitalcollections.anu.edu.au/ bitstream/1885/46225/5/01front.pdf. Accessed 10 August 2016. Clayman, Steven, and John Heritage. 2004. The News Interview—Journalists and Public Figures on the Air. Cambridge: Cambridge University Press. Connel, Ian. 1980. Television News and Social Contrast. In Culture, Media, Language: Working Papers in Cultural Studies, ed. Stuart Hall, Doothy Hobson, Andrew Lowe, and Paul Willis, 139–156. London: Hutchinson. de Saussure, Louis. 2007. Pragmatic Issues in Discourse Analysis. Critical Approaches to Discourse Analysis Across Disciplines 1 (1): 179–195. Dedaic, Mirjana N. 2005. Ironic Denial: Toboze in Croatian Political Discourse. Journal of Pragmatics 37 (5): 667–683. Dér, Csilla Ilona. 2017. Diskurzusjelöl˝ok a spontán beszélt nyelvben. habilitációs dolgozat, Budapest: ELTE-BTK. Downing, Angela. 2009. Surely as a Marker of Dominance and Entitlement in the Crime Fiction of P. D. James. Brno Studies in English 35 (2): 79–92. Ducrot, Oswald. 1984. Le dire et le dit. Paris: Minuit. Fairclough, Norman. 1995. Media Discourse. London: Edward Arnold. Fetzer, Anita. 2000a. Contextualization Cues: More- and Less-Fuzzy Hedges, Discourse Markers and Interpersonal Markers. Discourse Particles, Modal and Focal Particles and all that stuff … An International Conference on Particles, Linguistic Society of Belgium, Brussels. Fetzer, Anita. 2000b. Negotiating Validity Claims in Political Interviews. Text 20 (4): 1–46. Fetzer, Anita. 2007. Challenges in Political Interviews: An Intercultural Analysis. In Political Discourse in the Media, ed. Anita Fetzer and Gerda Lauerbach, 163–192. Amsterdam: John Benjamins Publishing Company. Fetzer, Anita. 2014. I Think, I Mean and I Believe in Political Discourse: Collocates. Functions and Distribution. Functions of Language 21 (1): 67–94.

3 Discourse Markers from a Critical Perspective …

87

Fetzer, Anita, and Elda Weizman. 2006. Political Discourse as Mediated and Public Discourse. Journal of Pragmatics 38: 143–153. Fischer, Kerstin (ed.). 2006. Approaches to Discourse Particles. Oxford: Elsevier. Fowler, Roger, Bob Hodge, Gunther Kress, and Tony Trew. 1979. Language and Social Control. London: Routledge. Fraser, Bruce. 1996. Pragmatic Markers. Pragmatics 6: 167–190. Furkó, Bálint Péter. 2007. The Status of Of Course as a Discourse Marker. HUSSE 8 Conference Proceedings. http://husse-esse.hu/wp-content/2007/04/ furko-peter-of-course-as-a-dm.doc. Accessed 20 December 2018. Furkó, Bálint Péter. 2013. The Functional Spectrum of Pragmatic Markers in Political News Interviews and Celebrity Interviews. Topics in Linguistics 11: 13–21. Furkó, Bálint Péter. 2017. Manipulative Uses of Pragmatic Markers in Political Discourse. Palgrave Communications 3: 17054. https://doi.org/10.1057/ palcomms.2017.54. Furkó, Bálint Péter, and Ágnes Abuczki. 2014. English Discourse Markers in Mediatised Political Interviews. Brno Studies in English 40 (1): 45–64. Furkó, Bálint Péter, András Kertész, and Ágnes Abuczki. 2019. Discourse Markers in Different Types of Reporting. In Indirect Reports—Perspectives in Pragmatics, Philosophy and Psychology, ed. Alessandro Capone, 243–276. Heidelberg: Springer Verlag. Greatbatch, David. 1986. Aspects of Topical Organisation in News Interviews: The Use of Agenda-Shifting Procedures by Interviewees. Media, Culture and Society 8 (4): 441–455. Gumperz, John. 1982. Discourse Strategies. Cambridge: Cambridge University Press. Heritage, John, and David Greatbatch. 1991. On the Institutional Character of Institutional Talk: the Case of News Interviews. In Talk and Social Structure: Studies in Ethnomethodology and Conversation Analysis, ed. B. Deirdre and D.H. Zimmermann, 93–137. Berkley: University of California Press. Hill, Jane H. 1999. Read My Article: Ideological Complexity and the Overdetermination of Promising in American Presidential Politics. In Regimes of Language: Ideologies, Polities, and Identities, ed. P.V. Kroskrity, 259–291. Santa Fe, NM: School of American Research Press. Ho, Victor. 2013. Strategic Use of Nouns and Pronouns in Public Discourse: The Case of the Fine-Tuning of the Medium of Instruction Policy in Hong Kong. Pragmatics 23 (1): 51–67. Holmes, Janet. 1988. Of Course: A Pragmatic Particle in New Zealand Women’s and Men’s Speech. Australian Journal of Linguistics 2: 49–74.

88

P. B. Furkó

Holmes, Janet. 1990. Hedges and Boosters in Women’s and Men’s Speech. Language & Communication 10: 185–205. Huspek, Michael. 1989. Linguistic Variability and Power: An Analysis of You Know/I Think Variation in Working-Class Speech. Journal of Pragmatics 13: 661–683. Ifantidou, Elly. 2001. Evidentials and Relevance. Amsterdam and Philadelphia: John Benjamins Publishing Company. Irimiea, Silvia. 2010. Rhetorical and Comparative Study of the Victory Speeches of Barack Obama and Mircea Geoana. JoLIE 3: 41–53. Kantara, Argyro. 2012. Adversarial Challenges and Responses in Greek Political Interviews: A Case Study. Critical Approaches to Discourse Analysis Across Disciplines 5 (2): 171–189. Khosravinik, Majid. 2010. The Representation of Refugees, Asylum Seekers and Immigrants in British Newspapers. Journal of Language and Politics 9 (1): 1–28. Lauerbach, Gerda. 2006. Discourse Representation in Political Interviews: The Construction of Identities and Relations Through Voicing and Ventriloquizing. Journal of Pragmatics 38: 196–215. Lerner, Gene H. 1994. Responsive List Construction. Language and Social Psychology 13: 20–33. Lewis, Diana. 2006. Discourse Markers in English: A Discourse-Pragmatic View. In Approaches to Discourse Particles, ed. Kerstin Fischer, 43–59. Amsterdam: Elsevier. McNair, Brian. 2011. An Introduction to Political Communication. London: Routledge. Mölder, Martin. 2010. Meanings of Democracy in Estonia: An Analysis of Focus Group Discussions. Critical Approaches to Discourse Analysis Across Disciplines 4 (1): 38–53. Németh, T. Enik˝o. 2014. Implicit Arguments at the Grammar-Pragmatics Interface: Some Methodological Considerations. Argumentum 10: 679–694. Norrby, Catrin, and Joanne Winter. 2002. Affiliation in Adolescents’ Use of Discourse Extenders. In Proceedings of the 2001 Conference of the Australian Linguistic Society. http://www.als.asn.au/proceedings/als2001/winter_norrby. pdf. Accessed 20 December 2018. Nuckolls, Janice. 1993. The Semantics of Certainty in Quechua and Its Implications for a Cultural Epistemology. Language in Society 22: 235–255. Šandová, Jana Kozubíková. 2010. Speaker’s Involvement in Political Interviews. MA thesis, Masarykova University, Brno, Czech Republic. https://is.muni. cz/th/tzbxy/Disertacni_prace.pdf. Accessed 10 January 2019.

3 Discourse Markers from a Critical Perspective …

89

Schegloff, Emmanuel. 1972. Notes on a Conversational Practice: Formulating Place. In Studies in Social Interaction, ed. David Sudnow, 75–119. New York: Free Press. Schiffrin, Deborah. 1987. Discourse Markers. Cambridge: Cambridge University Press. Schourup, Lawrence. 1999. Discourse Markers: Tutorial Overview. Lingua 107: 227–265. Simon-Vandenbergen, Anne-Marie. 1988. What Really Really Means in Casual Conversation and in Political Interviews. Linguistica Antverpiensia 22: 206– 225. Simon-Vandenbergen, Anne-Marie. 1992. The Interactional Utility of Of Course in Spoken Discourse. Occasional Papers in Systemic Linguistics 6: 213–226. Simon-Vandenbergen, Anne-Marie, P.R.R. White, and Karin Aijmer. 2007. Presupposition and ‘Taking-for-Granted’ in Mass Communicated Political Argument: An Illustration from British, Flemish and Swedish Political Colloquy. In Political Discourse in the Media, ed. Anita Fetzer and Gerda Lauerbach, 31–75. Amsterdam: John Benjamins Publishing Company. Sperber, Dan, and Deirdre Wilson. 2002. Pragmatics, Modularity and Mindreading. Mind and Language 17: 3–23. Stenström, Anna-Brita. 1990. Lexical Items Peculiar to Spoken Discourse. In The London-Lund Corpus of Spoken English: Description and Research, ed. Jan Svartvik, 137–175. Lund: Lund University Press. Stubbe, Maria, and Janet Holmes. 1995. You Know, Eh and Other Exasperating Expressions: An Analysis of Social and Stylistic Variation in the Use of Pragmatic Devices in a Sample of New Zealand English. Language & Communication 15: 63–88. Stubbs, Michael. 1996. Text and Corpus Analysis. Oxford: Blackwell. Swales, John. 1990. Genre Analysis: English in Academic and Research Settings. Cambridge: Cambridge University Press. Tranchese, Alessia, and Sole Alba Zollo. 2013. The Construction of GenderBased Violence in the British Printed and Broadcast Media. Critical Approaches to Discourse Analysis Across Disciplines 7 (1): 141–163. van Dijk, Teun A. 1993. Principles of Critical Discourse Analysis. Discourse and Society 4 (2): 249–283. van Dijk, Teun A. 2008. Discourse and Power. New York: Palgrave Macmillan. van Leeuwen, Theo. 1996. The Representation of Social Actors. In Texts and Practices: Readings in Critical Discourse Analysis, ed. Carmen Rosa, CaldasCoulthard, and Malcolm Coulthard, 32–70. London: Routledge.

90

P. B. Furkó

Wilson, John. 1993. Discourse Marking and Accounts of Violence in Northern Ireland. Text 13 (3): 455–475. Wilson, John. 2001. Political Discourse. In The Handbook of Discourse Analysis, ed. Deborah Schiffrin, Deborah Tannen, and Heidi E. Hamilton, 398–414. Boston, MA: Blackwell. Wodak, Ruth. 1989. Language, Power and Ideology. Amsterdam and Philadelphia: John Benjamins Publishing Company. Wodak, Ruth. 2007. Pragmatics and Critical Discourse Analysis: A CrossTheoretical Inquiry. Pragmatics & Cognition 15 (1): 203–234. Wodak, Ruth. 2009. The Discourse of Politics in Action: Politics as Usual. New York: Palgrave Macmillan. Wodak, Ruth, and Michael Meyer. 2009. Critical Discourse Analysis: History, Agenda, Theory, and Methodology. In Methods for Critical Discourse Analysis, ed. Ruth Wodak and Michael Meyer, 1–33. London: Sage. Zovko, Ivana. 2012. Contrastive Analysis of Discourse Markers in the Interviews with Presidents of Bosnia and Herzegovina and the U.S.A. Paper presented at 2nd International Conference on Foreign Language Teaching and Applied Linguistics (FLTAL’12), Sarajevo.

4 Discourse Markers from a Critical Perspective: A Case Study of Discourse Markers in Parliamentary Speeches

1

Introduction: Populism and Populist Discursive Strategies

Having identified the manipulative potential of discourse markers in Chapter 3, the aim of the present chapter is to take a closer look at manipulative discursive strategies through the empirical study of the manipulative potential of propositional and non-propositional lexical items (cf. Sect. 2.1 in Chapter 1) in general, and populist discursive strategies, in particular. The aim is to identify populist discursive strategies used by government and opposition parties in the course of parliamentary debates and to validate the findings of Furkó (2019), whose research focused on parliamentary speeches relating to (anti-)immigration and the immigration quota referendum of 2016. Populism is a term frequently used with reference to Hungary’s present government and especially its leader, Viktor Orbán, but rarely in connection with the centrist and/or left-liberal opposition parties currently present in Hungarian Parliament. However, Furkó (2019) found that parliamentary speeches given by government and opposition party members are equally characterized by populist discursive strategies. Moreover, the chapter will also aim to confirm that the analysis of discursive © The Author(s) 2020 P. B. Furk´o, Discourse Markers and Beyond, Postdisciplinary Studies in Discourse, https://doi.org/10.1007/978-3-030-37763-2_4

91

92

P. B. Furkó

strategies as linguistic manifestations of political goals provides a more dynamic approach to populist discourse than using observed (or often presupposed) ideological stances or political positions as bases for identifying populist and anti-populist stances or strategies. The analysis uses a mixed methodology of quantitative and qualitative approaches at the intersection of corpus linguistics, Critical Discourse Analysis and discourse marker research. In order to enhance the empirical validity and applicability of the concept of populist discursive strategy, a combination of De Cleen’s (2019) definition of populism and Wodak et al.’s (2009) definition and typology of discursive strategies will be used. As for the former, De Cleen argues for a discourse-theoretical definition of populism as a claim to represent ‘the people’ as well as a political logic centred around the identities of ‘the people’ and ‘the elite’, in which the meaning of ‘the people’ and ‘the elite’ is constructed through a down/up antagonism between ‘the people’ as a large powerless group and ‘the elite’ as “a small and illegitimately powerful group” (De Cleen 2019: 34). Discursive strategies, on the other hand, are defined as linguistic devices that underlie sociopolitical strategies, which are, in turn, defined as “more or less accurate plans for achieving a socio-political goal” (Wodak et al. 2009: 31). Discursive strategies are, thus, manifestations of sociopolitical strategies and are “systematically practised in order to assist or contradict a political action” (Küçükali 2015: 2). Wodak et al. (2009: 36ff ) provide an exhaustive typology of strategies associated with the discursive construction of national identity, which will be partially adopted for the present analysis with a view to identifying populist discursive strategies. Naturally, one has to keep in mind that there is an important distinction between the claim to represent “the people-as-nation” and the claim to represent “the people” as opposed to “the elite” (cf. De Cleen 2019: 34). Accordingly, linguistic manifestations of the macrostrategies of “positive self-representation” (Wodak et al. 2009: 39) and “negative other-representation” (ibid.: 42) will be considered as necessarily conjoined strategies in terms of opposition between “the people” and “the elite”. Although comparing the nature and frequency of nationalist and populist discursive strategies is beyond the scope of the research, the application of Wodak et al.’s taxonomy of discursive strategies will

4 Discourse Markers from a Critical Perspective …

93

highlight some of the similarities and differences between nationalist and populist discursive strategies.

2

Contextual Background

Hungary has recently attracted a considerable amount of international media attention due to its migration policy and Hungarian Prime Minister Viktor Orbán’s repeated statements that he wants to abandon liberal democracy in favour of an “illiberal state”. His major argument is that the global financial crisis in 2008 illustrated that “liberal democratic states can’t remain globally competitive” (Orbán 2014). Because of his anti-liberal, anti-elitist and nationalist rhetoric, Orbán has often been referred to as a figure emblematic of European populism in general and as a right-wing populist, in particular (source: bbc.co.uk). Korkut (2012) argues that among the many endogenic reasons for the rise of an illiberal democracy in Hungary the most prominent one is that the two “main motors of liberalization in Hungary”—the Alliance of Free Democrats (SZDSZ) and the Hungarian Socialist Party (MSZP)— have both fallen out of favour with their voters (2012: 38ff ). He goes on to state that it was the very elitism and the alienation of the people on the part of the SZDSZ that led to its downfall (ibid.), while the MSZP fell from favour largely due to a “moral crisis” (then President László Sólyom’s term1 ), following an event that is referred to in Hungarian political discourse as the leaking of the “Öszöd Speech”. The “Öszöd Speech”2 was given by socialist Prime Minister Ferenc Gyurcsány at an MSZP party conference in May 2006, in which he stated (in rather profane terms) that his party, the MSZP had been constantly telling lies about the state of the Hungarian economy to the public in the course of the pre-election campaign and did nothing but pretend to govern after the MSZP’s electoral victory. From the perspective

1 Népszabadság 2 Népszabadság

Online (2006). Online (2007).

94

P. B. Furkó

of discursive strategies, it is interesting to note that certain linguistic features of the “Öszöd Speech” such as “hazudtunk éjjel, hazudtunk nappal ” (we lied at night, we lied during the day) and “elk*rtuk”3 (we screwed up) have become popular catchphrases in anti-liberal and anti-elitist discourse. It is also interesting to consider that in his defence, Gyurcsány implicitly used the distinction between semantic/propositional and pragmatic/inferential content as an argument, stating that the latter cancelled the former in the “Öszöd Speech”, i.e. he never intended to imply that the MSZP actually lied, did nothing to govern, etc. (cf. http:// www.origo.hu/itthon/20061006gyurcsany.html). A second major argument was the mismatch between public exposure and the original target audience of his speech.4 The events above (and several others that are beyond the scope of the present analysis) resulted in a landslide victory of Fidesz in both the 2010 and 2014 parliamentary elections. Against this background, a nationwide referendum was held in Hungary on 2 October 2016, on whether to accept a future European Union quota system for resettling migrants, a measure interpreted by many in terms of Prime Minister Viktor Orbán’s fight against the EU’s migration policies. The question that was put to the public in the referendum was the following: “Do you agree that the European Union should have the power to impose the compulsory settlement of non-Hungarian citizens in Hungary without the consent of the National Assembly of Hungary?” Furkó (2019) analysed parliamentary speeches surrounding the immigration quota referendum, while the aim of the present study is to see whether or not similar discursive strategies can be observed in more recent parliamentary speeches.

3The

asterisk is used to tone down the profanity of the original lexical item. beszéd] persze nem állja ki a nyilvánosság próbáját úgy, ahogyan ott, nem a nyilvánosságnak szánt módon elhangzott”. “[the speech] and the way it was formulated was, naturally, not suitable or intended for wider publicity” (ibid.).

4 “…[a

4 Discourse Markers from a Critical Perspective …

3

95

Research Questions, Corpus and Methodology

The present chapter will analyse anti- and pro-immigration (pro- and anti-government) populist discourses surrounding the quota referendum with a view to answering two research questions: (1) What were the discursive strategies used by the government and opposition parties in the parliamentary debates relating to (anti-)immigration? (2) What differences and similarities can be observed, if any, between pro- and anti-immigration (anti- and pro-government) discursive strategies in terms of De Cleen’s (2019) notion of populism and Wodak et al.’s (2009) above-mentioned typology? (3) Can the findings of Furkó (2019) be confirmed or refuted on the basis of more recent data, i.e. have government and opposition party MPs been using similar discursive strategies since the immigration quota referendum of 2016? Furkó’s (2019) analysis was based on a 158,920-word corpus of speeches made in the Hungarian Parliament5 in which reference was made to immigration and/or the quota referendum during a period that spans 14 months: one year leading up to and two months after the referendum. With a view to replicating the methodology of the previous study, the above research questions are answered on the basis of quantitative and qualitative analyses of a 151,240-word corpus of speeches made in the Hungarian Parliament in which reference was made to immigration and/or Brussels during a period that also spanned 14 moths: between 8 May 2018 and 18 September 2019 which was the default query period used by the search engine parliament.hu used at the time the research was performed. The quantitative approach involved the use of corpus analytical tools such as concordancing (Key Word in Context, KWIC), keyness analysis 5The transcripts of parliamentary speeches are available and searchable at www.parlament.hu, the search engine provides filters for date as well as party affiliation of the speaker.

96

P. B. Furkó

(based on log-likelihood tests), cluster analysis and collocation searches. The qualitative analysis was performed through manual annotation of the keywords identified in the quantitative stage as well as through the identification of discursive strategies based on Wodak et al.’s (2009) typology. Thus, a combination of quantitative (corpus linguistics) and qualitative (critical discourse) analytical tools was adopted. Baker et al. (2008) vindicate the utility of corpus linguistic techniques in critical analyses of political discourse. However, they point out two caveats they observe as downsides of some of the relevant research: (1) many studies take a qualitative perspective and use corpora as mere repositories of examples and (2) others use corpora that are suitable for quantitative KWIC analysis in terms of descriptive statistical measures, but are too small for the measurement of (inferential ) statistical significance (Baker et al. 2008: 274ff ). Accordingly, the analysis of the above-mentioned medium-sized corpus enables us to calculate the statistical significance of several lexical items that can be associated with populist discursive strategies. With a view to replicating the methodology of the previous study, 51% of the speeches in the corpus were given by MPs from the ranks of the rightwing Fidesz, while 49% of the data were based on transcripts of speeches made by politicians affiliated to either the LMP or the MSZP, i.e. centrist or left-liberal opposition parties, respectively. With a view to comparing speeches given by MPs who are distant from each other on the political spectrum, discursive strategies used by MPs from the Jobbik (often described as a far-right populist party) as well as by independent MPs are outside the focus of the present analysis.

4

Characteristics of Parliamentary Speeches

Parliamentary speech as a sub-genre of political discourse has been researched from a variety of perspectives, both descriptive (cf. e.g. S˘aftoiu 2013) and critical (cf. e.g. Cheng 2013). Previous analyses highlight two of the genre-specific characteristics that make parliamentary speeches

4 Discourse Markers from a Critical Perspective …

97

especially amenable to the analysis of discursive strategies in general and populist discourse in particular. The first of these is the fact that, as van Dijk (1999: 29ff ) argues, parliamentary debates are doubly public. On the one hand, there is pressure on speakers to serve the interests of the people who have chosen MPs as their representatives, on the other hand, not only are parliamentary debates broadcast live, the written transcripts are available and searchable by the public at any time on the Internet, resulting in a complex constellation of temporal and interactional frames. Thus, it is safe to assume that there are at least three frames of interaction speakers have to keep in mind while preparing and delivering their speeches: the first interactional framework consists of MPs interacting in the debate itself, the second comprises participants and TV viewers, while the third frame of interaction combines features of the former two categories. It includes both classes of interactants and users of the parliamentary archives (in a different temporal frame). MPs frequently caution each other that their words will be on public record permanently, which, in the present day and age means that (traditional or online) TV viewers as well as publicists might comment on, post, tweet, etc. notable utterances or speeches and spread them through their social networks. A second characteristic feature of parliamentary speeches that increases the likelihood and frequency of the deployment of manipulative discursive strategies is a special combination of two often conflicting discursive practices: “institutionally ritualised discourse” and “individually tailored discourse” (Ilie 2010: 202). MPs need to follow procedures and observe a highly restrictive set of formal and content rules (in terms of, e.g. length of speech, topic choice, turn-sequence, type of turn, neutrality, objectivity, etc.), while maximizing the subjective “personal note” added to their speeches (S˘aftoiu 2013: 49). As will be seen in the following, propositional as well as non-propositional lexical items can both be utilized in the course of using the discursive strategies of personalization of self (animating the voice of the people) and depersonalization of the “elite”.

98

5

P. B. Furkó

Propositional Lexical Items and Indexicals Used as Manifestations of Populist Discursive Strategies

The first corpus linguistic features investigated in the discourse under investigation were the keyness6 of lexical items as well as the suffixes that can be associated with references to the identity of “the people” (Hungarians, the Hungarian nation, the Hungarian people, etc.), “us” (we, our, for us, etc.), “the will” (interest, benefit, good, etc.) as well as the antagonistic identity of “the elite”, “them” (in right-wing rhetoric it is either Brussels, i.e. the EU or immigrants) or “you” (referring to members of the Fidesz in centrist/left-wing rhetoric). It is important to keep in mind De Cleen’s (2019: 34) warning that the presence of the “us”/“them” distinction is not sufficient to constitute populist discursive strategies. A presence of a down/up, people/elite distinction needs to be complemented by a claim to represent the people-as-the-underdog (ibid.), which is why the results of quantitative research were substantiated by subsequent qualitative analysis. In the first stage of the keyness analysis, the “keyword list” feature of AntConc 3.4.4w was used and all lexical items in the two subcorpora (Fidesz sub-corpus/FSC and sub-corpus based on speeches given by members of the opposition parties/OPSC) were considered, the full list was then shortened with a focus on items that can be potentially associated with populist discourse. In the course of calculating keyness, each sub-corpus was used as a reference corpus with respect to the other. In order to calculate keyness, the log-likelihood test was used as a statistical measure. The log-likelihood value of 3.84 was considered critical (p < 0.05): if a lexical item had a value higher than the critical value, it was considered to have keyness in a particular corpus. The results are summarized in Table 1, while Table 2 summarizes Furkó’s (2019) findings based on debates surrounding the immigration quota referendum.

6 AntConc

3.4.4w was used as a concordancing programme as well as keyness analyser. For more information on keyness cf. http://www.laurenceanthony.net/software/antconc/.

842 91.4 54.2 186.6 16.6 6.6 831 98.3 90

31.6

OPSC/FSC OPSC/FSC OPSC/FSC FSC/OPSC FSC/OPSC FSC/OPSC FSC/OPSC OPSC OPSC

OPSC

Sub-corpus/reference corpus

Normalized frequency (number of tokens per hundred thousand words)

7.21

80.33 10.94

No statistical significance 55.35

No statistical significance

73.09

No statistical significance

46.04

298.43

Keyness (log-likelihood)

* character here is what is called a “wildcard” in corpus linguistics, which stands for any sequence of characters including zero, see http://www.laurenceanthony.net/software/antconc/. Hungarian is a primarily agglutinative language, thus derived as well as inflected forms need to be identified by using wildcards

a The

önök (formal alternative of “you”) maguk (~alternative of formal “you”) nézzék (~“you should take a look at…”) Brüsszel*a (for/to/in, etc. “Brussels”) nemzet* (“nation” and its derivatives) akarat (“will”) magyar* (Hungarian/Hungary) migráns (migrant) *ünket/*inket (pl. first-person possessive suffix, “our”) szeretnénk (we would like)

Relevant lexical item or morphological form

Table 1 Frequency and keyness of potentially populist uses of lexical items and suffixes in speeches given by governing and opposition parties in the period between 8 May 2018 and 18 September 2019

4 Discourse Markers from a Critical Perspective …

99

480.5 71.4 150 464.6 73.1 1087.8 102.4 735.3 484.1 64.6 195.1

26.8

OPSC/FSC FSC/OPSC FSC/OPSC FSC/OPSC FSC/OPSC FSC/OPSC FSC/OPSC FSC/OPSC FSC/OPSC FSC/OPSC

FSC/OPSC

Sub-corpus/reference corpus OPSC/FSC

Normalized frequency (number of tokens per hundred thousand words)

7.21

10.94

11.82

19.85 17.2 12.02

28.55 26.33

34.83

37.64

39.7

169.1

Keyness (log-likelihood)

Source Adopted from Furkó (2019: 348) a The * character here is what is called a “wildcard” in corpus linguistics, which stands for any sequence of characters including zero, see http://www.laurenceanthony.net/software/antconc/. Hungarian is a primarily agglutinative language, thus derived as well as inflected forms need to be identified by using wildcards

önök (formal alternative of “you”) nézzék (~“you should take a look at…”) Brüsszel*a (for/to/in, etc. “Brussels”) nemzet* (“nation” and its derivatives) akarat (“will”) magyar* (Hungarian/Hungary) migráns (migrant) nép (~“people as a nation”) emberek (~“people as individuals”) nekünk/bennünket (“us”, “for us”) *ünket/*inket (pl. first-person possessive suffix, “our”) szeretnénk (we would like)

Relevant lexical item or morphological form

Table 2 Frequency and keyness of potentially populist uses of lexical items and suffixes in speeches given by governing and opposition parties surrounding the immigration quota referendum of 2016

100 P. B. Furkó

4 Discourse Markers from a Critical Perspective …

101

If we compare the two tables, it is clear that lexical items that can be potentially associated with populist discursive strategies are more frequent and varied in the Fidesz sub-corpus than in the speeches of opposition MPs, both in speeches surrounding the immigration quota referendum (cf. Furkó 2019) and in the period between 8 May 2018 and 18 September 2019. Subsequent concordancing and manual annotation of the keywords confirmed Furkó’s (2019: 348) finding that these lexical items are used as polarization strategies contrasting the people’s will to the will of an out-group and co-occur with claims to represent the “will of the people”. For example, out of the 186 tokens of variant forms of the lemma “Brussels”, in the case of 171 Brussels is used metonymically to refer to the EU in contrast to “us” or “Hungarians” and no tokens present EU decisions in a favourable light. In the data used in previous research (Furkó 2019: 348) frequent right-context collocates of “Brussels” included “dictates” or “dictatorship” as well as “forceful”, “forced” or “demanded”. These collocates are still frequent in the present data under analysis (cf. examples 1 and 2), new collocates include “Soros” and “the Soros plan” (cf. example 3) and “immigration-friendly” as in (4): (1) Mi nem vagyunk hajlandók azt csinálni, amit Brüsszel diktál, ha az a magyaroknak nem jó. (We are not willing to do what Brussels dictates if it is not good for the Hungarian people) (FSC) (2) Brüsszel ebben a kérdésben is ránk akarja kényszeríteni saját akaratát. ( Brussels wants to force its own will on us in this issue as well.) (FSC) (3) Brüsszel és a Soros-hálózat akárhányszor is fenyegeti meg Magyarországot, a Fidesz és a KDNP szövetsége ki fog tartani a magyar nemzeti érdekek védelme mellett. (No matter how many times Brussels and the Soros-network try to threaten Hungary, the Fidesz-KDNP coalition will stand up for the interests of the Hungarian nation) (FSC) (4) Brüsszel újabb bevándorláspárti javaslatot szeretne elfogadtatni az uniós kormányf˝okkel. ( Brussels would like to push through new immigration-friendly proposals with EU state heads) (FSC)

Similarly, despite the second person grammatical form of the two keywords in the left-wing sub-corpus, most of the tokens are used as strategies to address the public, rather than the first-frame participants, in an

102

P. B. Furkó

effort to undermine the present government’s authority and ability to represent the people’s will: (5) Ez viszont azt jelenti, hogy önök a magyar családok helyett egyébként a bankok oldalára álltak ebben a kérdésben. (This means that you favour the banks’ interests over those of the Hungarian families) (OPSC) (6) Önök létrehozzák azt a világot, ahol már nincsen igazság, ahol már csak a Fidesz van, meg a fideszes káder van, meg azok az emberek vannak, akik letérdeltek önöknek. (You are creating a world where there is no truth, there’s only Fidesz and the people who bow down to you.) (OPSC)

6

Discourse Markers and Modal Adverbs Used as Manifestations of Populist Discursive Strategies

In addition to propositional lexical items and indexicals that add explicit content to populist messages, discourse markers, a set of nonpropositional lexical items were also analysed with a view to revealing implicit populist discursive strategies. In Chapter 3, I argued that discourse markers such as English you know, I mean, well, oh, etc., are highly marked for orality and the personal domain, thus, their very presence in mediatized political discourse is a sign of conversationalization and the increasingly blurred boundary between the public and personal domains. The conversationalization of political discourse is, according to Fairclough (1995), itself a manifestation of manipulative intent on the part of political actors, more specifically, a type of marketization of ideologies with a hidden consensusbuilding effect (1995: 51). I have also argued in the previous chapter that the analysis of the functional spectrum of discourse markers serves as a methodology for revealing additional strategies including polarization, suppression, recontextualizing and dramatization and provided both Hungarian and English examples.

4 Discourse Markers from a Critical Perspective …

103

Schirm (2009a) studied the Hungarian discourse markers hát (~“well”), ugye (~“surely”) and persze (~“of course”), while Schirm (2009b) described the functions of vajon (~“I wonder”) in parliamentary speeches and found the presence of these attitude-marking devices rather unexpected because of the structural and institutional constraints on parliamentary debates detailed in Sect. 4. Her conclusion is that discourse markers have a manipulative effect in adding subjectivity and a popularising rhetorical quality to a supposedly objective and neutral genre, which can be detected through the difference between default uses of discourse markers in everyday conversations and their marked functions in political discourse. Zimányi (2008: 116ff ) also analysed manipulative discursive strategies using Hungarian parliamentary speeches as data and pointed out that politicians tend to manipulate the emotions of the audience by asking face-threatening questions and giving face-threatening replies with a view to painting an unfavourable picture of a political elite. However, neither researcher has studied the use of discourse markers as manifestations of populist discursive strategies or compared speeches given by two different segments of the political spectrum, such as government and opposition MPs’ parliamentary speeches. Moreover, in the present study, all non-propositional and propositional subjectivizers (discourse markers as well as modal adverbs, cf. Sect. 4 in Chapter 8) that have keyness in either of the sub-corpora were considered, while discourse markers that do not have keyness but were identified as potential manifestations of manipulative intent in Chapter 3 of the present volume were also added to the list of resources for populist discursive strategies. Table 3 summarizes the results, while Table 4 is adopted from Furkó (2019: 351) as a point of reference. As in the case of propositional lexical items above, log-likelihood tests were used to calculate keyness in a particular sub-corpus, the log-likelihood value of 3.84 was considered critical (p < 0.05), if a lexical item had a value higher than the critical value, it was considered to have keyness in a given sub-corpus. If no statistical significance was established because of the small number of tokens, the dispersion of the lexical item was calculated by using variation coefficient (CV) values. The range of CV values is between zero and one, the lower

hát (~“well”) vajon (~“I wonder”) ugye (~“surely”) nyilvánvalóan (~“apparently”) persze (~“of course”) na (interjection) mondjuk (~“we have to admit”, “by the way”) úgymond (~“one could say”) ja (interjection) conjunctive meg (colloquial “and”) szinte (~“as if”) biztosan (~“for sure”) gyakorlatilag (~“practically”) bizony (~“for sure”) lehet (~“perhaps”) nagyjából (~“mostly”) Total

Discourse marker 31.6 13.3 3.3 11.6 23.3 11.6 33.3 6.6 8.3 33.03 16.6 3.33 28.3 36.6 170 5 435.76

48.6 (meg persze) 25.7 77.14

4.2

1.4 144.6

14.28 30 57.14

20 257.1 25.7 834.26

Normalized frequency in the FSC (tokens per 100,000 words)

64.2 10 30 24.2

Normalized frequency in the OPSC

LL LL LL LL

= = = =

4.64 17.85 14.56 127.61

No statistical significance LL = 25.64 LL = 10.1

No statistical significance LL = 75.09

No statistical significance

LL = 8.99 LL = 5.59 LL = 18.1

LL = 10.87 No statistical significance LL = 25.64 LL = 4.95

Keyness (log-likelihood)/dispersion (variation coefficient)

Table 3 Frequency and keyness of potentially manipulative discourse markers in speeches given by members of governing and opposition parties in the period of 8 May 2018 and 18 September 2019

104 P. B. Furkó

26 20 8 14 19 7 18 (+15 propos.) 3 0 16 15 1 32 9 161 4 353

45 (meg persze) 23 50 (+16 propos. mondjuk)

8

7 79

24 21 74

20 228 22 758

Normalized frequency in the FSC (tokens per 100,000 words)

49 11 42 55

Source Adopted from Furkó (2019: 351)

hát (~“well”) vajon (~“I wonder”) ugye (~“surely”) nyilvánvalóan (~“apparently”) persze (~“of course”) na (interjection) mondjuk (~“we have to admit”, “by the way”) úgymond (~“one could say”) ja (interjection) conjunctive meg (colloquial “and”) szinte (~“as if”) biztosan (~“for sure”) gyakorlatilag (~“practically”) bizony (~“for sure”) lehet (~“perhaps”) nagyjából (~“mostly”) Total

Discourse marker

Normalised frequency in the OPSC

CV = 0.38 LL = 16.28 LL = 14.89 LL = 185.29

LL = 2.71 LL = 24 LL = 19.9

N.A. LL = 50.75

CV = 0.45

LL = 12.6 LL = 10.04 LL = 18.34

LL = 9.13 CV = 0.19 LL = 27.57 LL = 39.75

Keyness (log-likelihood)/dispersion (variation coefficient)

Table 4 Frequency and keyness of potentially manipulative discourse markers in speeches given by members of governing and opposition parties

4 Discourse Markers from a Critical Perspective …

105

106

P. B. Furkó

the CV value (the closer it is to zero), the more unevenly a given lexical item is dispersed across the two (sub-)corpora. As tables show, the findings confirm those in Furkó (2019), in that with the exception of a few item, such as bizony (“for sure”) and vajon (“I wonder ”), opposition MPs use non-propositional lexical items in a wider range of contexts, yielding statistically significant differences in the frequency of these items overall. As for individual non-propositional items, similarly to earlier findings (cf. Furkó 2019: 351), 12 out of 16 lexical types yielded statistically significant differences in terms of frequency in speeches made by members of opposition parties. As in the case of propositional lexical items, discourse markers and modal adverbs were manually annotated for their discourse features, which confirmed their contribution to manipulative and/or populist discursive strategies. A random selection of 200 discourse markers (100 from each sub-corpus) was manually annotated for the mention of antagonistically opposed identities in the immediate left or right context. In the case of the FSC 79%, while in the OPSC 85% of PrMs co-occurred with explicit or implicit reference to antagonistic/opposing identities such as the EU, liberals or immigrants (in speeches made by government MPs), or the Fidesz government and its political partners (in speeches made by opposition party MPs). There are three salient discursive strategies that can be associated with the use of the non-propositional items under scrutiny: (1) evidential marking and its contribution to backgrounding, selective presentation and personal insults; (2) (semi-) interjections and their contribution to dramatization and emotional appeals; and (3) types of reporting and the contribution of discourse markers to voicing as disalignment and antagonizing. Evidential markers have been discussed in descriptive terms in Chapter 2; a critical approach to their potential manipulative effect was also detailed in Chapter 3. It was argued that in addition to marking the source and the reliability of information and knowledge (Ifantidou 2001: 3), they may also indicate how knowledge or information was acquired, e.g. through personal experience, inference or report (Nuckolls 1993: 235). Evidential markers that were identified in Tables 3 and 4 above include ugye (~“surely”), nyilvánvalóan (~“apparently”), persze

4 Discourse Markers from a Critical Perspective …

107

(~“of course”), mondjuk (~“we have to admit”, “by the way”), úgymond (~“in a way”), szinte (~“as if ”), biztosan/bizony (~“for sure”), gyakorlatilag (~“practically”), lehet (~“perhaps”) and nagyjából (~“mostly”). The default function of some of these evidential markers in spontaneous conversations (e.g. persze, biztosan, lehet ) is to express different degrees of agreement with the interlocutor’s previous utterance (e.g. “emphatic yes” function, cf. Lewis 2006), while in other contexts they often mark topic shifts, evaluations in narratives as well as the end of a list. However, in political discourse the range of functions evidential markers fulfils markedly different from their default uses. Interactional uses are rare, while most occurrences in parliamentary speeches can be grouped into two categories: (1) evidential markers are either used in anticipation of an alternative viewpoint, and/or the audience’s objections (heteroglossic uses, cf. Bakhtin 1987) or (2) evidential markers can background propositions that were previously foregrounded and/or highlight new arguments and statements by taking their truth value for granted. Both practices are used as discursive strategies by members of governing as well as opposition parties with differences in frequency (cf. Table 2). Heteroglossic uses are exemplified in 7a (FSC) and 7b (OPSC): (7a) Azt mondják, hogy akkor az embereket, a magyar embereket büntetni kell. Persze, ez nem újdonság, mert egyébként, amikor a rezsicsökkentés kapcsán a Tavares-jelentésben elkaszáltak bennünket, vagy amikor megpróbáltuk megvédeni emlékeznek, még 2013-ban itt a parlamentben a rezsicsökkentést, akkor is kik mellé álltak? (They say that then the people, the Hungarian people, should be punished. Of course, this is not a novelty, because, by the way, when we were mowed down by the Tavares report on utility cuts, or when we were trying to defend them, who in this Parliament was still in favor of a utility cut in 2013? ) (FSC) (7b) Önök a mai pénzünkb˝ol tesznek-e fel valamennyit erre a sikerre? Én nem látom. Persze, önök mondhatják, hogy mi vakok vagyunk, de o˝ szintén szólva, ebb˝ol semmit nem látok. (Do you put aside any money from today’s money for this success? I do not see that you do. Of course, you can say we’re blind, but frankly, I see nothing of it.) (OPSC)

Information structuring uses are exemplified in 8a (FSC) and 8b (OPSC):

108

P. B. Furkó

(8a) Ezért küzdünk addig, amíg lehet. Ugye, ebben nagy szerepe lesz a magyar választóknak is, vagyis hogy milyen többség˝u, bevándorláspárti vagy Európa hagyományaihoz, vallásához, kultúrájához igazodó új EP és Bizottság jön majd létre 2019 májusában. (That’s why we fight as long as we can. Of course, Hungarian voters will also have a big role to play in this regard when it comes to EP elections, whether or not the majority of the voters will be pro-immigration or pro-European in terms of traditions, religion and culture.) (FSC) (8b) Ha lehet egy óvatos kritikát megfogalmazni: ez nem a legjobban sikerült jelz˝o, kevesen fogják tudni ezt magukévá tenni, ugye azt mondták önök, hogy ez a biztonságos növekedés költségvetése. (If one can be wary of criticism: this is not the best signal, few people will be able to embrace it, [DM ugye ] you have said it is a budget for safe growth.) (OPSC)

Heteroglossic as well as backgrounding/foregrounding uses of evidential markers relate to the manipulative strategies of suppression (cf. Van Leeuwen 1996: 39) and selective presentation (cf. Pollak et al. 2011: 652). By shifting attention away from themes that are not congruent with their own beliefs and ideological aims, political actors can continue to pursue their own agenda. As 8a and 8b illustrate, suppression can be considered a populist discursive strategy in the speeches under scrutiny in that it is the antagonistic identity’s perspectives, beliefs and ideological aims that are being suppressed: in the case of the OPSC the Hungarian government’s, while in the case of the FSC the EU’s underlying intentions and aims are being suppressed and/or selectively represented. As was also mentioned in Sect. 5.1 of Chapter 3, from a cognitive pragmatic perspective (cf. Sperber and Wilson 1995: 217ff ), this strategy can be analysed as a process whereby politicians highlight elements of the mutual cognitive environment that, if appropriately combined, will lead the audience to the conclusions that the politician wants them to arrive at. Evidential markers are, therefore, inherently manipulative in parliamentary speeches, and the subtlety of the manipulation lies in the way politicians construct a “socially, politically and ideologically skewed reality” (in cognitive terms, a manipulative cognitive environment) rather than “deconstruct an existing objective reality” (Connel 1980, cited in Tranchese and Zollo 2013: 157).

4 Discourse Markers from a Critical Perspective …

109

In addition to the above functions, evidential markers often mark personal insults in the political discourse of opposition MPs, even though it is a strategy that has been mostly associated with right-wing populism (cf. Greven 2016): (9a) Nagyon nem meglep˝o az önök hozzáállása a nyugdíj dolgában. Emlékeztetni szeretném önöket, hogy a szolgálati nyugdíj ügyében bizony ez a kormány, ez a fideszes kormány volt az, amely lényegében semmibe vette a korábban a hazájukért küzd˝o katonákat, rend˝oröket, t˝uzoltókat, és szüntette meg a szolgálati nyugdíjat, változtatta szolgálati járulékká, mintegy degradálva ezeknek az embereknek a munkáját. (Your attitude towards retirement is not surprising, at all. I would like to remind you that it was this government, indeed , the Fidesz government, that basically ignored the soldiers, policemen, firefighters who had fought for their country and terminated their service pension, changing it into service contributions, people’s work.) (OPSC) (9b) Ha pedig olyan botcsinálta, Trócsányi László által kinevezett bírók dönthetnek az állam és a polgárai közötti jogvitákban, akik korábban megszokták, hogy utasításra cselekszenek, ett˝ol a beidegz˝odést˝ol bizony, pontosan tudjuk azt is, hogy nehezen fognak megszabadulni. (And if incompetent judges appointed by László Trócsányi, who have been accustomed to acting on instructions, will settle disputes between the state and its citizens, we know for sure that this reflex will be hard to get rid of.) (OPSC) (9c) Látszik, hogy nagyon hamar el akarják fogadni, azt akarják önök, hogy már januárban elkezdhessen felállni egy olyan rendszer, amely már az EP-választásra egy olyan bíróságot állít fel, amelyik gyakorlatilag a választási csalásokat el tudja tussolni. (Apparently , you want this bill passed very soon, you want to be able to start a system in January that already sets up a court for the EP election that can cover up practically every election fraud.) (OPSC)

Once again, personal insults underlie the populist discursive strategy of contrasting the people’s will with the actions of a political elite, while the larger textual context includes claims to represent the people’s will against antagonistic identity of “the elite”. The second group of discursive strategies discourse markers can be associated with is related to dramatization and emotional appeals. Schirm

110

P. B. Furkó

(2009a) observes that certain Hungarian discourse markers are incongruent with formal style as well as with objective, neutral argumentation, yet, they frequently appear in parliamentary speeches. She mentions interjections such as jaj, ja, na, no and semi-interjections (discourse markers that are, similarly to interjections, bleached of semantic content) such as hát and izé (Schirm 2009a: 170). Their incongruence is due to the fact that their use is taboo in formal contexts, when they do occur, they mark subjectivity, emotional content or verbal aggression (ibid.). Some of these items, hát, na, no and ja appear in both sub-corpora, but they are especially salient as left-wing discursive strategies: (10a) … Önök arról beszéltek, hogy megvédik ezt az országot, és itt is elmondta a képvisel˝o úr azt, hogy majd itt megállítják, meg hú, de jó lesz, meg itt aztán majd olyan biztonság lesz, hogy az csak na. (You are saying that you will protect this country and the honorable gentleman has just claimed they will stop [immigration], and it will be oh so great, people will be safe like crazy ) (OPSC) (10b) […] egyébként milyen forrás van erre. Na, ezekr˝ol olyan nagyjából semmit nem lehet tudni… (What resources are available for this? Oh well , there is not much we can know for sure in this regard.) (OPSC) (10c) … Ez nem illegális bevándorló, nem azért jön ide az az ember, mert o˝ dolgozni akar, hanem azért, mert még az utolsó családtagjait próbálja menekíteni, és megpróbál életben maradni. Na most, ezeket az embereket önök nem segítik, hanem ezeket az embereket támadják. (They are not illegal immigrants who come here for work, they come here so they can rescue the remaining family member and so that they can survive. Well then [DM na ] you’re not helping such people, you are attacking them.) (OPSC)

As these utterances are clearly intended for the second- and third-frame participants, rather than the speakers’ fellow MPs, discourse markers mark emotional appeals to the public, antagonizing the speakers’ political opponents as hypocrites (10a), non-transparent (10b) or as “the attacking force” (10c). Moreover (10a), also illustrates how reporting expressions, evidential markers and interjections co-occur and reinforce each other in creating

4 Discourse Markers from a Critical Perspective …

111

emotional appeals against the antagonistic identity’s actions and intentions. Furkó’s (2019) example data also confirms this point: (11) A kormány bármire rábökhet, és azt mondhatja, hogy ez egy kiemelt jelent˝oség˝u olimpiai ügy, és innent˝ol kezdve pontosan azt a folyamatot készítik el˝o, amit a vizes vb-n látunk, hogy majd el˝obbutóbb eszükbe jut, hogy ja, hát akkor létre kéne hozni egy törvénymódosítást, hogy a közbeszerzési szabályoktól el lehessen térni, akkor gyorsan a Market Zrt.-t meg lehet bízni meghívásos pályázaton, hogy ugyan már építsen meg ezt vagy azt, vagy Mészáros L˝orinc cégét, hogy jaj, kéne egy kisvasutat építeni, akkor gyorsan azt építse meg. (The government can point at anything and say that this is significant with a view to the Olympic Games, and from then on they can start preparing the process that we are all too familiar with, that we saw in the case of the Water Polo Games, they will soon realize oh well, [DM ja hát ] let’s implement an amendment by way of derogation from the Act on Public Procurement, then let’s quickly publish an invitation for bids especially designed for Market Ltd. so they can build something or another, or let’s design it oh [DM ja hát ] for L˝orinc Mészáros’s company, so that he can build a trenino line quickly.) (OPSC, example taken from Furkó 2019: 352)

The manipulative potential of direct and indirect speech as different types of reporting has been a widely researched area in CDA (for an overview, cf. Baker et al. 2008: 295ff ). However, as I argued in Chapter 2, a third type of reporting, referred to as voicing also needs to be considered when analysing spoken discourse. As I argued in Sect. 6 of Chapter 2 as well as in Sect. 5.3 of Chapter 3, voicing is different from both direct and indirect reports in that when using this strategy, speakers report an utterance that is probable, typical or likely to be heard or produced by a speaker other than the present one. Examples 10a and 11 illustrate this type of report of a hypothetical/imaginary utterance with a lower degree of pragmatic accountability, and the manner in which voicing the discourse of others lends itself to selective and/or biased representations of the antagonistic identity’s actions and intentions (cf. Lauerbach 2006).

112

P. B. Furkó

In parliamentary speeches, as (11) also illustrates, discourse markers, even (semi-)interjections such as ja, jaj, hát, often introduce imaginary propositions reported by the speaker, adding a dramatizing effect to the report and aligning or disaligning the animator with respect to the (hypothesized) source of the proposition. The Hungarian discourse marker úgymond (~“one could say”) is unique in its development as the pragmaticalization of a reporting verb used for voicing (cf. Dömötör 2008). As Tables 3 and 4 show, the number of tokens in the two corpora is not high enough to yield statistical significance. However, similarly to other non-propositional items, úgymond appears to be more frequent in opposition MPs’ speeches as a strategy of disalignment (12a and b) than in right-wing speeches (12c): (12a) 2010 óta úgymond süket fülekre lelünk, egyszer˝uen a kormány nem hajlandó ezt az összeget emelni, több mint 40 százalékát veszítette már reálértéken el ez a gyerekeknek járó juttatás. (Since 2010, we have been receiving deaf ears, if you like[DM úgymond ], as the government simply refuses to raise the necessary funds, and more than 40% of this allowance for children has already been lost in real terms.) (OPSC) (12b) Önök csak azokat a családokat szeretnék támogatni, akik úgymond érdemesek, vagy az önök meggy˝oz˝odése, ideológiája szerint méltók arra, hogy az állam is támogassa o˝ ket. (You are only willing to support families who are worthy in a way [DM úgymond ], who are, in your belief and according to your ideology, are worthy of being supported by the state.) (OPSC) (12c) Az Európai Parlament tavaly december 11-én szavazott err˝ol, és arra szólította fel az Európai Bizottságot, hogy idén márciusig dolgozza ki a programot, mert úgymond sürg˝os. (European Parliament voted on this issue on 11th December and called on the European Commission to develop the program by March this year, saying that it was allegedly urgent.) (FSC)

Once again, the disalignment implied by the use of úgymond underlies the populist discursive strategy of contrasting the people’s will with the actions of a political elite, the antagonistic identity being Brussels in the case of government MPs’ speeches, and the government in the case of opposition MPs’ utterances.

4 Discourse Markers from a Critical Perspective …

7

113

Conclusions, Directions for Further Research

In the present chapter, I have taken an empirical approach to populist discursive strategies used by (right-wing) government and (centrist/leftwing) opposition MPs in the wake of the migration quota referendum of 2016 and compared the findings with those in Furkó (2019) which used a corpus based on the parliamentary debates surrounding the migration quota referendum. The findings confirm those in earlier research in that when comparing populist discursive strategies used by government and opposition MPs, one can observe that both groups use very similar strategies (antagonizing, selective presentation, patronizing, polarization, dramatization and emotional appeals). Differences can be observed in terms of linguistic manifestations, i.e. the frequencies with which particular linguistic resources are used by different political actors. MPs affiliated with the Fidesz prefer explicit means of constructing “the will of the people” as well as the antagonistic identity of a “political-economic elite”, indicated by the finding that propositional lexical items that can be associated with these concepts have keyness in the speeches given by the government’s MPs both in Furkó (2019) and in the present dataset. On the other hand, anti-government voices prefer to utilize cognitively more implicit linguistic resources to realize the same discursive strategies, as is shown by the keyness of discourse markers and interjections associated with subjectivity, backgrounding, antagonizing, dramatization and voicing. As was mentioned, the use of discourse markers is a sign of the conversationalization of public discourse and is especially incongruent in parliamentary speeches, i.e. primarily monologic, pre-planned speeches delivered in highly formal contexts. It is not surprising that members of left-wing and centrist political parties resort to conversationalization (a form of equalization between the private and public spheres) to a greater degree than members of Fidesz, a conservative and right-wing party that eschews traditional values and appears to use a more normative and conservative rhetorical style as well. It was also seen that the use of discourse markers appears to underlie macro-strategies such as relativization and perpetuation in populist

114

P. B. Furkó

discourse, as well as the micro-strategies of (de-)personalization, suppression, antagonizing and dramatizing. It is unlikely that the strategic use of discourse markers is unique to populist discourse, further research needs to identify other contextual parameters that explain the predominance of implicit discursive strategies in general, and the salience of nonpropositional items in particular.

References Baker, Paul, Costas Gabrielatos, Majid KhosraviNik, Michal Krzyzanowski, Tony McEnery, and Ruth Wodak. 2008. A Useful Methodological Synergy? Combining Critical Discourse Analysis and Corpus Linguistics to Examine Discourses of Refugees and Asylum Seekers in the UK Press. Discourse & Society 19 (3): 273–306. Bakhtin, Mikhail. 1987. The Dialogic Imagination: 4 Essays. Austin: University of Texas Press. Cheng, Jennifer E. 2013. Exclusive and Inclusive Constructions of ‘Australia’ in the Australian Parliament. Critical Approaches to Discourse Analysis Across Disciplines 7 (1): 51–65. Connel, Ian. 1980. Television News and Social Contrast. In Culture, Media, Language: Working Papers in Cultural Studies, ed. Stuart Hall, Doothy Hobson, Andrew Lowe, and Paul Willis, 139–156. London: Hutchinson. De Cleen, Benjamin. 2019. The Populist Political Logic and the Discursive Construction of ‘the People’. In Imagining the Peoples of Europe: Political Discourses Across the Political Spectrum, ed. Jan Zienkowski and Ruth Breeze. Amsterdam: John Benjamins Publishing Company. Dömötör, Adrienne. 2008. Az úgy mond-tól az úgymond ig, Egy diskurzusjelöl˝o elem története az ómagyar kortól napjainkig. Magyar Nyelv˝or 132: 37–52. Fairclough, Norman. 1995. Media Discourse. London: Edward Arnold. Furkó, Bálint Péter. 2019. Populist Discursive Strategies Surrounding the Immigration Quota Referendum in Hungary. In Imagining the Peoples of Europe: Political Discourses Across the Political Spectrum, ed. Jan Zienkowski and Ruth Breeze, 343–363. Amsterdam: John Benjamins Publishing Company. Greven, Thomas. 2016. The Rise of Right-Wing Populism in Europe and the United States—A Comparative Perspective. Friedrich Ebert Stiftung, May.

4 Discourse Markers from a Critical Perspective …

115

Ifantidou, Elly. 2001. Evidentials and Relevance. Amsterdam and Philadelphia: John Benjamins Publishing Company. Ilie, Cornelia. 2010. European Parliaments Under Scrutiny: Discourse Strategies and Interaction Practices. Amsterdam and Philadelphia: John Benjamins Publishing Company. Korkut, Umut. 2012. Liberalization Challenges in Hungary Elitism, Progressivism, and Populism. Houndmills: Palgrave Macmillan. Küçükali, Can. 2015. Discursive Strategies and Political Hegemony: The Turkish Case. Amsterdam and Philadelphia: John Benjamins Publishing Company. Lauerbach, Gerda. 2006. Discourse Representation in Political Interviews: The Construction of Identities and Relations Through Voicing and Ventriloquizing. Journal of Pragmatics 38: 196–215. Lewis, Diana. 2006. Discourse Markers in English: A Discourse-Pragmatic View. In Approaches to Discourse Particles, ed. Kerstin Fischer, 43–59. Amsterdam: Elsevier. Népszabadság Online. 2006. Morális válság Magyarországon [Moral Crisis in Hungary]. http://nol.hu/archivum/archiv-417754-228454. Accessed 28 August 2017. Népszabadság Online. 2007. A teljes balatonöszödi szöveg [The Complete Transcript of the ‘Öszöd Speech’]. http://nol.hu/archivum/archiv-417593228304. Accessed 28 August 2017. Nuckolls, Janice. 1993. The Semantics of Certainty in Quechua and Its Implications for a Cultural Epistemology. Language in Society 22: 235–255. Orbán, Viktor. 2014. Beszédek [Speeches]. http://www.miniszterelnök.hu. Accessed 28 August 2017. Pollak, Senja, Roel Coesemans, Walter Daelemans, and Nada Lavraˇc. 2011. Detecting Contrast Patterns in Newspaper Articles by Combining Discourse Analysis and Text Mining. Pragmatics 21 (4): 647–683. S˘aftoiu, R˘azvan. 2013. The Discoursive Practice of Addressing in the Romanian Parliament. In The Pragmatics of Political Discourse, ed. Anita Fetzer, 47–65. Amsterdam: John Benjamins Publishing Company. Schirm, Anita. 2009a. A diskurzusjelöl˝ok a parlamenti beszédekben. In Hatékony nyelvi, idegen nyelvi és szakmai kommunikáció interkulturális környezetben, ed. Kukorelli Katalin, 168–175. Dunaújváros: Dunaújvárosi F˝oiskola. Schirm, Anita. 2009b. A parlamenti beszédek a diskurzuselemzés szemszögéb˝ol - esettanulmány a vajon diskurzusjelöl˝or˝ol. In Új módszerek az alkalmazott nyelvészeti kutatásban, ed. Gecs˝o Tamás and Sárdi Csilla, 250–255. Budapest and Székesfehérvár: Tinta Könyvkiadó.

116

P. B. Furkó

Sperber, Dan, and Deirdre Wilson. 1995. Relevance: Communication and Cognition. New York: Blackwell. Tranchese, Alessia, and Sole Alba Zollo. 2013. The Construction of GenderBased Violence in the British Printed and Broadcast Media. Critical Approaches to Discourse Analysis Across Disciplines 7 (1): 141–163. van Dijk, Teun A. 1999. ¿Qué es el análisis del discurso político? [What Is Political Discourse Analysis]. In Análisis del discurso social y político [Social and Political Discourse Analysis], ed. Teun van Dijk and Ivan Rordrigo Mendizábal, 9–103. Quito, Ecuador: Abya-Yala. van Leeuwen, Theo. 1996. The Representation of Social Actors. In Texts and Practices: Readings in Critical Discourse Analysis, ed. Carmen Rosa Caldas-Coulthard and Malcolm Coulthard, 32–70. London: Routledge. Wodak, Ruth, Rudolf de Cillia, Martin Reisigl, and Karin Liebhart. 2009. The Discursive Construction of National Identity. Edinburgh: Edinburgh University Press. Zimányi, Árpád. 2008. A politikai-közéleti nyelv érzelmi. In Az agressziókutatásról interdiszciplináris keretben, ed. Zimányi Árpád, 114–122. Eger: Líceum Kiadó.

Part II Discourse Markers Across Languages

5 The Use of Discourse Markers in Business English Textbooks: Issues in L2 Communicative Competence and Learners’ Input

1

Introduction

In previous chapters, it was seen that discourse markers are linguistic items that do not change the basic meaning of utterances but are essential for the organization and structuring of discourse and for marking the speaker’s attitudes to the proposition being expressed. From a cognitive perspective, discourse markers play an important role in regulating the processes of pragmatic inferences, in other words, in guiding hearers in their efforts to find out what is not explicitly stated but is implied by a given utterance. I have also illustrated the trend whereby both theoretical and empirical discourse marker research has been rapidly expanding mainly from a descriptive (cf. Chapters 1 and 2) perspective and have argued for the need of taking more critical stances (cf. Chapters 3 and 4). Another trend in discourse marker research has been the exponential growth of detailed analyses of a variety of items in a wide range of languages as well as across languages (for an overview, cf. Maschler and Schiffrin 2015), but it is often observed that even cross-linguistic analyses rarely concern second language acquisition (SLA) or foreign language learning (FLL) issues (cf. Vyatkina 2016). © The Author(s) 2020 P. B. Furk´o, Discourse Markers and Beyond, Postdisciplinary Studies in Discourse, https://doi.org/10.1007/978-3-030-37763-2_5

119

120

P. B. Furkó

The present chapter aims to contribute to the discourse marker/SLA/FLL research interface from the perspective of English as a second language (ESL) learners’ input, i.e. the utterances in widely used textbooks that provide illustrations and examples of discourse marker uses (cf. Furkó 2017). After the discussion of the major issues related to the concept of communicative competence as well as the role of discourse markers in shaping ESL speakers’ communicative competence, some remarks will be made about the possible sources of the difficulties that may hinder the acquisition and/or learning of discourse markers. Finally, the results of a case study will be presented, which aimed at mapping the functional spectrum of discourse markers as they are used in selected Business English (BE) textbooks.

2

Components of Communicative Competence

Developing communicative competence has been at the forefront of TEFL/TESL classes for decades. Since Hymes (1972) introduced the concept into the world of linguistics in his pioneering work, a vast number of attempts have been made by theoreticians and empirical researchers alike to create models and specify its components, some with the aim to provide theoretical bases for language teaching and testing practices (for an overview, cf. Liu 2006; Roever et al. 2014). One of the first, and up to the present-day most influential models, proposed by Canale and Swain (1980), further refined by Canale (1983), explains the concept in terms of three component competences, which are grammatical, sociolinguistic and strategic. Grammatical competence is the mastery of the language code, that is, of the phonetic, morphological, syntactic, semantic and orthographic rules. Sociolinguistic competence is conceived of as comprising two sets of rules: sociocultural and discourse rules. Sociocultural rules specify ways of using language appropriately in a given social situation, as such, they are concerned with appropriate vocabulary, style, politeness and register. Discourse rules, on

5 The Use of Discourse Markers in Business English Textbooks …

121

the other hand, determine the ways that language structures are combined to produce unified, coherent texts in different modes in speaking and writing. Strategic competence in this model consists of verbal and non-verbal communication strategies which are called into action to compensate for breakdowns in communication, due to insufficient knowledge in one or another component of communicative competence. In other words, strategically competent language users are able to get their message across successfully in the face of difficulties. A pragmatic component per se was added by Bachman, and then Bachman and Palmer in the 1990s. In their approach (Bachman and Palmer 1996), communicative competence, or to use their term, communicative language ability, consists of two broad areas: language knowledge and strategic competence. Language knowledge has two main components—organizational and pragmatic knowledge—which complement each other. Organizational knowledge consists of knowledge of linguistic units and the rules of joining them together on the sentence and text level. Pragmatic knowledge includes illocutionary and sociolinguistic competence, where illocutionary competence concerns the knowledge of how to perform acceptable language functions (Liu 2006: 7), while sociolinguistic competence equals the ability to use language appropriately to the social context. As opposed to Canale and Swain’s work, which focuses on sociolinguistics and its interaction with other components, Bachman and Palmer emphasize the central role of strategic competence, which they present as “…a set of metacognitive components, or strategies, which can be thought of as higher order executive processes that provide a cognitive management function in language use” (1996: 70). Furthermore, while the Canale and Swain model is relatively simple and accessible and, therefore, serves in many places even today as a frame of reference for TEFL/TESL classes, Bachman and Palmer’s framework is multidisciplinary and rather complex in nature. On top of that, it sets communicative competence within the context of language assessment rather than instruction. According to Liu (2006) and Bachman and Palmer (1996) were the first to develop pragmatic competence “as a competence in its own right within a learner’s overall communicative competence” (Liu 2006: 7), and

122

P. B. Furkó

this component was also adapted by the Common European Framework of Reference (CEFR for short). The model presented in the CEFR comprises three competences described in terms of knowledge and skills: linguistic, sociolinguistic and pragmatic. Linguistic competence refers to the dimensions of language as a system, for example: the use of lexical, syntactic, semantic resources, in order to form well-structured messages. Sociolinguistic competence is concerned with the sociocultural conditions of language use in so much that it determines the rules of politeness, the norms governing linguistic behaviour between different social groups (sexes, classes, etc.), and the differences in register and dialects. Pragmatic competence comprises the functional use of linguistic resources (production of language functions, speech acts), as well as the mastery of discourse cohesion and coherence, and the identification of text types and forms, irony and parody (p. 13). Unlike all the other models reviewed above, the one in the CEFR does not present strategic competence as a component of communicative language competence. Instead, strategic competence is defined as strategy use in the broadest sense: the use of communication strategies not only to compensate for the lack of knowledge in a particular area of language, but a whole range of non-compensatory communication strategies, as well. Another novelty of the CEFR model is the separation of sociolinguistic and pragmatic competences, and the designation of discourse competence as a part of pragmatic competence, rather than the other way round.

3

The Role of Discourse Markers in Shaping Learners’ Communicative Competence

No matter which model one adopts, one cannot but agree with Farhady (2005), who argues that communicative competence is so vast in domain and complex in nature that even native speakers may not fully achieve it.

5 The Use of Discourse Markers in Business English Textbooks …

123

Nevertheless, as emphasized above, developing language learners’ communicative competence has been the ultimate aim of FL/SL (L2) teaching for more than four decades, and the concept has served as the basis for working out communicative curricula. It is also important to consider the specific contribution of discourse markers to the perception and production of target language discourse. Discourse markers are used by native speakers consistently and with great precision in order to achieve a wide range of functions. They are used as frames in the interaction, for example, to make it easier for the hearer to understand how the different units of the discourse are related; discourse markers can serve to express solidarity between interlocutors and as such can function as positive politeness devices; discourse markers can be used as hedges or boosters; they mark backgrounded/foregrounded information in the case of narratives, quotes or lists; alternatively, discourse markers can be used to simply signal that the speaker needs more time to formulate a response or to process the information that was recently made available. In terms of the specific components of communicative competence, the acquisition of discourse markers contributes a great deal to the development of sociolinguistic competence, in that discourse markers play an important role in the negotiation of the (power) relationship between speaker and hearer, increasing or reducing social distance between interlocutors, diminishing authority for solidarity (cf. Lin 2010: 1173), as well as setting the appropriate level of formality or informality. With regard to pragmatic competence, discourse markers enable learners to increase or decrease the force with which propositions are expressed, selecting appropriate strategies to implement communicative acts (cf. Fraser 1990), as well as providing hearers with processing instructions in order for them to make appropriate inferences. Discourse markers’ contribution to discourse competence is in marking relationships between units of discourse or making relationships more explicit, signalling transitions, for example, from a main point to a supporting detail, in other words, “orchestrating coherent discourse” (Lin 2010: 1173) by highlighting or backgrounding information (cf. Sect. 5.1 in Chapter 3), and “signposting key points” (Lin 2010: 1180). As far as strategic competence is concerned, discourse markers often introduce the expression of

124

P. B. Furkó

communicative difficulties (finding the appropriate word/phrase), as well as appeals for the hearer’s understanding, while they might also precede paraphrasing, rephrasing and reformulation (cf. Sect. 3 in Chapter 6). It also follows from the above that non-native speakers who use discourse markers inappropriately or do not use them at all are prone to being misunderstood and to having difficulty in getting the communicative message across. However, being misunderstood is only one of the negative consequences of the absence or inappropriate use of discourse markers. Other side effects include a giveaway of the NNS’s status as an outsider to the given speech community (cf. Boxer 1993) and might also incur negative judgements on the NNS’s personality and character, as is illustrated in the following quotation: if a foreign learner says five sheeps or he goed, he can be corrected by practically every native speaker. If, on the other hand, he omits a well, the likely reaction will be that he is dogmatic, impolite, boring, awkward to talk to etc., but a native speaker cannot pinpoint an ‘error’. (Svartvik 1980: 171)

Thus, unlike grammatical mistakes or inappropriate lexical choices that reflect on the speaker’s level of proficiency, the inappropriate use or absence of discourse markers reflects on the learner’s personality, marking him/her as disfluent in the target language and out of place in the target culture in subtle ways, “ways that interlocutors (even students of language and linguistics) may not recognize” (Hellermann and Vergun 2007: 160). Consequently, even if it is unreasonable to expect students to attain native-like proficiency in all aspects of communicative competence, ESL teachers should make every effort to help them avoid negative judgements concerning their social attributes such as politeness, likeability and friendliness.

5 The Use of Discourse Markers in Business English Textbooks …

4

125

Characteristics of Discourse Markers as Sources of Learners’ Difficulties

The formal and functional characteristics discussed in Sect. 2 of Chapter 1 make discourse markers difficult to describe, teach and practice in a classroom (i.e. non-natural) environment. It was seen that discourse markers are (1) devoid of semantic content, (2) dependent on the (local and global) context and the sequence of talk for their interpretation, and (3) they are (in a purely syntactic sense) optional; thus omitting discourse markers does not change the propositional meaning of the utterance. (4) Discourse markers are also “extremely” multifunctional and, thus, display type as well as token ambiguity, as a result of which a particular occurrence of a discourse marker may simultaneously fulfil a variety of interpersonal and discourse functions. Prototypical discourse markers such as well, you know, I mean are also (5) characteristic of spoken (unplanned) discourse and oral style, and, as a result, they are often stigmatized by laypeople. Due to the first four properties, native or proficient speakers of English have difficulties describing the use and functional spectrum of even the most frequently used discourse markers such as well, you know, of course, right, etc. In addition, even if NSs find ways to describe close-to-thecore meanings/functions of particular discourse markers, their intuitions about more opaque uses are less reliable. As for the fifth property of discourse markers—i.e. the fact that they are primarily features of oral style and, as a result, are frequently stigmatized by laypeople—it is important to make a distinction between discourse-pragmatic and strategic uses of discourse markers, on the one hand, and between appropriate use and overuse, on the other. If discourse markers are used exclusively for strategic purposes, i.e. in order to try to fix or anticipate a (possible) breakdown in communication, they should, by all means, be avoided or complemented by a variety of other, more explicit metacommunicative strategies, such as asking interlocutors to provide more detail, to clarify or simply to repeat their previous utterance, as well as appealing to the hearer’s understanding, sympathy, etc. Although discourse markers are, for the most part, interchangeable when they are used for strategic functions, they are not interchangeable in their

126

P. B. Furkó

various discourse-pragmatic functions and are by no means sporadically sprinkled at various points in the discourse. As for the latter argument, it is easy to prove that even such highly stigmatized discourse markers as you know are not randomly distributed: the various functions they fulfil are determined by the host unit in which they appear as well as the discourse unit they take in their scope (cf. e.g. Fox Tree and Schrock 2002: 731).

5

Mapping the Functional Spectrum of Discourse Markers in a Corpus of Business English Textbooks

5.1

Research Process

After an overview of discourse markers’ contribution to communicative competence and the difficulties that teachers can anticipate when giving instructions about the appropriate use of discourse markers, I now turn to the results of a corpus-based study that was aimed at finding out how discourse markers are perceived and dealt with in the context of teaching Business English (henceforth, BE). The research was aimed at mapping the functional spectra of the discourse markers well and of course in a selection of BE textbooks in order to find out if learners receive enough and adequate input concerning the use of these items. Well was selected because it appears to be the most widely investigated discourse marker in the literature (cf. Schourup 2001) and, as a result, textbook writers have access to detailed descriptions of the various contexts of its use. The selection of the second item was motivated by the findings based on a contrastive study of English of course and Hungarian persze (Furkó 2011), which suggested that Hungarian learners of English might overuse of course as a result of negative pragmatic transfer. After compiling the corpus that comprises ten BE textbooks, a concordance analysis (Key Word in Context, henceforth KWIC) of well and of course was performed, content words and non-discourse marker uses were eliminated through a combination of automated and manual

5 The Use of Discourse Markers in Business English Textbooks …

127

annotation (cf. Sect. 3 in Chapter 1), and the remaining tokens were tagged according to the functional categories that were described in previous accounts of the two discourse markers (as described in the next section). Finally, the D-value of each item was calculated, i.e. the number of tokens with discourse marking functions divided by the total number of occurrences (cf. e.g. Sect. 5 in Chapter 1 or Sect. 5.4 in Chapter 3).

5.2

The Functional Spectrum of Well in Naturally Occurring Discourse

One of the earliest accounts of well as a discourse marker is found in Lakoff (1973), who observes that answers might be prefaced by well (1) if the answer is an indirect one (Lakoff 1973: 458), (2) if the information supplied with the utterance prefaced by well is only part of the answer (1973: 459), (3) in cases “where the speaker senses some sort of insufficiency in his answer” (1973: 463). Svartvik (1980) identifies the primary use of well as a “sharing device” (Svartvik 1980: 168). He agrees with the functions Lakoff (1973) identified with regard to answers, and supplements them with a number of other functions, which he subsumes under the categories of “qualifiers” and “frames”. Well as a qualifier, indicates or marks (1) agreement, positive reaction or attitude (2) reinforcement, (3) an incomplete answer to a wh-question, (4) a non-direct or qualified answer (Svartvik 1980: 173ff.). Well as a frame (1) shifts the topic focus to one of the topics which have already been under discussion, (2) introduces explanations, clarifications, etc. or (3) indicates the beginning of direct speech (Svartvik 1980: 174ff.). In addition, Svartvik claims that well also functions on the level of discourse techniques: “as floorholder, hesitator, or initiator” (1980: 176). According to Schiffrin, well is primarily a “response marker” used “when the options offered through a prior utterance for the coherence of the upcoming response are not precisely followed” (Schiffrin 1987: 127). She also states that even in cases when it is used outside of question–answer sequences, “well locates a speaker as a respondent to one

128

P. B. Furkó

level of discourse and allows a temporary release from attention to others” (ibid.). She identifies a whole range of sub-functions on the basis of her corpus compiled from sociolinguistic interviews, such as (1) well after utterances where the speaker’s assumption is inaccurate, (2) well as disagreement minimizer, especially before unexpected responses, (3) well used in reaction to an insufficiently broad wh-question, (4) well marking information search, (5) well before skip-connecting, (6) well used in narratives, especially before story abstracts, and (7) emphatic well, especially in cases where it is used to elicit information that previous questions failed to elicit or before a request for clarification. Most recent accounts (e.g. Schourup 2001; Fuller 2003; Aijmer 2013) also use Svartvik’s and Schiffrin’s description of well as a starting point. However, Norrick (2001) added that well in oral narratives has a unique function unlike in other discourse types and speech activities: well initiates and concludes narrative action, guides listeners “back to the main sequence of narrative elements following interruptions and digressions”, and is invoked by listeners “to re-orient the primary teller to the expected order of narrative presentation” (2001: 849). On the basis of the above categories and components of communicative competence, five super-functions emerged as appropriate for the tagging of well in BE textbooks (Furkó and Mónos 2013), while using corpora of naturally occurring spoken discourse as points of reference (cf. Furkó 2007): [1] Conversation management / interactional use: topic shift / topic management / skip-connecting // turn-taking/turn management. (e.g. Well, let’s move on to the next topic.) [2] Information management / monologic use: quote / shift of perspective // contrast with previous statement or question / unexpected development. (e.g. So Mom said, well my next free day’s like October fourth) [3] Well in narrative management: digressions, new developments, story preface, abstract, evaluation etc. (e.g. And, next thing you knew, it was just overcast, … well the smoke all blew in.) [4] Interpersonal (pragmatic) functions: mitigated disagreement / face management // “aggressive questions” (speaker is given no space by the question) // “considered questions”. (e.g. A: You said you never made

5 The Use of Discourse Markers in Business English Textbooks …

129

the horseshoes. – B: Well, when we put them on a horse’s hoof, they’re already made.) [5] Miscellaneous strategic functions (usually in discourse marker clusters, accompanied by attributable silence) e.g. delay, lexical search (e.g. there’s millions of ligaments, and millions of … tendons, you know, well not millions, but…)

5.3

The Functional Spectrum of Of Course in Naturally Occurring Discourse

Evidential markers have been discussed in descriptive terms in Chapter 2, a critical approach to their potential manipulative effect was also detailed in Chapter 3. It was seen that evidential markers “signal the degree of confidence, positive or negative, weakly or strongly, held by the speaker about the truth of the basic message” (Fraser 1996), and that as discourse markers they “indicate a speaker’s attitude regarding the validity of certain information, e.g. whether it is certain, probable, or untrustworthy” (Nuckolls 1993: 235). It was also clear from preceding chapters that in present-day English (both BrE and AmE) oral conversations, of course is the most frequently used evidential marker. Section 5.1 in Chapter 3 detailed the various functions of course fulfils in naturally occurring conversations as well as mediatized political interviews. For the purposes of the present research, the same annotation scheme was used for tagging the various tokens of of course as the one used for coding discourse marker uses of well (cf. Sect. 5.2 above). Thus, the various sub-functions of of course that were identified in the literature were subsumed under the following super-functions: [1] Conversation management: response marker, feedback signal, topic change signal. A: I’d like to arrange to have my car serviced. B: Yes, of course. [2] Information management (monologic uses): lists / sequences, new information, shared background knowledge.

130

P. B. Furkó

Changes in the climate can effect the living things on the planet – and of course, part of these changes are caused by man as well. [3] Story structure: mainly marks side sequences and new developments. And, of course, I didn’t end up going. [4] Interpersonal functions: personal-centre switches, persuasion, solidarity. You have to agree, of course, that … [5] Miscellaneous strategic functions (usually in discourse marker clusters and / or accompanied by silence and hesitation): self correction, lexical search, filler. And this could all, of course, I mean… end badly.

Naturally, because of the inherent multifunctionality of discourse markers in general and well and of course in particular, it was at times difficult to identify a single most salient function. In previous research (Furkó and Mónos 2013), two expert annotators were employed and inter-annotator agreement was reached in all cases, while the present study built on the previous annotation and also used annotational bias (cf. Spooren and Degand 2010), in that general interpersonal functions (e.g. politeness, hedging) were subordinated to more specific functions (e.g. narrativity, dispreferred second, etc.) in cases where both were relevant.

5.4

The Functional Spectrum of Well and Of Course in a Corpus of Business English Textbooks

The present chapter builds on the findings of previous research into the functional spectrum of discourse markers in a corpus of BE textbooks (Furkó and Mónos 2013). In the previous study, the selection of textbooks was informed by our colleague’s recommendations and, there-

5 The Use of Discourse Markers in Business English Textbooks …

131

fore, reflected on the teaching of BE in Hungarian classrooms. The corpus comprised the following textbooks (detailed information about the authors, date and place of publication is given in the Sources section). It is important to note that the dialogues and the transcripts were extracted from each textbook and instructions as well as non-dialogic texts (e.g. newspaper articles, business letters, etc.) were disregarded when determining D-values and frequencies (occurrences per ten thousand words) of the two discourse markers. 1. Viczena et al. 2003. 1000 Questions, 1000 Answers—Business English. 2. Mascull 2005. Advanced Business Vocabulary in Use. 3. MacKenzie 2006. Professional English in Use—Finance. 4. Sweeney 2003. English for Business Communication. 5. Lees-Thorne 1994. English On Business—Practical English for International Executives. 6. Cotton-Robbins 1994. Business Class: Student’s Book. 7. Ockenden 1987. Situational Dialogues. 8. Radványi-Görgényi 2003. English for Business and Finance—Haladó üzleti és pénzügyi nyelvkönyv. 9. Vándorné 1988. Fels˝ofokú angol társalgási és külkereskedelmi nyelvkönyv. 10. Powell 2003. Presenting in English—How to Give Successful Presentations. Table 1 summarizes the results gained after looking for explicit descriptions of discourse markers, indexing tokens of well and of course in terms of the 5 super-functions described above, tagging tokens of of course according to utterance position and calculating the D-value of well in the individual textbooks: In the present study, more recent (published over the last 6–8 years) and up-to-date textbooks were used with a view to comparing the results of the corpus analysis with previous findings. The corpus comprised the following BE textbooks (for detailed information cf. list of sources at the end of the chapter):

Well as a lexical item Well as a discourse marker D-value of well (%) Super-function(s) of well Of course in initial position Of course in medial/final position Super-functions of of course

6

20.7

2,4

21

5

2

14.9

1,2,4

8

4

2,4

yes/3 tasks 23

2

10

57

Descriptions/tasks no/no

1

N.A.

0

0

2

10.5

2

17

no/no

3

1,2

4

1

1,2,4,5

21.4

6

22

no/no

4

1,4

7

3

1,2,4,5

65.2

30

no/1 task 16

5

2

21

13

1,2,4,5

58.9

105

no/1 task 73

6

1,4

1

10

4

25.8

8

23

no/no

7

2

14

5

1,2,4,5

54.4

43

36

no/no

8

2,4

15

4

1,2,4,5

33.3

39

no/1 task 78

9

1,2,3,4

5

7

1,2,3,4,5

63

41

yes/8 tasks 24

10

Table 1 The use and functional spectrum of well and of course in BE textbooks adopted from Furkó and Mónos (2013: 142–143)

132 P. B. Furkó

5 The Use of Discourse Markers in Business English Textbooks …

133

1. Barrall-Barrall. 2011. Intelligent Business. Skills Book. 2. Dubicka-O’Keeffe. 2007. Market Leader —An Advanced Business English Course Book. 3. Evans. 2000. Powerhouse—An Intermediate Business English Course. 4. Gutjahr-Mahoney. 2009. Oxford Business English—English for Sales and Purchasing. 5. Lafond-Vine-Welch. 2010. Oxford Business English—English for Negotiating. 6. Miles. 2010. 250 Ways to Say It in Business English. 7. Seefer. 2011. Business English, Tenth Edition. 8. Shellabear. 2011. False Friends in Business English. 9. Taylor-Zeter. 2011. Career Paths—Business English. 10. Thomson, Kenneth. 2007. Oxford Business English—English for Meetings. Table 2 shows the results after looking for explicit descriptions of discourse markers, indexing tokens of well and of course and calculating the D-value of well in the individual textbooks: The results show that only 2 of the 10 textbooks contained instructions about the use of discourse markers in the previous study (using BE corpus 1, BEC1), while all of the (more recent) textbooks that comprise the BE corpus in the present study (BEC2) contain at least one paragraph of information about the use of discourse markers and/or some exercises that follow up on the instructions. The terminology used in the textbooks is rather heterogeneous: Seefer (2011), for example, refers to discourse markers as “mild interjections”, Evans (2000) uses the term “linking word”, while Dubicka and O’Keefe (2007) make reference to “discourse devices”. The two textbooks with explicit instructions in the earlier study contained gap-filling exercises where students were asked to complete extracts of conversations using a given list of words or expressions which include well and of course, while most of the more recent textbooks prefer multiple choice questions, where the disruptors include discourse markers as well as propositional lexical items. While the increase in explicit instructions provides cause for optimism, no significant differences can be found between the D-values and functional spectra of the two discourse markers in earlier and more recent

Descriptions/tasks Well as a lexical item Well as a discourse marker D-value of well (%) Super-function(s) of well Of course in initial position Of course in medial/final position Super-function(s) of of course

57 (20.44 pttw) 64.04

1,2,4,5

4

10

2,4

38 (31.6 pttw)

1,2,4,5

4

1

1,2,4

61.3

y/y 32

2

y/y 24

1

2

3

0

1,2,4,5

57 (91.28 pttw) 96.6

y/y 2

3

1,2,4

12

18

2,4,5

58.62

17 (9 pttw)

y/y 12

4

1,4

1

6

1,4

11 (5.33 pttw) 35.48

n/y 20

5

1,4

1

19

1,4,5

30

6 (4 pttw)

y/y 14

6

1,2,4

14

6

1,2

5.9

6 (1.44 pttw)

y/y 95

7

2

2

0

1

28.5

2 (6.26 pttw)

y/n 7

8

1

0

4

1,2,4

61.53

40 (9.89 ptww)

y/y 25

9

Table 2 The use and functional spectrum of well and of course in BE textbooks analysed in this chapter

1

0

2

1,2,5

38.1

8 (19.83 pttw)

y/y 13

10

134 P. B. Furkó

5 The Use of Discourse Markers in Business English Textbooks …

135

textbooks. In both cases, a chi-square test was performed and a statistically significant difference was found between the overall D-values of the BECs and the D-value given in the London-Lund Corpus (86%). With respect to BEC1 (44%) and the LLC, the difference is significant at χ2 = 146.5 (p < 0.01), while the D-value difference between BEC2 (47.9%) and the LLC is significant at χ2 = 120.56 (p < 0.01). The values, naturally, vary across individual textbooks, the ones that come closest to the values in natural conversation are LeesThorne (1994) and Powell (2003) in the BEC1, and Dubicka and O’Keeffe (2007) in BEC2. However, the difference between the Dvalues in these textbooks and the D-value observed in the LLC is still significant at χ2 Lees-Thorne x LLC = 35.93, χ2 Powell x LLC = 43.94, χ2 Dubicka-O’Keeffe x LLC = 40.05, p < 0.01 in all three cases. One can also observe that the higher the occurrence of well as a discourse marker the more likely it is used in three or more of its superfunctions in the both BEC1 and BEC2; however, super-function three (the use of well in narratives) is absent in all but one textbook (Powell 2003). In addition, well occurs 19.9 times per ten thousand words in the BEC2, which compares somewhat unfavourably to the LKC reference corpus, which, even though smaller in size, contains 1426 discourse marker uses, i.e. 26 tokens of well per ten thousand words. The uneven distribution is illustrated by a Juilland’s D1 of 0.57 (CV = 42.5%) across the textbooks in the BEC2, and a Juilland’s D of 0.62 (CV = 7.5%) between the components of the BEC2 and the CI corpus used as a reference corpus (cf. Sect. 4 in Chapter 1). In the case of of course, it is utterance-initial position that serves as an indicator of spoken as opposed to written style: in the reference corpus of course is used utterance initially in over 90% of its occurrences, whereas in the BEC1 corpus of course was used in medial or final position 51% of the time, which indicates that of course is used more frequently in its more formal, i.e. written functions. There is some improvement in this respect in the BEC2: in 68% of its occurrences, of course appears in

1The more unevenly an item is distributed across two or more corpora, the closer Juilland’s D is to 1, and CV is to 0%, cf. Gries (2008), and Sect. 4 in Chapter 1.

136

P. B. Furkó

utterance-initial position. However, there is a statistically significant difference between the normalized frequency of utterance-initial of course in both BEC1/2 and CI, which was used a reference corpus. The difference is significant at χ2 BEC1-CI = 186.7778 (p < 0.001) and at χ2 BEC2-CI = 53.7778 (p < 0.001). The functional distribution of of course also confirms this finding: the most salient super-function of of course in the BEC1 was in terms of backgrounding/foregrounding information, rather than marking interpersonal relations or managing conversation. These functions are somewhat more salient in the BEC2.

6

Conclusion

Discourse markers make an important contribution to efficient (business) communication, whether between native or non-native speakers of English. Discourse markers are used by NSs consistently and with great precision in order to achieve a wide range of functions, while the overuse of a limited number of discourse markers or the inappropriate use of particular discourse markers might reflect badly on a NNS’s personality, rather than their language proficiency. Because of the extreme contextdependence and non-propositional nature of discourse markers, native or proficient speakers of English usually have difficulty describing the use and functional spectrum of even the most frequently used discourse markers such as well, you know, of course, right, etc. In addition, even if they find ways to describe functions that are close to the semantic core of particular discourse markers, their intuitions about more opaque (strategic, interpersonal, conversational, etc.) uses are less reliable. As a consequence, teachers, irrespective of their level of proficiency or their status as native speakers, need to raise their meta-pragmatic awareness of the non-propositional lexical items under discussion; in other words, they need to obtain explicit instructions and explanations about the functions and uses of individual discourse markers in particular contexts. On the basis of the above analysis, one can observe that there has been an increased awareness of the importance of discourse markers in

5 The Use of Discourse Markers in Business English Textbooks …

137

BE textbooks over the last fifteen years, illustrated by the presence of explicit instructions and exercises aimed at targeting the use of discourse markers. However, if one uses BE textbooks as a corpus of utterances, the frequency, functional spectra and D-values of two of the most salient discourse markers is a far cry from the quality input students should be receiving as characteristic of naturally occurring conversations. It is a commonplace in TEFL and TESP that the availability of input as well as the salience of the relevant linguistic features in the input is of paramount importance (cf. e.g. Bardovi-Harlig 2001). Since the dialogues in BE textbooks have been found to be lacking on both counts, it appears that teachers need to compensate for the inadequate input that is provided by textbooks and they are also responsible for making the use of discourse markers more salient. Sources for Business English Corpus 1: Cotton, D.; Robbins, S. 1994. Business Class: Student’s Book. Longman: BUCL ELT Series. Lees, G.; Thorne, T. 1994. English On Business—Practical English for International Executives. New York: Longman ELT. MacKenzie, I. 2006. Professional English in Use—Finance. Cambridge: CUP. Mascull, B. 2005. Advanced Business Vocabulary in Use. Cambridge: CUP. Ockenden, M. 1987. Situational Dialogues. NY: Longman. Powell, M. 2003. Presenting in English—How to Give Successful Presentations. Boston: Thomson-Heinle. Radványi, T.; Görgényi, I. 2003. English for Business and Finance— Haladó üzleti és pénzügyi nyelvkönyv. Budapest: KJK-KERSZÖV. Sweeney, S. 2003. English for Business Communication. Cambridge: CUP. Vándorné Murvai, M. (Ed.) 1988. Fels˝ofokú angol társalgási és külkereskedelmi nyelvkönyv. Budapest: Közgazdasági és Jogi Könyvkiadó. Viczena, A.; Sz˝oke, A.; Molnár, J. 2003. 1000 Questions, 1000 Answers—Business English. Székesfehérvár: Lexika Kiadó.

138

P. B. Furkó

Sources for Business English Corpus 2: Barrall, Irene, and Nikolas Barrall. 2011. Intelligent Business. Skills Book. Harlow: Pearson Longman. Dubicka, Iwonna, and Margaret O’Keeffe. 2007. Market Leader — Advanced Business English Course Book. Harlow: Pearson Longman. Evans, David. 2000. Powerhouse—An Intermediate Business English Course. Harlow: Pearson Longman (linking words!). Gutjahr, Lothar, and Sean Mahoney. 2009. Oxford Business English— English for Sales and Purchasing. Oxford: OUP. Lafond, Charles, Sheila Vine, and Birgit Welch. 2010. Oxford Business English—English for Negotiating. Oxford: OUP. Miles, Andrew D. 2010. 250 Ways to Say It in Business English. Barcelona: English for Business (ebook). Seefer, Carolyn M. 2011. Business English, 10th ed. Mason: Cengage Learning. Shellabear, Stephanie. 2011. False Friends in Business English. Freiburg: Haufe-Lexware. Taylor, John, and Jeff Zeter. 2011. Career Paths—Business English. Newbury: Express Publishing. Thomson, Kenneth. 2007. Oxford Business English—English for Meetings. Oxford: OUP.

References Aijmer, Karin. 2013. Understanding Pragmatic Markers—A Variational Pragmatic Approach. Edinburgh: Edinburgh University Press. Bachman, Lyle F., and Adrian S. Palmer. 1996. Language Testing in Practice. Oxford: Oxford University Press. Bardovi-Harlig, Kathleen. 2001. Evaluating the Empirical Evidence: Grounds for Instruction in Pragmatics. In Pragmatics and Language Teaching, ed. Gabriele Kasper and Kenneth R. Rose, 11–32. Cambridge: Cambridge University Press. Boxer, Diana. 1993. Complaints as Positive Strategies: What the Learner Needs to Know. TESOL Quarterly 27: 277–299.

5 The Use of Discourse Markers in Business English Textbooks …

139

Canale, Michael. 1983. From Communicative Competence to Communicative Language Pedagogy. In Language and Communication, ed. Jack C. Richards and Richard W. Schmidt, 2–27. London: Longman. Canale, Michael, and Merrill Swain. 1980. Theoretical Bases of Communicative Approaches to Second Language Teaching and Testing. Applied Linguistics 1: 1–47. Dubicka, Iwonna, and Margaret O’Keeffe. 2007. Market Leader—Advanced Business English Course Book. Harlow: Pearson Longman. Evans, David. 2000. Powerhouse An Intermediate Business English Course. Harlow: Pearson Longman. Farhady, Hossein. 2005. Language Assessment: A Linguametric Perspective. Language Assessment Quarterly 2 (2): 147–164. Fox Tree, Jean E., and Josef C. Schrock. 2002. Basic Meanings of You Know and I Mean. Journal of Pragmatics 34: 727–747. Fraser, Bruce. 1990. Perspectives on Politeness. Journal of Pragmatics 14: 219– 236. Fraser, Bruce. 1996. Pragmatic Markers. Pragmatics 6: 167–190. Fuller, Janet M. 2003. The Influence of Speaker Roles on Discourse Marker Use. Journal of Pragmatics 35: 23–45. Furkó, Bálint Péter. 2007. The Status of Of Course as a Discourse Marker. HUSSE 8 Conference Proceedings. http://husse-esse.hu/wp-content/2007/04/ furko-peter-of-course-as-a-dm.doc. Accessed 20 December 2018. Furkó, Bálint Péter. 2011. A Contrastive Study of English Of Course and Hungarian Persze. In Proceedings of the HUSSE 10 Conference, Linguistics Volume, ed. Balogné Bérces Katalin, Földváry Kinga, and Mészárosné Kóris Rita, 95–106. Debrecen: Hungarian Society for the Study of English. Furkó, Bálint Péter. 2017. A pragmatikai jelöl˝okr˝ol az üzleti szaknyelv oktatásában (On Pragmatic Markers in the Context of Teaching Business English). Modern Nyelvoktatás 23 (2–3): 34–37. Furkó, Bálint Péter, and Katalin Mónos. 2013. The Teachability of Communicative Competence and the Acquisition of Pragmatic Markers—A Case Study of Some Widely-Used Business English Coursebooks. Argumentum 9: 132–148. Gries, Stefan Th. 2008. Dispersion and Adjusted Frequencies in Corpora. International Journal of Corpus Linguistics 13 (4): 403–437. Hellermann, John, and Andrea Vergun. 2007. Language Which Is Not Taught: The Discourse Marker Use of Beginning Adult Learners of English. Journal of Pragmatics 39: 157–179. Hymes, Dell. 1972. On Communicative Competence. In Sociolinguistics: Selected Readings, ed. John B. Pride and Janet Holmes, 269–293. Harmondsworth: Penguin Books.

140

P. B. Furkó

Lakoff, Robin. 1973. Questionable Answers and Answerable Questions. In Issues in Linguistics: Papers in Honor of Henry and Renée Kahane, ed. B.B. Kachru, R.B. Lees, Y. Malkiel, A. Pietrangeli, and S. Saporta, 453–467. Urbana, IL: University of Illinois Press. Lees, Gerald, and Tony Thorne. 1994. English on Business—Practical English for International Executives. New York: Longman ELT. Lin, Chia-Yen. 2010. ‘… That’s Actually Sort of You Know Trying to Get Consultants in …’: Functions and Multifunctionality of Modifiers in Academic Lectures. Journal of Pragmatics 42: 1173–1183. Liu, Jianda. 2006. Measuring Interlanguage Pragmatic Knowledge of EFL Learners. Frankfurt am Main: Peter Lang. Maschler, Yael, and Deborah Schiffrin. 2015. Discourse Markers: Language, Meaning, and Context. In Handbook of Discourse Analysis, ed. Deborah Tannen, Heidi E. Hamilton, and Deborah Schiffrin, 189–221. Malden and Oxford: Wiley Blackwell. Norrick, Neal, R. 2001. Discourse Markers in Oral Narrative. Journal of Pragmatics 33: 849–878. Nuckolls, Janice. 1993. The Semantics of Certainty in Quechua and Its Implications for a Cultural Epistemology. Language in Society 22: 235–255. Powell, Mark. 2003. Presenting in English—How to Give Successful Presentations. Boston: Thomson-Heinle. Roever, Carsten, Catriona Fraser, and Catherine Elder. 2014. Testing ESL Sociopragmatics. Frankfurt am Main: Peter Lang. Schiffrin, Deborah. 1987. Discourse Markers. Cambridge: Cambridge University Press. Schourup, Lawrence. 2001. Rethinking Well. Journal of Pragmatics 33: 1025– 1060. Seefer, Carolyn M. 2011. Business English, 10th ed. Mason: Cengage Learning. Spooren, William, and Liesbeth Degand. 2010. Coding Coherence Relations: Reliability and Validity. Corpus Linguistics and Linguistic Theory 6 (2): 241– 266. Svartvik, Jan. 1980. ‘Well’ in Conversation. In Studies in English linguistics for Randolph Quirk, ed. Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik, 167–177. London: Longman. Vyatkina, Nina. 2016. What can Multilingual Discourse-annotated Corpora do for Language Learning and Teaching? In TextLink: Structuring Discourse in Multilingual Europe Conference Handbook, ed. Bálint Péter Furkó, Csilla Dér, Liesbeth Degand, and Bonnie Webber, 21–24. Debrecen: Debrecen University Press.

6 Discourse Markers in Scripted Discourse I: Issues of (Under)Specification in the Translation of Reformulation Markers

1

Introduction

The aim of the present chapter is to explore some of the theoretical and practical issues pertaining to the translation of discourse markers in general, and reformulation markers in particular. In the first part of the chapter, categorial features of discourse markers will be revisited as sources of difficulty for translation. Next, I will look at some of the genrespecific features pertaining to the translation of scripted discourse and subtitles. After providing an overview of previous research into reformulation and reformulation markers, in the empirical part of the chapter I will present the results of a case study on the translation of the English reformulation markers I mean and actually into Hungarian. By way of concluding, I will argue that the findings are in line with the universal translation strategy of explicitation, but a wider repertoire of translation strategies is needed in order to achieve dynamic equivalence in the target text.

© The Author(s) 2020 P. B. Furk´o, Discourse Markers and Beyond, Postdisciplinary Studies in Discourse, https://doi.org/10.1007/978-3-030-37763-2_6

141

142

1.1

P. B. Furkó

Contrastive Approaches to Discourse Markers

As mentioned in the previous chapter, a recent trend in discourse marker research has been the exponential growth of detailed analyses of a variety of items in a wide range of languages as well as across languages. At the same time, contrastive studies have moved from their traditional linguistic fields of semantics and lexicology into the areas of pragmatics and discourse analysis (cf. Maschler and Schiffrin 2015). As a consequence, an increasing number of case studies are aimed at gaining insight into the functions and distributions of discourse markers across languages, thereby attempting to find translation equivalents and translation correspondences across a variety of languages. Similar to an approach to content words using translation equivalents in order to establish semantic fields, a contrastive perspective on discourse markers is aimed at mapping the functional spectrum of a given discourse marker across a wide range of bi- or multilingual contexts. As a result of the extreme multifunctionality and context-dependence of discourse markers, one can expect a larger number of correspondences between discourse markers across languages than, for example, between translation equivalents of nouns or verbs. Still, many argue (cf. e.g. Crible 2018: 7ff ) that finding translation correspondences is in many ways a more reliable method of describing individual discourse markers than providing paraphrases and glosses, or establishing co-occurrence patterns, exemplified by part I of the present volume and the majority of monolingual research. In addition, studies at the interface of discourse marker research and translation studies can provide new perspectives for both disciplines. While finding translation equivalents is a reliable way of mapping individual discourse markers’ functional spectra and can also serve as a heuristic for “establishing semantic-pragmatic fields” (Degand 2009: 174), such studies can be useful for analysing translation strategies, as well.

6 Discourse Markers in Scripted Discourse I: Issues of …

2

Preliminary Considerations

2.1

The Translation of Discourse Markers—Theory and Practice

143

The translation of discourse markers is a notoriously difficult task because of the very properties that were discussed in Sect. 2 of the introduction, i.e. non-propositionality, context-dependence, extreme multifunctionality and primarily interpersonal and/or textual functions. As a corollary of such criterial features, discourse markers do not change the basic meaning of utterances, but are essential for the organization and structuring of discourse as well as for marking the speaker’s attitudes to the proposition being expressed. Since discourse markers do not change the basic meaning of utterances, a straightforward translation strategy is to omit them in the target text. The resulting translation, in certain contexts, might not lose any of the propositional content of the source text, but will lose a variety of communicative effects, such as the naturalness of ordinary, everyday conversation or the speaker’s attitude to the words being uttered (cf. Furkó 2013: 23). The difference between a conversational exchange including and omitting a discourse marker can, in several cases, be captured along various social and functional dimensions such as the solidarity/social distance or affective/referential scales (cf. Holmes 2008), as well as along the lines of the spoken/written, planned/unplanned continua. Consequently, omitting discourse markers in the translation conspicuously and repeatedly can result in a text that does not match either the social dynamics or the intended style/register of the source text. In addition to omission, translators have a variety of options depending on the extent to which they want to (or are able to) convey the subtleties of the source text and the linguistic means by which they choose to do so. Nida’s (1964) classical distinction between formal equivalence and dynamic equivalence is especially relevant to the translation of discourse markers, since a target text that “focus[es] attention on the message” (Nida 1964: 159) will yield a radically different discourse marker equivalent from a target text where the translator has observed “the principle of equivalent effect” (ibid.).

144

P. B. Furkó

If the translator strives for formal equivalence, he or she can use the same (at times closely corresponding) discourse marker in the target text every time a particular discourse marker occurs in the source text. However, this might not turn out to be the right translation strategy, since research has shown that it is impossible to find one-to-one correspondences between discourse markers in two different languages and there are a variety of “false friends” across typologically related and unrelated languages as well (cf. e.g. Lewis 2006; Lauwers et al. 2012). Thus, even when it comes to the functional spectra of cognates such as Swedish alltså, Norwegian altså, and German also, one can find partial overlaps rather than close correspondences (cf. Aijmer and Simon-Vandenbergen 2009). Moreover, as Schirm (2015) and Polcz (2012) both point out, mirror translations can cause changes in the illocutionary force of the utterance in the target text, so that the target text might convey a weaker or stronger pragmatic force than the source text, in certain cases, the target text utterance might also fulfil a radically different pragmatic function (cf. Polcz 2012: 15ff ). Equivalent effect is equally difficult to achieve, because of the various social, stylistic, interpersonal and other functions that are simultaneously conveyed by a particular use of a discourse marker. As a result, a series of compromises will have to be made, which are, naturally, the staple of a translator’s job. However, translating pragmatic effects, i.e. non-conceptual meaning, requires greater flexibility in handling translation options, which range from lexical items such as target text discourse markers, modal particles and conjunctions, through the use of whole clauses as well as grammaticalized forms (cf. Aijmer and Simon-Vandenbergen 2003, Cuenca 2008). The present chapter will partly adopt the coding scheme for the translation of discourse markers across languages presented in Crible et al.’s (forthcoming) study. The study focuses on the underspecification of discourse markers in the translation of TED talks, starts from Spooren’s (1997) monolingual understanding of underspecified discourse markers and expands it to multilingual data. Crible et al. argue that from

6 Discourse Markers in Scripted Discourse I: Issues of …

145

a monolingual perspective, underspecification corresponds to the mismatch between a discourse marker’s semantics and its enriched interpretation in context (U1), while from a multilingual perspective, underspecification relates to a translation strategy, i.e. a mismatch between the original and the translated discourse marker, or the lack of a discourse marker in the target text. Accordingly, a distinction is made between two types of mismatch: (1) the discourse marker is omitted in the translation, tagged as ; (2) the discourse marker is translated by a more polyfunctional, less specified one, tagged as : (1) I was playing golf and my cleat got stuck. I mean, it hurt a little but I kept playing. Éppen golfoztam, amikor beakadt a stoplim. (no discourse marker, ) Kicsit fájt, de azért folytattam. (2) But, Dr. House, you’ve been so awesome. I mean, I really, totally trust you. De Dr. House, maga annyira kedves volt… és (’and’, ) én annyira megbízom magában.

In the present research, I also found it useful to add two categories of specification as well. tag has been used for a translation strategy whereby a target language discourse marker with a broad reformulative functional spectrum is used, and the tag marks a translation strategy whereby the lexical item (discourse marker or non-discourse marker, e.g. modal particle, focus structure, etc.) used in the target text has a more specific or narrower functional range than the discourse marker in the SC: (3) It’s got to be something. I mean, she didn’t come back because she likes you. Kell lenni itt valaminek. Mármint ( ~‘I mean’, ), nem jön vissza csak azért, mert kedvel téged. (4) Well, they’re doing it wrong. I mean, for Hank. Szerintem nem jól csinálják. Legalábbis (‘at least’, ) Hank.

146

2.2

P. B. Furkó

From Scripted Discourse to the Language of Subtitles

For the present study, a parallel corpus of subtitles was compiled in two languages (cf. Furkó 2014) in order to investigate the translation of reformulation markers from English into Hungarian. Several preliminary remarks are in order with regard to the study of dramatized dialogues in general and subtitles in particular. While parallel texts are ideal for optimal comparability between languages (Kenning 2010) and have been considered as flexible and accurate ways to compare discourse markers (Cettolo et al. 2012; Samy and González-Ledesma 2008), some methodological issues concerning data gathering as well as the nature/constitution of data for analysis need to be raised. Brown and Yule’s classical study, for example, differentiates between the “constructed-data” approach and the “performance-data” approach (Brown and Yule 1983: 20). One of the criteria Van Dijk sets up to define the term discourse is that it must be actual language in use, i.e. authentic and not invented language data (van Dijk 1985: 2). More recently, Chovanec distinguishes between data based on prototypical kinds of interpersonal communication such as “real-life conversations” (Chovanec 2011: 243) and data based on less prototypical types of interaction such as “television dialogue and other kinds of scripted dialogue” (ibid.). Both Chovanec (2011) and Dynel (2011) argue in favour of the legitimacy of the latter type of data in the field of language studies in general and discourse analysis in particular. Dynel (2011) observes that scripted discourse mirrors “language users’ everyday communicative patterns” (2011: 43) and invokes “an illusion of real-life conversations” (ibid.). Furkó (2010) argues that—similarly to linguists, who rely on their own intuitions in order to make grammaticality judgements— the discourse analyst who looks at dramatized dialogues, relies on “script writers’ intuitions about conversational mechanisms and communicative strategies” (Furkó 2010: 114). Moreover, since the scriptwriter’s intuitions and skills manifest themselves in the “verisimilitude of fictional interactions” (Dynel 2011: 43), the study of scripted data strikes up a balance between the “armchair approach” to linguistics (i.e. theorizing

6 Discourse Markers in Scripted Discourse I: Issues of …

147

about the implications of linguistic phenomena on the basis of constructed examples, cf. e.g. Hudson: 2001), and field methods that rely on the study of real-life conversations. If one accepts that (good) scriptwriters’ skills and intuitions are reliable with respect to conversational mechanisms and communicative strategies, one can presume that corpora based on film as well as TV scripts contain discourse markers in a wide range of communicative contexts and with an adequate range of textual and interpersonal functions. Moreover, Furkó (2011) found that co-occurrence patterns (discourse marker clusters, collocations, speech act-discourse marker pairings, etc.) observed in scripted data closely correspond to those established in research based on real-life conversations. Another reason why the analysis of scripted discourse is amenable to the study of discourse markers is that the function of a particular discourse marker is easier to identify in dramatized dialogues than in naturally occurring conversations. Scripted discourse is “thoroughly considered before being uttered” (Mattsson 2009: 2), as a result of which scriptwriters tend to make calculated use of every single linguistic (as well as paralinguistic) means of communication, including the use of discourse markers. In addition, information about the previous discourse, the social background, social status as well as some of the more dynamic personal characteristics of the speaker is more readily available than in the case of naturally occurring language data. Such information is very useful, often indispensable, in disambiguating particular uses of discourse markers. Naturally, one would expect different patterns of use with respect to functions associated with performance phenomena such as selfrepetition/self-correction, false starts and lexical search. While unplanned conversations are replete with mispronunciations, misformulations, overlaps, etc. (often marked by discourse markers, cf. Schirm 2010, 2011), such occurrences are scarce in scripted data (cf. e.g. Richardson 2010). Dramatized dialogues are “stripped of conversational chuff such as ah, y’know, well and ummm” (Lucey 1996: 168), so that “the viewer’s understanding is not impeded or even precluded” (Dynel 2011: 45). The second issue concerning the data used for the present research is genre-related. Subtitled versions of dramatized dialogues are constrained

148

P. B. Furkó

by a variety of genre-specific characteristics. Hatim and Mason (2000) group these constraints into four categories: (1) constraints that demonstrate the shift in mode from speech to writing, (2) technical constraints such as available space and the pace of the soundtrack dialogue, (3) the impossibility of back-tracking and (4) the requirement of matching the visual image (Hatim and Mason 2000: 430–431). Goris (1993) notes the levelling effect of these constraints, as a consequence of which one can expect that discourse markers will tend to occur in subtitles in the source text when they have a precisely controlled effect in the dialogue and they will tend to be omitted even in the source text subtitle when no apparent significance is attached to their use. In the target text subtitles, I expected no further levelling compared to the source text subtitles, because the former is based on the latter, rather than on the source text soundtrack, and because I assumed that once an source text discourse marker is selected for inclusion in the subtitle, the significance of its communicative effect is duly noted by the translator.

3

Previous Accounts of Reformulation

3.1

Definitions and Lists of Reformulation Markers

Before describing the different stages of the empirical part of the research, a few words are in order about reformulation in general and reformulation markers (RMs) in particular. Bordería (2014) notes that RMs “have not always received the attention they deserve in the literature on discourse markers” (Bordería 2014: 108). While there is still a scarcity of research on RMs as a subgroup of discourse markers in the relevant Anglo-Saxon literature (cf. Sect. 3.2 below), reformulation as a linguistic phenomenon in general has been described from a variety of perspectives. Halliday and Hasan concentrate on reformulation in general and distinguish between avowal and correction; the former is “an assertion of ‘the facts’ in the face of real or imaginary resistance (‘as against what you might think’)” (1976: 254), while the latter is a process whereby “one formulation is rejected in favour of another (‘as against what you have been told’)” (ibid.). Linguistic items that express avowal include in fact,

6 Discourse Markers in Scripted Discourse I: Issues of …

149

actually and as a matter of fact, while I mean and (or) rather are correction markers, according to Halliday and Hasan. Quirk et al. (1985: 1308ff ) list four types of reformulation: (1) reformulation which involves rephrasing or rewording (i.e. it is based on “linguistic knowledge”), (2) reformulation based on background knowledge (“factual knowledge”), (3) more precise formulation and (4) revision. Similarly to Halliday and Hasan, Quirk et al. take a broad view of RMs and list a variety of words, phrases or even clauses that can express one of the four types of reformulation (e.g. in other words, technically, that is, more specifically, that is to say, or … as he is usually called ). A discourse coherence approach is provided by Fraser (1999), Del SazRubio (2003) as well as Del Saz-Rubio and Fraser (2003). Fraser (1999) lists RMs such as I mean, in particular, namely, parenthetically, that is to say as a sub-class of the class of elaboration markers, which, in turn, constitute a sub-class of discourse markers. Del Saz-Rubio (2003) provides the most extensive list of RMs and defines reformulation as a “recharacterization of the message conveyed by the whole previous discourse segment S1, or one of its constituents” (2003: 211ff ). He also states that a reformulation holds between “a source discourse segment S1 […] and a reformulated segment, or S2” along with the presence of a RM that displays “the type of relationship accomplished between the two linked discourse segments” (ibid.). Del Saz-Rubio and Fraser (2003: 4ff ) list five types of reformulation: 1. paraphrase of a constituent (expressed by e.g. that is), 2. recasting of the intended meaning by the speaker (expressed by e.g. in other words), 3. revision of an implication of the prior message (expressed by e.g. worse still ), 4. correction (e.g. that is to say), 5. request for information (e.g. are you saying …). As is apparent from the list and types of RMs, a novel aspect of del Saz and Fraser’s typology is that they differentiate between self-initiated reformulation (items 1–4) and other-initiated reformulation (item 5).

150

P. B. Furkó

Table 1 provides a summary list of the RMs that is mentioned in the literature discussed above. Based on the criterial features for manual annotation described in Sect. 2 of Chapter 1 as well as on automated annotation of the above items, actually, I mean, in other words, in short and in brief are the items on the lists that both qualify as prototypical discourse markers and receive a USAS tag (“discourse bin”, cf. Sect. 4 in Chapter 1). The other items are compositional, propositional and receive a variety of nondiscourse marker USAS tags, such as (“exclusivizers”), (“deciding”) or (“grammatical bin”). Of the four, actually and I mean are frequent in the corpus (288 and 130 tokens, respectively), in other Table 1 RMs listed by author and year of publication Halliday and Hasan (1976) instead, rather, on the contrary, at least, I mean, that is, in other words, to put it another way, for instance, for example, thus, actually Quirk et al. (1985) in other words, technically, that is, or…, at that, more specifically, that is to say, or rather, I mean Fraser (1999) that is to say, for example, more precisely, I mean, in particular, namely, parenthetically Del Saz-Rubio and Fraser (2003) in other words, technically, that is, at that, more specifically, more precisely, that is to say, or rather, I mean, namely, in a word, all in all, for instance, or better, to put it simply, on second thought Del Saz-Rubio (2003) (Or) better (yet/still) In sum/to summarize/to sum up Especially More accurately/to be more accurate For example/for instance, e.g. More clearly/to be more clear/to I mean make things clear In a few words/to put it in a few More exactly/to be more exact words More precisely/to be more precise In a nutshell (in sum} More simply/in simple(r) terms/Put In a sense (in other words) more simply/to put it more simply In conclusion/to conclude More specifically/to be more specific In more technical terms Or rather In one word/In a word/in words of Particularly/in particular one syllable Say (for example/for instance) In other words/To put it in other Technically speaking words That is/that is to say/i.e. (id est) In short/in brief/in a nutshell

6 Discourse Markers in Scripted Discourse I: Issues of …

151

words, in short and in brief have been excluded because of their absence from the corpus. With the exception of (or) rather, and say, the remaining RMs listed in previous research do not even qualify as peripheral discourse markers, since they express (context-independent) propositional meaning; furthermore, they are compositional, syntactically integrated and are unmarked for oral style. Once the list of RMs has been narrowed down to I mean and actually, the next step is to look at the various pragmatic and discourse functions of the two discourse markers identified in previous research. As will be seen in the next section, mapping the functional spectrum of I mean and actually will be relevant in the course of part two of the research process, i.e. in the course of looking for correspondences between Hungarian counterparts of the two items and their discourse-pragmatic functions.

3.2

The Functional Spectrum of I Mean

In Crystal and Davy (1975: 97ff ), I mean is glossed as “in other words”, “what I have been saying amounts to the following”, “my specific meaning is that”. Its main function is in clarifying the meaning of the speaker’s immediately preceding expression, other functions include marking a restatement of the previous utterance, providing extra information and/or a fresh angle about a previous topic as well as marking a change of mind. As was mentioned earlier, according to Halliday and Hasan (1976) I mean expresses correction, more specifically, an additive conjunctive relation (expository apposition) or an adversative conjunctive relation (correction of wording). Schiffrin (1987: 296) defines I mean as a marker of the “speaker’s upcoming modification of the meaning of his/her own prior talk”. Its two main functions are “expansions of ideas” and “explanations of intention” (ibid.). Swan (1997: 159) argues that I mean introduces explanations, additional details, expressions of opinion and corrections, while it can also

152

P. B. Furkó

serve as “a general-purpose connector of ‘filler’ with little real meaning”. Further functions include “softening” and “gaining time” (ibid.). Jucker and Smith (1998) approach the function of I mean from the perspective of processing information. They differentiate between reception markers (e.g. oh, okay), which mark a reaction to information provided by another speaker and presentation markers, which accompany and modify the speaker’s own information. Information-centred presentation markers such as like modify the information itself, while addresseecentred presentation markers (I mean is included here) relate the information to the presumed knowledge state of the addressee. González (2004) concentrates on the functions of discourse markers with reference to story structure. Her research shows that the two most common functions of I mean in narratives are to mark (1) reformulation of previous information and (2) internal evaluation of the events presented in the narrative. Parallel to the functions described above, Koczogh and Furkó (2011) found that discourse marker uses of I mean include marking topic shift, elaboration, explanation, clarification, specification, false start, contrast, disagreement, conclusion, emphasis, explanation of speaker’s intention and self-correction. Moreover, Beeching (2016: 184ff ) distinguishes between non-canonical (non-pragmatic) and canonical (pragmatic) usages of I mean, the latter of which include clarification, exemplification, elaboration and reformulation. Once again, a high-granularity coding scheme has been abandoned in favour of three codes for the basic reformulative functions of I mean.

3.3

The Functional Spectrum of Actually

Lenk (1998) describes actually as a globally oriented topical marker, which performs topical actions such as closing digressions, returning to a prior topic, changing topics, introducing a new topic and inserting a subjective aside. Smith and Jucker (2000) provide a Relevance-Theoretic account of actually and propose that, on the one hand, it marks “an apparent discrepancy between propositional attitudes of conversational partners”

6 Discourse Markers in Scripted Discourse I: Issues of …

153

(2000: 207), on the other, it introduces “the presentation of a counterclaim” (2000: 208). Lumping actually together with in fact and well they state that conversationalists use actually to introduce repairs to the common ground. Biber et al. (2000: 869) classify actually as a stance adverbial expressing “epistemic-actuality”. It is important to note that, according to their research, actually occurs seven times as frequently in conversational data as in fiction, news reports and academic discourse. Oh (2000) compares the function of actually with that of in fact. Both express counter-expectation, i.e. the fact that what the speaker is about to say goes against (the hearer’s) expectations. However, while actually announces the denial of an assertion, in fact announces a strengthening of the assertion. Taglicht (2001) differentiates between a discourse marker and a nondiscourse marker use of actually. In its non-discourse marker use, actually is an integral element in the sentence structure and is used scalarly and truth-insistently (cf. Sect. 5 in the introduction). Actually as a discourse marker, on the other hand, is a marker of mild contradiction, surprise, topic change or is used as a pragmatic softener.

4

The Study

4.1

Data and Methodology

For the present study, a parallel corpus with two sub-corpora have been compiled: the Language A corpus (henceforth LAC) consists of the dialogues in the first four seasons of the TV show House (also known as House M.D. © NBC Universal Television) and constitutes a corpus of 109,909 words, while the LBC is a collection of the corresponding Hungarian translations. In the course of compiling the two subcorpora, whenever possible, I made a point of comparing the scripts and the transcripts with the official subtitles. For the LAC, I extracted the relevant dialogues from the television transcripts database (available at tvtdb.com). In order to make electronic search and concordance easier, LBC was compiled from the Hungarian subtitles of the relevant episodes;

154

P. B. Furkó

however, a keyword corpus containing the occurrences and translations of RMs was also used and was compared with the transcripts of the Hungarian-dubbed version of the TV series. The second stage of the first part of the research process involved making a concordance (Key Word in Context, KWIC) of the lexical items mean and actually, eliminating content words and non-RM uses. In the course of stage 3, I extracted discourse marker uses of I mean and actually together with their contexts, matched them with the Hungarian translations in the LBC and highlighted the keywords that served as translation equivalents in the LBC. Finally, stage 4 involved compiling the “Key Word Corpus” (henceforth KWC) by extracting the translation equivalents and their contexts in the LBC and aligning them with corresponding Key Words and Contexts in the LAC. In the course of the second part of the research process, I performed both a quantitative and a qualitative analysis of the various tokens of the two RMs in the KWC. In the first stage of this part of the research, the individual tokens were tagged according to their Hungarian translations (or the lack of a translation equivalent) as , , and (cf. Sect. 2.1 above). In the second stage, the same tokens were tagged in terms of the pragmatic functions/categories that were identified in the course of previous research into I mean and actually (as described in the previous sections). With a view to using a low-granularity coding scheme, three tags were used, each corresponding to one of three major reformulative functions described in Sect. 3.1 above. The tag (“correction”) was used in cases where I mean and actually had cancellative functions (e.g. “two, I mean, three people came”), (“clarification”) was used when the reformulation was anaphoric, i.e. recasting the previous utterance, providing its more precise formulation. Finally, (“elaborative”) was used whenever the RM was cataphoric, i.e. pointing to the following utterance, rather than correcting or rephrasing the previous one. As a last stage of this part of the process, I looked for correspondences between translation equivalents and functions as well as between functions/categories and clusters. The third part of the research process was aimed at mapping the functional spectrum of Hungarian RMs mármint and vagyis, which emerged

6 Discourse Markers in Scripted Discourse I: Issues of …

155

as the most salient Hungarian RMs and which also qualify as discourse markers. First, I made a concordance (KWIC) of mármint and vagyis in the LBC. Next (stage 2), I tagged tokens of mármint and vagyis in terms of the pragmatic functions I identified on the basis of all the occurrences in the corpus (rather than on the basis of previous research). Stage 3 involved tagging tokens of mármint and vagyis according to the source items/meaning relations in the LAC. Finally, as a final stage (stage 4) of the third part of the process, I looked for correspondences between the functions of mármint and vagyis and the English source items.

4.2

Findings

Part 2/stage 1 of the research process yielded the following results, based on 288 tokens of actually. Actually is translated as igazából (“really”, “truly”, ) in 44.09% of the cases, while actually lacks a translation equivalent () in 20.13% of its contexts of use. In third place, at 12.15%, we find the expressions pontosabban (“to be more exact”) and egészen pontosan (“to be [quite] exact”), both of which correspond to the third type of reformulation given by Quirk et al. (1985), i.e. “more precise formulation”, i.e. . S˝ot (“what is more”, ) occurs as a translation in close to 8% of the contexts (7.98%); it is interesting to consider that “what is more” occurs as an expanded, emphatic form of the additive “and” type of conjunctive relation in Halliday and Hasan (1976) and not as an element that expresses reformulation. Vagyis (~“that is”, ), another typical Hungarian RM, also occurs in close to 8% of the cases. Less frequent translations include illetve (~“or rather”), tudja mit? (“you know what?”) na jó, csak (“all right, [I was] just […ing]”), the latter two correspond to more opaque (far-from-the semantic-core) uses of actually, but, because of their more specific meanings than the source text discourse marker, they have been tagged as . As for the translations of I mean, based on 130 tokens mármint, a pragmaticalized combination of már (“yet”, “now”, “ever”) and mint (“like”), occurs as a translation equivalent in 35.38% of the cases (tagged as ). I mean is not translated () in 26.15% of the cases. Úgy

156

P. B. Furkó

értem (“I mean [it like that]”, ), an expression, which has not undergone pragmaticalization to the extent the English I mean has, occurs as a translation equivalent in 22.3% of the cases. In fourth place, we find a variety of discourse markers and discourse marker-like expressions such as (úgy) gondolom (“in my view”, “I think”, ), mondjuk (“let’s say”, “one has to admit”, ), egyébként (“by the way”, ) and például (“for example”, ). Vagyis (“that is”, ), a Hungarian discourse marker, which, as was seen earlier, corresponds to actually in several contexts, appears as a translation equivalent of I mean in 9.23% of the cases. Less frequent translations include hiszen (“since”, “surely”, ) and persze (“of course”, “naturally”, ). One can observe no statistically significant differences between the two source text discourse markers in terms of the distribution of specification/underspecification as translation strategies (χ 2 = 0.5897, p = .442538). However, there are statistically significant differences between the translation of I mean and actually in terms of the frequency of the sub-strategies of both specification (χ 2 = 63.8792, p < .00001) and underspecification (χ 2 = 4.3077, p = .03794). In the case of actually, S2 is more likely to be used as a specification strategy and U2a as an underspecification strategy, while I mean is more likely to be specified by S1 (a corresponding Hungarian RM) than S2, and equally likely to be underspecified by the strategy of omission or the use of a more underspecified discourse marker. As mentioned above, part 3/stage 3 of the research process was aimed at investigating the back translations of the most frequent translation equivalents of English I mean and actually. I found that Hungarian vagyis occurs 270 times, while mármint occurs 66 times in the LBC. In 43% of the cases, vagyis is a translation of a compositional/propositional item (i.e. non-discourse marker) with the verb mean; in 11% of the contexts, vagyis is a translation equivalent of you mean and is used for correcting the interlocutor’s utterance. In such cases, it is not used for selfcorrection, which means that it is considered a RM only in the framework of Del Saz-Rubio and Fraser (2003), where RMs of type 5 are “requests for information”. In 11% of the contexts, vagyis is a translation of a (non-propositional) use of so, i.e. it is used as an S2 strategy, while,

6 Discourse Markers in Scripted Discourse I: Issues of …

157

based on the LBC, the functional spectra of actually and vagyis overlap in 9% of the contexts. This stage of the research also revealed that there are a number of contexts where the translator felt the need to include vagyis in the Hungarian version even though there is no RM, discourse marker or any linguistic expression of reformulation, self- or other correction, etc. in the original text, thereby specifying the reformulative coherence connection between two discourse segments. The functional spectra of vagyis and I mean overlap in only 2% of the contexts. Other, less frequent contexts include utterances where we find meaning, basically, at least and that is in the original, in such cases vagyis is considered to be a U2b strategy because of its wider functional spectrum than the corresponding source text discourse markers and modal particles. Despite the small number of occurrences, it is clear from the corpus that mármint is used as an S1 strategy and serves as the translation equivalent of I mean in the majority of its functions (68%). In 20% of the cases, it is a U2b strategy expressing (underspecifying) other correction (you mean), while in a few cases it is the translation of the discourse marker like as well as meaning and meaning what?, which serve as requests for specification addressed to the interlocutor. Similarly to the case of vagyis, there are a number of contexts where mármint is used in the translation but there is no explicit expression of reformulation in the original.

4.3

Discussion, Translation Strategies and Implications

On the basis of the large number of tokens (both in the case of I mean and actually) and the small number of source text discourse markers that are not translated one can observe that compared to previous research into the use of discourse markers in subtitles (cf. e.g. Nocella 2000; Mattsson 2009) there seems to be less of a levelling effect in English subtitles and their Hungarian translations than in other language combinations. As for translation strategies, the data suggests that translators prefer specification to underspecification, and they achieve it by using one or

158

P. B. Furkó

two target text counterparts of an source text discourse marker in the majority of cases. However, as was demonstrated above, the functional spectra of target text and source text discourse markers show only a partial overlap, as in the remaining contexts translators either choose an alternative, conceptual-compositional linguistic device to convey reformulation, self-correction, elaboration, etc. (S2 or U2b), or omit the discourse marker in the target text altogether (U2a). Moreover, even in the former case translators do not avail themselves of the Hungarian language’s capacity to convey “illocutionary nuances through sequential focus” (Hervey 1998: 17), thus, they do not observe Hervey’s guidelines for the translation of illocutionary phenomena from English into Hungarian. A possible reason is that they concentrate on the textual rather than the interpersonal functions of the discourse markers under scrutiny. Thus, (5) below, where the source text discourse marker is omitted and focus is expressed through fronting the personal pronoun in the target text () is the only such example identified in the corpus: (5) Tommy: [stops playing] It’s been, like, two hours. Brandy: And we will keep waiting. I mean, you’ve heard this guy play and you know what he does, so can you just shut up, Tommy? És tovább is várni fogunk. [no discourse marker] Te is hallottad o˝t. Tudod, hogy mire képes. Szóval pofa be, Tommy! (House M. D. © NBC Universal Television)

A more frequent functional equivalence strategy can be observed in utterances when I mean is used emphatically or as a booster of the force of its host utterance, in such contexts hiszen (“since”, “surely”, ) appears as a Hungarian translation equivalent: (6) What does it matter where they are? I mean he’s in pain. You gotta do something. Mit számít, hogy hol vannak? Hiszen fájdalmai vannak. Tenniük kell valamit! (House M. D. © NBC Universal Television)

6 Discourse Markers in Scripted Discourse I: Issues of …

159

(7) Mom: Is it possible the problem isn’t his blood? It’s just psychological? I mean, he almost killed himself. Lehetséges, hogy a baj nem a vérével van? Hogy csak pszichológiai? Hiszen majdnem megölte magát. (House M. D. © NBC Universal Television)

One can also observe that the underspecification strategy of omitting the discourse markers in the target text () is especially salient in narrative discourse. The data revealed that I mean (unlike you know) is usually not translated even when it marks a new unit in the narrative structure: (8) You know he just started school when Anne died? I was a mess. Still adjusting to being a parent, much less a single parent. You know I used to put cold pancakes in his lunchbox [chuckles]. I mean, that was the only thing that I could make that he would eat. Tudja, akkor kezdte az iskolát, mikor Ann meghalt. Össze voltam zavarodva. Szül˝onek lenni is nehéz, nem hogy egy egyedülálló szül˝onek. Tudja, mindig hideg palacsintát kapott tízóraira. [no discourse marker] Ez volt az egyetlen, amit meg tudtam csinálni és meg is ette. (House M. D. © NBC Universal Television)

5

Conclusions, Limitations and Directions for Further Research

In the present chapter, I have explored the translation strategies with special reference to (under)specification, concerning the use of discourse markers in general and RMs in particular. I argued that, even though discourse markers do not normally convey conceptual meaning, their omission in the target text may result in increased processing effort, unintended conversational implicatures, misrepresented interpersonal dynamics, or, what is even worse, the absence of the naturalness of ordinary conversations. This last effect is highly undesirable when translating dramatized discourse, since dramatized discourse is, by definition,

160

P. B. Furkó

scripted discourse designed in a way that it does not sound like scripted discourse. The case study in the second part of the chapter was aimed at illustrating that there is a great deal to be gained by the cross-fertilization of discourse marker research and translation studies for both fields of inquiry. More specifically, it was argued that translators should be increasingly aware of functional equivalence strategies that are specific to the translation of illocutionary phenomena, for example, the strategy of using sequential focus to convey a particular discourse marker’s emphatic function and beware of using S1 and S2 strategies in functional domains where the source text and target text discourse markers are false friends. A limitation of the study has been its focus on RMs, a subgroup of discourse markers. It is important to consider a wider range of discourse marker types, since the different subgroups of discourse markers (reformulation markers, evidential markers, general extenders, etc.) have different degrees of communicative transparency, and, as a result, cause different degrees of difficulty for the translator. Moreover, because of the genre-specific constraints posed by the translation of subtitles (outlined in Sect. 2.2), the validity of the findings is limited to the dataset under scrutiny. Therefore, in the following chapters I turn to addressing these issues by looking at the translation of a wider range of discourse markers in both scripted and literary discourse.

References Aijmer, Karin, and Anne-Marie Simon-Vandenbergen. 2003. The Discourse Particle Well and Its Equivalents in Swedish and Dutch. Linguistics 41 (6): 1123–1161. Aijmer, Karin, and Anne-Marie Simon-Vandenbergen. 2009. Pragmatic Markers. In Handbook of Pragmatics: 2009 Installment, ed. Jan-Ola Östman and Jef Verschuere, 1–29. Amsterdam and Philadephia: John Benjamins Publishing Company. Beeching, Kate. 2016. Pragmatic Markers in British English—Meaning in Social Interaction. Cambridge: Cambridge University Press.

6 Discourse Markers in Scripted Discourse I: Issues of …

161

Biber, Douglas, Stig Johansson, Geoffrey Leech, Susan Conra, and Edward Finegan. 2000. Longman Grammar of Spoken and Written English. London: Longman. Bordería, Salvador Pons. 2014. Paths of Grammaticalization in Spanish O Sea. In Discourse and Pragmatic Markers from Latin to the Romance Languages, ed. Chiara Ghezzi and Piera Molinell, 108–135. Oxford: Oxford Scholarship Online. Brown, Gillian, and George Yule. 1983. Discourse Analysis. Cambridge: Cambridge University Press. Cettolo, Mauro, Christian Girardi, and Marcello Federico. 2012. WIT: Web Inventory of Transcribed and Translated Talks. In Proceedings of the 16th EAMT Conference, 261–268, 28–30 May 2012, Trento, Italy. Chovanec, Jan. 2011. Humour in Quasi-Conversations: Constructing Fun in Online Sports Journalism. In The Pragmatics of Humour Across Discourse Domains, ed. Marta Dynel, 243–264. Amsterdam and Philadelphia: John Benjamins Publishing Company. Crible, Ludivine. 2018. Discourse Markers and (Dis)fluency. Amsterdam and Philadelphia: John Benjamins Publishing Company. Crystal, David, and Derek Davy. 1975. Advanced Conversational English. London: Longman. Cuenca, Maria-Josep. 2008. Pragmatic Markers in Contrast: The Case of Well. Journal of Pragmatics 40: 1373–1391. Degand, Liesbeth. 2009. On Describing Polysemous Discourse Markers. What Does Translation Add to the Picture? In From Will to Well. Studies in Linguistics Offered to Anne-Marie Simon-Vandenbergen, ed. Stef Slembrouck, Miriam Taverniers, and Mieke Van Herreweghe, 173–183. Gent: Academia Press. Del Saz-Rubio, M. Milagros. 2003. An Analysis of English Discourse Markers of Reformulation. Servei de Publicacions de la Universitat de València. Del Saz-Rubio, M. Milagros, and Bruce Fraser. 2003. Reformulation in English (unpublished manuscript). http://people.bu.edu/bfraser/ Reformulation%20Marker%20Papers/deSaz%20&%20Fraser%20-% 202003%20-%20RF%20in%20English.doc. Accessed 10 January 2019. Dynel, Marta. 2011. Stranger Than Fiction?—A Few Methodological Notes on Linguistic Research in Film Discourse. Brno Studies in English 37 (1): 41–61. Fraser, Bruce. 1999. What Are Discourse Markers? Journal of Pragmatics 31: 931–952.

162

P. B. Furkó

Furkó, Bálint Péter. 2010. As Good as It Gets—Scripted Data in Discourse Analysis. Argumentum 6: 13–123. Furkó, Bálint Péter. 2011. A Contrastive Study of English Of Course and Hungarian Persze. In Proceedings of the HUSSE 10 Conference, Linguistics Volume, ed. Balogné Bérces Katalin, Földváry Kinga, and Mészárosné Kóris Rita, 95–106. Debrecen: Hungarian Society for the Study of English. Furkó, Bálint Péter. 2013. The Presence and Absence of Pragmatic Markers in Naturally-Occurring and Scripted Discourse. In Presences and Absences— Transdisciplinary Essays, ed. Katarína Labudova and Nóra Séllei, 23–38. Tyneside: Cambridge Scholars Publishing. Furkó, Bálint Péter. 2014. Perspectives on the Translation of Discourse Markers: A Case Study of the Translation of Reformulation Markers from English into Hungarian. Acta Universitatis Sapientiae, Philologica 6 (2): 181–196. González, Montserrat. 2004. Pragmatic Markers in Oral Narrative—The Case of English and Catalan. Amsterdam and Philadelphia: John Benjamins Publishing Company. Goris, Olivier. 1993. The Question of French Dubbing: Towards a Frame for Systematic Investigation. Target 5 (2): 169–190. Halliday, M.A.K., and Ruqaiya Hasan. 1976. Cohesion in English. London: Longman. Hatim, Basil, and Ian Mason. 2000. Politeness in Screen Translating. In The Translation Studies Reader, ed. Lawrence Venuti, 430–445. London and New York: Routledge. Hervey, Sándor G.J. 1998. Speech Acts and Illocutionary Function in Translation Methodology. In The Pragmatics of Translation, ed. Leo Hickey, 10–24. Clevedon: Multilingual Matters. Holmes, Janet. 2008. An Introduction to Sociolinguistics. Person: Longman. Hudson, Richard A. 2001. Sociolinguistics. Cambridge: Cambridge University Press. Jucker, Andreas H., and Sara W. Smith. 1998. And People Just You Know Like ‘Wow’: Discourse Markers as Negotiating Strategies. In Discourse Markers: Descriptions and Theory, ed. Andreas H. Jucker and Yael Ziv, 171–201. Amsterdam: John Benjamins Publishing Company. Kenning, Marie-Madeleine. 2010. What Are Parallel and Comparable Corpora and How Can We Use Them? In The Routledge Handbook of Corpus Linguistics, ed. A. O’Keeffe and M. McCarthy, 487–500. London and New York: Routledge. Koczogh, Helga, and Bálint Péter Furkó. 2011. It’s Just Like, Dude, Seriously, It’s Been a Bad Week, I Mean, Kind of Thing: Gender Differences in the

6 Discourse Markers in Scripted Discourse I: Issues of …

163

Use of the Discourse Markers ‘You Know’ and ‘I Mean’. Argumentum 7: 1–18. Lauwers, Peter, Gudrun Vanderbauwhede, and Stijn Verleyen (eds.). 2012. Pragmatic Markers and Pragmaticalization: Lessons from False Friends. Amsterdam: John Benjamins Publishing Company. Lenk, Uta. 1998. Marking Discourse Coherence. Tübingen: Narr. Lewis, Diana. 2006. Contrastive Analysis of Adversative Relational Markers, Using Comparable Corpora. In Pragmatic Markers in Contrast, ed. Karin Aijmer and Simon-Vandenbergen, 139–153. Oxford: Elsevier. Lucey, Paul. 1996. Story Sense: Writing Story and Script for Feature Films and Television. New York: McGraw-Hill. Maschler, Yael, and Deborah Schiffrin. 2015. Discourse Markers: Language, Meaning, and Context. In Handbook of Discourse Analysis, ed. Deborah Tannen, Heidi E. Hamilton, and Deborah Schiffrin, 189–221. Malden and Oxford: Wiley Blackwell. Mattsson, Jenny. 2009. The Subtitling of Discourse Particles: A Corpus-Based Study of Well, You Know, I Mean, and Like, and Their Swedish Translations in Ten American Films. Göteborgs: Göteborgs Universitet. https://gupea.ub. gu.se//handle/2077/21007. Nida, Eugene Albert. 1964. Toward a Science of Translating. Leiden: E. J. Brill. Nocella, Delia Chiaro. 2000. The British Will Use Question Tags, Won’t They? The Case of Four Weddings and a Funeral. In Tradurre il Cinema. Atti del Convegno del 29–30 Novembre 1996, ed. Chistopher Taylor, Downloaded on 17 August 2019. Trieste: Università degli Studi di Trieste. Oh, Sun-Young. 2000. Actually and In Fact in American English: A Data-Based Analysis. English Language and Linguistics 4: 243–268. Polcz, Károly. 2012. Konvencionálisan indirekt beszédaktusok az angol-magyar filmfordításban. Unpublished PhD dissertation, ELTE, Budapest. http:// doktori.btk.elte.hu/lingv/polczkaroly/diss.pdf. Accessed 27 December 2018. Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik. 1985. A Comprehensive Grammar of the English Language. London: Longman. Richardson, Kay. 2010. Television Dramatic Dialogue: A Sociolinguistic Study. New York: Oxford University Press. Samy, Doaa, and Ana González-Ledesma. 2008. Pragmatic Annotation of Discourse Markers in a Multilingual Parallel Corpus (Arabic-Spanish-English). In Proceedings of the International Conference on Language Resources and Evaluation, LREC 2008, 26 May–1 June, Marrakech, Morocco. http://www.mtarchive.info/LREC-2008-Samy.pdf. Accessed 10 January 2019.

164

P. B. Furkó

Schiffrin, Deborah. 1987. Discourse Markers. Cambridge: Cambridge University Press. Schirm, Anita. 2010. Diskurzusjelöl˝ok a parlamenti beszédekben. In Hatékony nyelvi, idegen nyelvi és szakmai kommunikáció interkulturális környezetben, ed. Kukorelli Katalin, 168–175. Dunaújváros: Dunaújvárosi F˝oiskola. Schirm, Anita. 2011. A diskurzusjelöl˝ok funkciói: a hát, az –e és a vajon elemek története és jelenkori szinkrón státusa alapján. Unpublished PhD Dissertation, Szeged. Schirm Anita. 2015. A diskurzusjelöl˝ok fordítása az elmélet és a gyakorlat oldaláról. Conference poster, MANYE, Budapest, 30 March–1 April. Smith, Sara W., and Andreas Jucker. 2000. Actuall y and Other Markers of an Apparent Discrepancy Between Propositional Attitudes of Conversational Partners. In Pragmatic Markers and Propositional Attitude, ed. Gile Andersen and Thorstein Fretheim. Amsterdam and Philadelphia: John Benjamins Publishing Company. Spooren, Wilbert. 1997. The Processing of Underspecified Coherence Relations. Discourse Processes 24 (1): 149–168. Swan, Michael. 1997. Practical English Usage. Oxford: Oxford University Press. Taglicht, Josef. 2001. Actually, There’s More to It Than Meets the Eye. English Language and Linguistics 5 (1): 1–16. van Dijk, Teun A. 1985. Handbook of Discourse Analysis. London: Academic Press.

7 Discourse Markers in Scripted Discourse II: The Representation and Translation of Irish English Stereotypes in Contemporary Cinematography

1

Introduction

In the previous chapter, I demonstrated how a subgroup of discourse markers and reformulation markers are translated in cases where the source texts are scripted dialogues. An assumption behind the analysis was that scripted conversations are written with a view to resembling naturally occurring conversations as much as possible; consequently, a translator’s major task is to produce a target text that appears as natural to the target text audience as the source text is to its respective target audience. The present chapter focuses on a more specific translation issue: the translation of dialect, or, to use Thomas’s (1983) term, pragmalect, i.e. the pragmalinguistic features of a given speech community. The discussion also takes up the issue of personality extrapolations regarding the use of discourse markers, which was introduced in Sect. 3 of Chapter 5, and argues that this issue is relevant to intralingual communication, i.e. communication between native speakers of (different varieties of ) English. The aim of the chapter is twofold: firstly, it is to examine cinematographic representations of Irish English (IrE) pragmalinguistic features with special reference to the use of IrE discourse markers in scripted (stylized) dialogues; the second aim is to contrast the stereotypes that can be © The Author(s) 2020 P. B. Furk´o, Discourse Markers and Beyond, Postdisciplinary Studies in Discourse, https://doi.org/10.1007/978-3-030-37763-2_7

165

166

P. B. Furkó

observed in cinematographic representations with the actual pragmalinguistic features of IrE based on recent findings of variational pragmatics. The structure of the chapter is as follows: after an overview of some of the ways in which the concepts of “stereotype” and “stereotyping” are defined and used in sociolinguistics and sociopragmatics, the performance of the speech acts of requests, compliment responses and thanks minimizers are provided as examples of pragmatic stereotyping in the films Intermission and The Guard. In the second part of the chapter, the use of discourse markers in present-day Irish English will be discussed and contrasted with the highly stereotypical representation of Irish English in a corpus of 12 IrE movies and series episodes. Similarly to Chapter 6, the present chapter will also discuss the translation strategies of specification and underspecification in the Hungarian translations of the source text. Regarding the performance of speech acts, the present discussion will provide a qualitative, rather than quantitative, perspective because of the small number of examples; however, the size of the test corpus will make the keyness of relevant discourse markers amenable to statistical analysis.

1.1

The Concepts of “Stereotype” and “Stereotyping”

It is often noted in sociolinguistics (cf. e.g. Jaworski and Coupland 2006: 44) that conversational partners make judgements about each other’s social background as well as personality traits on the basis of trivial linguistic choices in terms of pronunciation features, intonation, lexical or morpho-syntactic variants. Pragmalinguistic features such as the use of discourse markers or the manner in which particular speech acts are performed are even more prone to trigger social stereotyping as well as “personality extrapolations” (cf. Tannen 2006a: 460). It was argued in Sect. 3 of Chapter 5 that the inappropriate use of discourse markers might reflect negatively on ESL learners’ personality traits, rather than on their L2 proficiency. Moreover, as Thomas (1983) notes with reference to cross-cultural pragmatic failure, offensive national stereotypes such as

7 Discourse Markers in Scripted Discourse II …

167

“the abrasive German”, “the obsequious Japanese”, “the insincere American” and “the standoffish Briton” can most likely be traced back to insufficient exposure to the respective speech communities’ pragmalinguistic practices. In intracultural communication, where communication problems are less often anticipated, it is even more important to raise awareness about differences in pragmalinguistic and sociopragmatic norms. In the classical Labovian model, the term “stereotype” is used with reference to a particular type of linguistic variant. According to Labov (1972), there are three types of linguistic variants: “indicators”, “markers” and “stereotypes”. The three-way distinction reflects the social significance/stratification of a particular variant as well as the degree to which speakers are aware of the social significance of particular forms. In the case of “indicators”, members of a particular speech community attach no social significance to the use/choice of a particular form, while “markers” reflect social stratification which speakers are usually aware of. “Stereotypes”, in Labov’s classification, are variants that are readily perceived and commented on (rightly or wrongly) as salient in the speech of particular social or ethnic groups, nationalities, etc. An item of any linguistic level can be a stereotype in this respect, whether it is a phonological, suprasegmental, morphological, syntactic, lexical or pragmalinguistic feature of the particular variety it is associated with. It is important to note that the variationist paradigm was originally developed for the analysis of the social stratification of phonological features, and the methodology was later extended for the study of other formal features such as morpho-syntactic and lexical variation. Variationist studies of discourse-pragmatic features are even more recent. Moreover, as Pichler (2010, 2016) notes, studies of phonological and morphosyntactic variation and change have been “relatively homogeneous and congruent in focus and methodology” (2010: 582), while there is no conceptual uniformity in the study of discourse-pragmatic variation due to the heterogeneity of the lexical items themselves (2016: 7) and a resulting “lack of a coherent set of methodological principles” (2010: 582). The formal-functional distinction is relevant to the analysis of stereotypes in the Labovian sense, as well. The identification of formal stereotypes is straightforward: formal variants complement each other and can

168

P. B. Furkó

be substituted by each other, and they have to occur in a particular place for the sentence to be considered well-formed by native speakers. In contrast, the use of functional features is seldom obligatory, they are rarely in complementary distribution and it is usually the frequency of a particular functional variant which provides a clue to its distribution across different discourse genres, social/regional/ethnic groups, etc. If we take IrE as an example, formal stereotypes about IrE include the use of [t] in think, the use of dental, rather than alveolar [t] and [d], the use of the afterperfective, resultative-perfective, etc. (cf. Hickey 2007: 207ff ). Although there are a few pragmalinguistic features that are easy to identify, such as the use of discourse markers (discussed in Sect. 3 below), most functional stereotypes take the form of impressionistic qualifications of Irish conversational style such as Irish “indirectness”, “retinence”, “evasiveness” and “implicitness” (cf. Moynahan 1995: 178). Another basic difference between formal and functional stereotypes is the fact that while the use of formal variants is completely arbitrary, the choice of pragmalinguistic features is motivated by sociocultural norms, which, in turn, are influenced by fundamental cultural values (cf. Spencer-Oatey 2000: 43).

1.2

“Stereotype” and “Stereotyping” in Interactional Sociolinguistics and Sociopragmatics

In interactional linguistics and sociopragmatics, the term “stereotype” is used very similarly to the way it is used in social psychology, i.e. as a set of “beliefs shared by in-group members about how one’s own and other groups are characterized by certain traits and behavioural tendencies which may be positive or negative” (Bourhis and Maass 2005: 1587). Stereotyping has evaluative as well as cognitive aspects, as for the former, Quasthoff defines stereotypes as “the over-generalizing, evaluative judgements that everybody disapproves of but nobody seems to be able to do without” (1978: 2), while Gumperz posits that conversationalists rely on certain knowledge schemas, that is, “stereotypes about variant ways of speaking to categorize events, infer intent and derive expectations about

7 Discourse Markers in Scripted Discourse II …

169

what is likely to ensue” (1982: 130). In the context of cross-cultural communication, stereotyping refers to the process whereby conversational partners extend negative impressions of their interlocutor to the social group and/or speech community their interlocutor is ostensibly a member of, usually as a result of repeated experiences of miscommunication (cf. Tannen 2006b: 362). In the following, I am going to explore some of the most basic stereotypes about IrE communicative style—“implicitness”, “indirectness” and “conventional pessimism” (cf. Kallen 2005)—at the pragmalinguistic and sociopragmatic level. It is important to note that when considering functional variants such as the use of discourse markers and pragmatic strategies of performing a variety of speech acts, the distinction between stereotype in the Labovian sense and stereotype in the social psychological sense is blurred, since evaluative judgements are implicit in the attribution of discourse-pragmatic features or sociopragmatic norms to a particular group of speakers, such as speakers of IrE.

2

Examples of Discourse-Pragmatic Stereotyping

2.1

Request Strategies

Barron (2008) observes that while there is a great degree of difference between stereotypical IrE request strategies and the actual pragmalinguistic forms used in naturally occurring data, the underlying general stereotype of increased indirectness and decreased affirmativeness is confirmed by Irish sociopragmatic norms. Accordingly, a stereotypical IrE request is “you couldn’t confirm this in writing, could you?” as opposed to the stereotypically British “could you confirm this in writing, please?” (Scharf and Mac Mathúna 1998: 161 quoted in Barron 2008: 38). However, contrary to the common stereotype, Barron’s (2008) corpus-based analysis shows that IrE speakers decrease the force of the request by means of a whole range of strategies other than negative formulations and tag questions: there are significant differences in lexical upgrading as well as external downgrading. Speakers of IrE use more upgraders even in

170

P. B. Furkó

non-standard/high-solidarity contexts; moreover, they use more pre- and post-grounders and, in other words, invest more effort into the external modification (explanation, justification, pretext, etc.) of requests. In cinematographic representations of IrE, one can find elaborate requests, ample strategies of indirectness and the use of a great deal of internal and external modification, which is especially salient in informal/highsolidarity contexts (Hungarian translation is provided in brackets): (1) Reporter to interviewee: Sorry, Sally. Could you step out of the shot for a minute? Just for a minute. (Elnézést Sally. Kiállna a képb˝ol egy percre? Csak egy perc.) (Intermission) (2) Policeman to criminal: Why d’you kill McCormick, if you don’t mind me asking? (Miért ölte meg McCormickot, ha megtudhatom?) (The Guard ) (3) Boyfriend to girlfriend: Listen, do you mind if we stay in tonight? I’m knackered, do you mind? (Figyelj, nem baj ha itthon maradunk ma este? Kivagyok, nem gond?) (Intermission) (4) Mother to daughter: And listen, I wasn’t going to say it to you, but since we’re having this discussion…would you not…Would you not get yourself a bit of Immac or something?1 (És figyelj, nem akartam mondani, de mivel épp err˝ol beszélünk… nem lehetne, hogy hogy elmenj valami kozmetikushoz.) (Intermission)

As Szili (2002) proposed, the performance and indirectness of a particular speech act are a highly complex issue and depend on a variety of contextual factors. Strategy clustering in (1) might be caused by social distance, and in (2), it might be ironic attitude, while (3) and (4) include directives that are highly face threatening (cf. Verschueren 1999: 24). Thus, it is not the performance of the individual speech acts that convey a sense of Irish retinence and overt politeness, but the cumulative effect of the constant recurrence of strategy clustering, especially in informal contexts such as in (3) and (4). As for the translation of the above directives, one can observe that strategy clustering is present in the target text, while functional rather 1 In

this example, additional contextual parameters (e.g. size of the imposition) may also warrant the formulation of the request in such indirect/elaborate terms.

7 Discourse Markers in Scripted Discourse II …

171

than formal equivalence is sought in matching the individual strategies of minimizing the imposition (just a minute), negative formulation (would you not ), time distancing (could ), pre-grounders (listen), etc. Moreover, the Hungarian translation in (4) is even more indirect than the source text because of the implicitation of a culturally embedded product name (Immac). An alternative translation strategy could be explicitation through the use of a general word rather than a product name (foundation instead of Immac, e.g. nem lehetne, hogy használj alapozót vagy valami ~‘could you perhaps use some foundation or something’ ), which would enable the translator to provide a functional equivalent the general extender (or something ), another stereotypical marker of Irish indirectness, as will be seen in Sect. 3 below.

2.2

Compliment Responses (CRs)

Speakers face a dilemma when responding to compliments because of conflicting sociocultural norms: if a compliment is accepted, the complimentee might appear immodest or vain; conversely, if a compliment is not accepted, the complimentee’s opinion is in contrast to that of the complimenter (cf. Szili 2004). In terms of Leech’s (1983) theory of the politeness principle, there is a clash between the maxim of agreement and the maxim of modesty. Consequently, as Herbert (1998) concludes, the ideal response to a compliment is to accept it, if possible, in a way that does not imply self-praise and/or expresses modesty. Once again, there are interesting differences between the types of CRs one finds in IrE and in other varieties of English. Schneider (1999) found that realization types are more varied in IrE than in either AmE or EE; IrE speakers prefer non-agreement micro-CRs to agreement microCRs, and at the same time, macro- and micro-CRs are more evenly distributed. While the Irish informants employed 15 different CR strategies, the American informants used only ten. Moreover, to the conversational gambit of “Nice shirt”, the American informants clearly preferred the macro-strategy of acceptance over refusal (e.g. “Yeah, it’s nice, isn’t it?”, “I’m glad you like it”.), whereas the Irish informants’ first preference was for a non-acceptance macro-strategy (“Do you really like it?”, or “I wasn’t

172

P. B. Furkó

very happy with it”.), followed closely by CRs that fulfilled the macrofunction of acceptance. As a result, the modesty and agreement maxims are given more equal weight in IrE than, for example, in AmE, where the agreement maxim is more highly valued in CRs than the maxim of modesty. In the cinematographic representations of IrE under scrutiny, the macro-strategy of non-acceptance appears to be stereotyped, as—unlike in naturally occurring data, where modesty and agreement are given close to equal weight—there is not a single instance of a CR that implies acceptance: (5) A: “You will not get the crowd to cry hosanna until you ride into town on an ass.” Nietzsche. (“A tömeg mindaddig nem fog hozsannát kiáltani… amíg nem lovagolunk be egy szamáron a városba.” Nietzsche.) B: Yeah, that’s a good one. Good quote, nice one. (Igen, ez jó. Szép idézet.) A: Oh, f… it. (Francba.) (The Guard ) (6) A: Good shot, man. (Jó dobás volt.) B: Nice one, John. (Szép volt, John.) C: Get off my groceries, will you, you knacker? (Szállj le a zöldségekr˝ol, jó?) (Intermission) (7) A: That’s nice. Is that real leather? (Ez valódi b˝or?~’Is that real leather?’) B: Whywhat am I, f…ing cheap? (Azt hiszi sóher vagyok?~ ‘Do you think I’m cheap’) (The Guard )

In the translations, we can observe that the micro-strategy of topic avoidance is mirrored in the target text in the case of (5) and (6). In (7), however, the head act of compliment is missing from A’s turn (probably for technical reasons detailed in Sect. 2.2 of Chapter 6); thus, B’s response does not come across as a CR in the target text and does not contribute

7 Discourse Markers in Scripted Discourse II …

173

to the cumulative effect of other compliment refusals that reinforce the cultural stereotype in the source text.

2.3

Responses to Thanks (RTs)

When responding to thanks, the speaker is faced with a different type of dilemma than in the case of CRs: whether to accept the expression of gratitude and express pleasure at performing the (verbal or non-verbal) action the thanker is grateful for (cf. e.g. great pleasure, anytime), or to minimize/play down the favour or effort invested into performing the action (cf. e.g. no problem, don’t mention it ). In general terms, when the thankee chooses the former strategy, he/she addresses the thanker’s (or his/her own) positive face, thus foregrounding solidarity and/or common ground, while playing down the favour incurs a negative politeness (i.e. distance-oriented) attitude. In certain contexts, it is also appropriate to reciprocate with another token of thanks or thank you. As far as the differences between IrE and other varieties of English are concerned, the following observations can be made (cf. Schneider 2005): speakers of IrE use more varied realization types of RTs, they frequently combine conventionalized RTs and they use internal modification (modification of the head act) more frequently than AmE or EE speakers. Moreover, IrE speakers use supportive moves (external modification) more extensively than speakers of other varieties. Just as in the case of CRs, the findings suggest that IrE speakers invest more effort and creativity into minimizing thanks; moreover, formulaic and conventionalized expressions are almost invariably accompanied by non-formulaic supporting moves. In cinematographic representations of IrE, instead of elaborate RTs, supporting moves and RT combinations, we find strategies of RT avoidance, topic shift and credit shift: (8) A: (speaker of AmE) I wanna thank you for all your help. Really. It’s much appreciated. (Köszönök mindent. De tényleg, nagyon hálás vagyok.)

174

P. B. Furkó

B:

(speaker of IrE) Well … Take care of yourself, now, Wendell. ([no DM] Vigyázzon magára, Wendell.) (The Guard )

(9) A: (IrE speaker mother to her son) Thanks for taking me out, Gerry. You’re a good boy. Always a good boy. (Köszönöm, hogy elhoztál Gerry. Jó fiú vagy, mindig is az voltál.) B: (IrES) Oh, stop it. ([no DM] Hagyd abba.) (The Guard ) (10) A: The DEA believe it to have somewhere in the vicinity of $500 million worth of cocaine on board. (A Kábítószer- ellenes Ügyn˝okség szerint 500 millió dollár érték˝u kokain van a fedélzetén.) B: Er, that’s half a billion, lads. Half a billion. (Az fél milliárd, emberek.) A: Er, thank you, Inspector Stanton. Yes, that’s half a billion. – (Köszönjük Stanton nyomozó, az fél milliárd, igen.) B: [averts his eyes from speaker A] (Intermission)

The above examples illustrate that, with the exception of the microstrategies of topic shift in (8) and non-verbal RT avoidance in (10), the target text does not convey the adequate strategy clusters of RTs that are present in the source text. The discourse markers well and oh in (8) and (9), respectively, are omitted in the target text, resulting in a higher degree of directness and unintended bluntness displayed by the characters’ portrayal.

3

Stereotypes Incurred by the Use of Discourse Markers

In the previous section, discourse markers were discussed as PFMs of the micro-strategies of requests, CRs and TMs, and it was seen that their absence from the target text can lead to stronger or weaker pragmatic force or an altogether different pragmatic function. The analysis so far has been mainly qualitative, and suggestions for translators were based on stipulated IrE stereotypes and a comparison with research based on present-day, non-scripted IrE. In the present section, I focus on the

7 Discourse Markers in Scripted Discourse II …

175

stigmatization and stylizing effects discourse markers can be associated with and complement the previous analysis by quantitative data. Non-linguists, including language mavens (cf. e.g. Cameron 1995: 50), often associate discourse markers such as I mean, you know, well and like with spoken discourse and oral style, at best, and with the language of women, adolescents and the illiterate, uneducated, underprivileged, etc., at worst. A variety of discourse markers have also been stereotyped as markers of superficiality as well as lack of intelligence, cf. e.g. like as a marker of “Valley Girl speech” (Blythe et al. 1990: 224). In Britain, certain discourse markers such as like and you know have become stereotypes of the (supposedly inferior) dialect of American English, as is exemplified by the following email joke, in which the writer gives advice to speakers of AmE: Generally, you should raise your vocabulary to acceptable levels. Look up ‘vocabulary’. Using the same twenty-seven words interspersed with filler noises such as ‘like’ and ‘you know’ is an unacceptable and inefficient form of communication. Look up ‘interspersed’. (example taken from Fuller 2003: 370)

It is well known that people’s attitudes to features of any variety of language reflect their attitudes to the speakers of that variety (cf. Honey 2007: 70); thus, negative stereotyping of discourse marker uses of like, I mean and you know reflects negative attitudes to social groups and social situations such items are (often wrongly) associated with. What is more, as is clear from the email joke above, the stereotypical function assigned to discourse markers is strategic rather than interpersonal or discourse organizational; in other words, some native (naïve) speakers perceive discourse markers’ role as limited to stalling and marking lexical search. It has been a well-documented fact in the sociolinguistic literature for some time now that laypeople’s intuitions (and stereotypes) about their own language use are not reliable, what is more, they frequently get embarrassed when confronted with their own speech output in the form of the recordings or transcripts of sociolinguistic interviews (cf. e.g. Romaine and Lange 1991: 261).

176

P. B. Furkó

In short, as was demonstrated in previous chapters, the functional spectrum of discourse markers extends well beyond lexical search and stalling, while it was also observed that native speakers use discourse markers consistently and with great precision in order to fulfil a wide range of communicative needs. Like, for example, rather than being a meaningless social noise is a discourse marker of “high frequency and versatility” (Andersen 2001: 221). Its functional spectrum ranges from (11a) marking explanation, (b) emphasis/new information, to (c) quotative, (d) approximative and (e) softening functions. The list of functions is far from exhaustive, for a more detailed functional spectrum of like cf. e.g. Mattsson (2009) or Beeching (2016). (11) (a) My roommate never cleans when I ask him to. Like, I asked him yesterday to clean, and he never did it. (b) This guy is so cool. I mean, he’s like the coolest person you could meet. (c) I went to the clerk to ask him where the beer was, and he’s like, ‘I don’t know, I’m new here’, so I’m like, yeah, sure, like, you should know this, man! (d) I missed like 40 questions on the exam. (e) Could you, like, loan me $100?

Utterance-initial and utterance-medial instances of non-propositional like—such as the ones in the examples above—occur in similar contexts, with similar distribution and frequency in IrE, AmE and BrE (cf. Kallen 2006). Despite this fact, the stereotypical use of like in IrE is in utterance-final position; moreover, it is typically associated with Irish tentativeness and uncertainty. One can observe this stereotype in a variety of cinematographic representations of IrE, in both Intermission and The Guard, for example, non-propositional uses of like occur invariably in utterance-final position if the speaker is Irish, as in the following examples: (12) A [he is] deliberately disrupting my lecture.

7 Discourse Markers in Scripted Discourse II …

B

177

Ah, I’m only having a bit of fun, like. (Csak szórakoztam kicsit, ne vegye ám komolyan.) (The Guard )

(13) You know, I’ve been reading…where they’re smuggling the cocaine out of Colombia in little submarines… Submarines they’ve built themselves, like. ([no DM] Utána olvastam, ezek a kokaint Kolumbiából kis tengeralattjárókon csempészik át [no DM]. Olyanokon, amiket saját maguk építettek.) (The Guard ) (14) Bobbie Gentry said she didn’t know what it was. Supposed to be mysterious, like. (Bobbie Gentry azt mondta, hogy nem tudja mi miatt volt. Nem szabad egyértelm˝unek lenniük [no DM].) (The Guard ) (15) A: Have I got a ronnie? (Van bajeszom? B: A what? (Micsoda?) A: A ronnie, mustache, like. (Bajesz, bajuszféle.) (Intermission)

As for the translation of like and you know, the above examples illustrate some of the translation strategies and the resulting underspecification of like in the target text. In (13) and (14), U2a is used (cf. Sect. 2.1 in Chapter 6), that is, the discourse markers and (any of ) their stylizing effects are completely missing from the target text. In (12), the modal particles csak (“just”) and ám (“by no means”) are used as explicitation/specification strategies. While these items can be considered as S2 strategies (cf. Sect. 2.1 in Chapter 6), the stylizing effect of IrE utterancefinal like is most likely lost on the target audience. In (15), another S2 translation strategy can be observed, and this time the Hungarian affix -féle is used to explicate the conceptual meaning of like, but not its procedural meaning or stylistic effects. Another discourse marker that displays a unique functional spectrum in IrE is sure. In AmE and BrE, the discourse marker sure is primarily used as a feedback signal, marker of an affirmative answer and/or compliance with a request, as well as an attitudinal disjunct with an interpersonal function of inviting agreement from the person or persons addressed (cf. Quirk et al. 1985: 446). In IrE, in addition to its standard

178

P. B. Furkó

functions, sure marks emphasis, reinforcement, epistemic stance, mockery and a range of additional interpersonal and discourse-organizational functions (cf. Moreno 2005). In Irish films—unlike the discourse marker like—sure displays a wide range of functions that are also present in AmE and BrE usage; nevertheless, the utterance-final position of sure appears to be, once again, stereotyped, as it is more salient in cinematographic representations of IrE than in naturally occurring or even literary data (cf. Moreno 2005): (16) I’m Irish, sure. Racism’s part of my culture. (De hát én ír vagyok, a rasszizmus a kultúránk része.) (The Guard ) (17) A: Are we to assume you’re satisfied with the service? (Remélem elégedett volt velünk.) B: I’m totally drained, sure. (Kipurcantam, fikarcnyi er˝o sincs már bennem [no DM].) (The Guard ) (18) You’re after blowing your cover, sure. (Neked l˝ottek az biztos.) (Intermission) (19) It’s the new f…ing business, sure. Everyone’s doing it. (Ez az új cucc b*meg, bizony. Mindenki ezt nyomja [no DM].) (Intermission)

It is interesting to observe that the translations of utterances (15) through (18) all go to some lengths in explicating the informality, nonstandard quality of the source text, such as the use of the target text discourse marker (hát, ~‘well’, ) in (16), slang in (17), (18) and (19) and even obscenity of (19). (19) uses the additional S1 strategy of a target text discourse marker with a similar conceptual as well as procedural meaning. Naturally, except for the S1 strategy, it is hard to tell if certain target text features are intended as renderings of the source text discourse marker sure or other non-standard source text features, whether syntactic (e.g. IrE perfective after ) or lexical ones (e.g. drained, f*ing ). The rendering of source text discourse markers is more probable if one considers that the translations of other source text discourse markers (cf. especially examples 5, 8, 9, 13, 14) included very few S1 and S2 strategies.

7 Discourse Markers in Scripted Discourse II …

4

Quantitative Perspectives on IrE Stereotypes

4.1

Keyness Analysis and Automated Semantic Tagging of IrE Scripted Discourse with Reference to AmE Scripted Dialogues

179

After an overview of the discourse markers and other pragmalinguistic features that are (rightly or wrongly) associated with IrE, I now turn to statistical measures of identifying the most salient lexical items in IrE scripted dialogues. In order to test the keyness of like and sure in cinematographic representations of IrE, the data used for the qualitative phase above was extended, and a test corpus of 103,368 words (IrE corpus, IEC) was compiled based on the subtitles of The Guard, Intermission, Circle of Friends, Leap Year, In the Name of the Father, Once, The Wind that Shakes the Barley, The Butcher Boy as well as all six episodes of the series entitled Derry Girls. At this stage, the scripted discourse corpus (SD), introduced in Chapter 1, was used as a reference corpus, which was based on seasons 1 through 4 of House, M.D. In the first stage of the keyness analysis, the “keyword list” feature of AntConc 3.4.4w was used and all lexical items in the two subcorpora (IEC and SD) were considered; the full list was then shortened with a focus on items that can be potentially associated with stylizing and stereotyping. Consequently, the IEC was automatically tagged using USAS (cf. Sect. 1) with a view to distinguishing between discourse marker and non-discourse marker uses of the relevant lexical items. In the course of calculating keyness, each corpus was used as a reference corpus with respect to the other. In order to calculate keyness, the log-likelihood test was used as a statistical measure (cf. Sect. 5 in Chapter 4). The log-likelihood value of 3.84 was considered critical (p < 0.05): if a lexical item had a value higher than the critical value, it was considered to have keyness in a particular corpus. The results are summarized in Table 1, and potential discourse markers are highlighted in bold italics.

180

P. B. Furkó

Table 1 Keyness analysis of lexical items in the IEC and SD sorted by test corpus and keyness

Lexical item

Test corpus/reference corpus

Normalized frequency (number of tokens per hundred thousand words)

Keyness (log-likelihood)

f*ing Jesus aye now man da wee lads cos please mean like know sure surely cause because figure okay gotta um unless hey think

IEC/SD IEC/SD IEC/SD IEC/SD IEC/SD IEC/SD IEC/SD IEC/SD IEC/SD IEC/SD IEC/SD IEC/SD IEC/SD IEC/SD IEC/SD SD/IEC SD/IEC SD/IEC SD/IEC SD/IEC SD/IEC SD/IEC SD/IEC SD/IEC

258 119 83 448 202 81 64 59 24 107 140 402 522 145 5 131 242 44 208 61 31 40 110 355

402.170 161.021 129.380 128.861 113.669 104.060 99.763 91.969 37.411 27.660 27.232 10.966 6.534