New Trends on Metadiscourse: An Analysis of Online and Textual Genres 9783031366895, 9783031366901

Table of contents :
Contents
List of Figures
List of Tables
1 Introduction
References
Part I Metadiscourse and Academic Genres
2 So What Do We Have Here? An Engineering Lecturer’s Metadiscursive Use of Rhetorical Questions in L1 and English-Medium Instruction
Introduction
Background
Method
Theoretical Framework and Coding
Results
Differences in Use of Rhetorical Questions in EMI and L1
Functions of Micro-Rhetorical Questions
Lecturer’s Stimulated Recall
Discussion
Appendix 1: Transcription Conventions
References
3 Distribution Patterns of Stance Features in English and Russian Conference Presentations
Introduction
Literature Review
Academic Conference Presentation as a Genre
Metadiscourse
Methods
Analysis
Stage A. Data Collection
Stage B. Structural Models of CPs
Stage C. Compilation
Hedges
Boosters
Stage D. Comparison
Stage D. Distribution
Stage E. Conference Presentations vs 3-Minute Thesis Presentations
Discussion
Conclusion
Notes
Appendix
References
4 A Metadiscoursal Approach to Academic Writers’ Construal of Identities Across Brief Reports and Case Reports in Medical Science
Introduction
Context
Register
Genre
Metadiscourse
Method
Analysis
Interactive Resources
Code Glosses
Transitions
Evidentials
Endophoric Markers
Frame Markers
Interactional Resources
Attitude Markers
Hedges
Boosters
Self-Mentions
Engagement Markers
Conclusion
References
5 Metadiscourse Learning Trajectories in Multilingual Learners: A Focus on Attitude Markers and Hedges
Introduction
Literature Review
The Role of Interactional Metadiscourse in Argumentative Writing
Interpersonal Metadiscourse in Multilingual Writing
Research Question
Method
Participants and Learning Context
Instruments
Data Collection Procedure
Data Analysis Procedures
Results and Discussion
Conclusion and Pedagogical Implications
Appendix A
Appendix B: Task Topics
Spanish Topics
Catalan Topics
English Topics
Appendix C: Attitude Markers and Hedges Coded
References
6 A Contrastive Analysis of Metadiscourse by Native and EFL Lecturers in Chinese University MOOCs
Introduction
Literature Review
Theoretical Research on Metadiscourse
Empirical Research on Metadiscourse
Methodology
An Analytical Framework Based on Hyland’s Interpersonal Model
Corpus
Procedure and Reliability
Results and Discussion
General Distribution of Metadiscourse in CCUM
Interactive Dimension
Interactional Dimension
Differences in Metadiscourse Between Chinese EFL Lecturers and English Native Lecturers
Transition Markers
Frame Markers
Evidentials
Self-Mentions
Engagement Markers
Conclusion and Implications
Appendix
References
Part II Metadiscourse and Non-academic Genres
7 Using Twitter for Public Dissemination and Engagement with Science: Metadiscourse on the Twitter Accounts of Scientific Organisations
Introduction
Literature Review
Twitter Use by Organisations
Extending the Framework of Metadiscourse for the Analysis of Digital Genres
Method and Corpus
Results
Communicative Functions of Tweets
Metadiscourse in Tweets
Interactive Resources
Interactional Resources
Conclusions
References
8 Persuasion Through Interactional Metadiscourse of Management Statements of European Renewable Energy Companies
Introduction
Corporate Annual Reports and Management Statements
The Concept of Metadiscourse and Its Persuasive Function
Use of Stances and Modal Verbs
Use of Pronouns
Corpus Description and Methodological Approach
Results and Discussion
Use of Attitudinal Stances
Use of Epistemic Stances
Modal Verbs
Use of Pronouns
Conclusion
References
9 On the Metadiscursive Dimension of Travel Blog Posts: A Cross-Linguistic Analysis
Introduction
Materials and Methods
Blogger’s Self in English and Italian Travel Blogs
Subjectivity Markers in English and Italian: Quantitative Overview
First-Person Subject Pronouns EN I and IT io/mi
Personal Object Pronouns EN ‘me’ and IT ‘me’
Possessive Adjectives EN ‘my’ and IT ‘mio/mia/miei/mie’
Engagement Markers
Conclusions
References
10 ‘I Think, You Know…’: A Corpus-Based Analysis of Metadiscourse in Malaysian Online Podcasts
Introduction
Methodology
Findings
Conclusion
References
Index

Citation preview

New Trends on Metadiscourse An Analysis of Online and Textual Genres Edited by Begoña Bellés-Fortuño Lucía Bellés-Calvera Ana-Isabel Martínez-Hernández

New Trends on Metadiscourse

Begoña Bellés-Fortuño · Lucía Bellés-Calvera · Ana-Isabel Martínez-Hernández Editors

New Trends on Metadiscourse An Analysis of Online and Textual Genres

Editors Begoña Bellés-Fortuño English Studies Department Universitat Jaume I/IULMA Castellón de la Plana, Spain

Lucía Bellés-Calvera English and German Philology Department Universitat de València/IULMA Valencia, Spain

Ana-Isabel Martínez-Hernández Universitat Jaume I/IULMA Castellón de la Plana, Spain

ISBN 978-3-031-36689-5 ISBN 978-3-031-36690-1 https://doi.org/10.1007/978-3-031-36690-1

(eBook)

© The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023 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. This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Paper in this product is recyclable.

Contents

1

Introduction Lucía Bellés-Calvera and Begoña Bellés-Fortuño

Part I 2

3

4

1

Metadiscourse and Academic Genres

So What Do We Have Here? An Engineering Lecturer’s Metadiscursive Use of Rhetorical Questions in L1 and English-Medium Instruction Sarah Khan and Marta Aguilar-Pérez Distribution Patterns of Stance Features in English and Russian Conference Presentations G. Gatiyatullina, M. Solnyskina, R. Kupriyanov, and E. Gafiyatova A Metadiscoursal Approach to Academic Writers’ Construal of Identities Across Brief Reports and Case Reports in Medical Science Sabiha Choura

11

39

69

v

vi

5

6

Contents

Metadiscourse Learning Trajectories in Multilingual Learners: A Focus on Attitude Markers and Hedges Sofía Martín-Laguna

105

A Contrastive Analysis of Metadiscourse by Native and EFL Lecturers in Chinese University MOOCs Dongyun Zhang and Diyun Sheng

129

Part II 7

8

9

10

Metadiscourse and Non-academic Genres

Using Twitter for Public Dissemination and Engagement with Science: Metadiscourse on the Twitter Accounts of Scientific Organisations María-José Luzón Persuasion Through Interactional Metadiscourse of Management Statements of European Renewable Energy Companies Maria Cristina Urloi and Miguel F. Ruiz-Garrido

163

191

On the Metadiscursive Dimension of Travel Blog Posts: A Cross-Linguistic Analysis Giuliana Diani

223

‘I Think, You Know…’: A Corpus-Based Analysis of Metadiscourse in Malaysian Online Podcasts Syamimi Turiman and Siti Aeisha Joharry

245

Index

259

List of Figures

Fig. 2.1 Fig. Fig. Fig. Fig. Fig. Fig. Fig.

2.2 3.1 3.2 3.3 3.4 4.1 4.2

Fig. 4.3 Fig. 4.4 Fig. 4.5 Fig. 4.6 Fig. 4.7

Metadiscourse functions: Macro RhQs (topic shift) and micro RhQs (shift within topic) per 1000 words Micro RhQs in L1 and EMI per 1000 words Boosters Hedges Attitude markers Self-mentions Metadiscourse layer Distribution of the communicative functions of code glosses across BRs and CRs Distribution of the communicative functions of transitions across BRs and CRs Distribution of the communicative functions of evidentials across BRs and CRs Distribution of the communicative functions of endophoric markers across BRs and CRs Distribution of the communicative functions of frame markers across BRs and CRs Distribution of the communicative functions of attitude markers across BRs and CRs

26 29 54 54 55 55 76 80 82 84 86 87 90

vii

viii

List of Figures

Fig. 4.8 Fig. 4.9 Fig. 4.10 Fig. 4.11 Fig. 5.1 Fig. 5.2 Fig. 6.1 Fig. 6.2 Fig. 6.3 Fig. 6.4 Fig. 7.1 Fig. 7.2 Fig. 8.1 Fig. Fig. Fig. Fig. Fig. Fig. Fig.

8.2 8.3 8.4 8.5 8.6 10.1 10.2

Fig. 10.3 Fig. 10.4 Fig. 10.5

Distribution of the communicative functions of hedges across BRs and CRs Distribution of the communicative functions of boosters across BRs and CRs Distribution of the communicative functions of self-mentions across BRs and CRs Distribution of the communicative functions of engagement markers across BRs and CRs Learning trajectories in the mean rate of use of attitude markers in Spanish, Catalan, and English Learning trajectories in the mean rate of use of hedges in Spanish, Catalan, and English The distribution of 20 Chinese universities providing selected MOOCs Proportion ranking of different categories of metadiscourse in CCUM Proportion and ranking of different categories of metadiscourse in SCCL Proportion and ranking of different categories of metadiscourse in SCNL Tweet in the NCSE Twitter account (used with permission) Tweet with exophoric marker in the NCSE Twitter account (used with permission) General distribution of interactional markers across corpora Distribution of attitudinal stances across corpora Distribution of epistemic stances across corpora Distribution of modal verbs across corpora Distribution of pronouns across corpora Distribution of inclusive-we and exclusive-we ‘You know’ used to signal shared knowledge or experience ‘You know’ used to appeal for acceptance in interrogative form ‘You know’ used as fillers between arguments ‘I think’ used as a hedge to sound less assertive when providing factual information ‘I think’ used as ‘utterance launcher’ (Biber et al., 1999) when used in sentence initials

91 93 94 96 117 118 137 141 148 149 170 179 204 205 209 212 213 214 252 253 253 255 255

List of Tables

Table 2.1 Table 2.2 Table Table Table Table Table Table Table Table Table Table

3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 4.1 4.2

Table 4.3 Table 4.4 Table 4.5

Summary of quantitative differences in RhQs between L1 and EMI Micro RhQs and the lecturer’s answers to them in L1 and EMI Size of CoConPres (min, tokens) Size of CoConPres moves by language (tokens) MDMs frequency by language Distribution of stance features Stance features by genre (1000 tokens and %) Stance features by language Metadiscourse markers in English presentations Metadiscourse markers in Russian presentations Hyland’s model of metadiscourse (2005, p. 49) Communicative functions (Swales, 1990; Swales & Feak, 2001) Frequencies of metadiscoursal types Frequencies of interactive resources Distribution of interactional resources across BRs and CRs

23 28 45 45 51 53 56 60 61 63 75 75 77 79 88

ix

x

List of Tables

Table 5.1 Table 6.1 Table 6.2 Table 6.3 Table 6.4 Table 6.5 Table Table Table Table

6.6 7.1 7.2 8.1

Table 9.1

Table 9.2 Table 10.1 Table 10.2 Table 10.3

Descriptive statistics of attitude markers and hedges at Times 1, 2, and 3 in Spanish, Catalan, and English An analytical framework based on Hyland’s interpersonal model Detailed information of selected MOOCs Description of sub-corpora of English native lecturers and Chinese EFL lecturers The general distribution of metadiscourse in CCUM Comparative results of metadiscourse between SCNL and SCCL Metadiscourse items in CCUM Function of tweets Types of tweets and examples Management statements (in English version) retrieved in April 2020 from firms’ websites Subjective markers in the word frequency lists of the EN and IT travel blog corpora (within the first 20 positions) Frequency distribution of engagement markers in the two language corpora Description of the spoken corpus Top 15 words and Ngrams in the corpus Collocates for ‘you know’ and ‘I think’ in the corpus

116 136 138 139 141 150 156 171 172 200

228 236 248 249 250

1 Introduction Lucía Bellés-Calvera and Begoña Bellés-Fortuño

Metadiscourse has been widely used as an umbrella term in the field of discourse analysis for the ways producers interact with their texts— either written or oral—as well as with their readers and listeners (Hyland, 2017, 2019). In other words, this concept is regarded as “discourse about discourse”, as stated by Flowerdew (2015, 17). Hence, interpersonal relationships can be found within communication, either in academic or non-academic contexts (Abdi, 2002; Ädel, 2006; Dafouz-Milne, 2008; Duruk, 2017). Metadiscourse elements in academic genres have been widely studied in literature (Hyland, 2005, 2013; Lorés-Sanz et al., 2010) where writing has been the main focus of research to see the different metadiscoursal L. Bellés-Calvera (B) Universitat de València/IULMA, Valencia, Spain e-mail: [email protected] B. Bellés-Fortuño Universitat Jaume I/IULMA, Castellón de la Plana, Spain e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 B. Bellès-Fortuño et al. (eds.), New Trends on Metadiscourse, https://doi.org/10.1007/978-3-031-36690-1_1

1

2

L. Bellés-Calvera and B. Bellés-Fortuño

variables. The use of metadiscourse as a tool for managing texts and the effect on writers and readers have been analyzed. However, due to digitalization, academic and non-academic communication practices have undergone a transformation process over the past two decades, leading to the emergence of hybridized genres/forms (D’Angelo et al., 2021). The volume presented here is an attempt to depict the new communication scenario. The book is arranged into two main parts where different perspectives are taken when examining written, spoken, and digital genres: Part I: Metadiscourse and Academic Genres Part II: Metadiscourse and Non-academic Genres The first part—Chapters 2 to 6—provides an overview of the metadiscoursal practices found in lecturers’ and multilingual learners’ speech, conference presentations as well as reports and case reports in medical science. As for the second part, there are four contributions—Chapters 7 to 10—devoted to non-academic texts analyzing communication strategies on digital media (i.e., podcasts, blogs, Twitter, and management statements of European renewable energy companies). What can be clearly observed is that multilingualism and social networks have become a popular area of research among scholars interested in new metadiscoursal practices. The book opens with the section entitled “Academic texts”. Chapter 2, by Sarah Khan and Marta Aguilar, from Universitat de Vic and Universitat Politècnica de Catalunya respectively, revolves around a lecturer’s metadiscursive use of rhetorical questions in Catalan, the lecturer’s L1, and the changes produced when English is the medium of instruction. In line with previous research, this study explores the multi-faceted nature of these questions (Crawford-Camiciottoli, 2004; Dafouz & Sanchez-Garcia, 2013; Rigol-Verdejo & Sancho-Guinda, 2015) in order to identify their metadiscursive function (Hyland, 2005). The findings indicate that these rhetorical questions were mainly employed for content purposes as they functioned as macro and micro topicalizers (Chaudron & Richards, 1986). The authors also stress that complex metadiscursive devices tend to be replaced by simpler ones due to the cognitive effort that entails giving a lecture in a second language.

1 Introduction

3

In Chapter 3, Galiya Gatiyatullina, Marina Solnyskina, Roman Kupriyanov, and Elzara Gafiyatova, from Kazan Federal University, present a comparative study where they investigate the distribution patterns of stance features in English and Russian conference presentations related to the field of medical biotechnology. To achieve this goal, they collected a 45,018-word corpus consisting of recorded English and Russian presentations delivered in different English-speaking and Russian-speaking areas, all of them posted on a variety of video channels, such as NIH (genome.gov), Cell and Gene Therapy Conference, UniverTV, FutureBiotech, and RusOncoWeb. The results obtained prove that there is a trend for Russian scholars arguing in an explicit way, which happens to be less common in English spoken scientific domains. With regard to the distribution patterns of hedges, there is a higher frequency rate in English, particularly when reviewing literature and describing the experiments conducted. Presenters’ cultural differences may explain such a choice of discourse markers. Following Hyland’s (2005) interpersonal model of metadiscourse, Chapter 4 examines the metadiscursive features produced by academic writers across brief reports and case reports in medical science and how these linguistic elements affect the creation of their identities. To this end, Sabiha Choura, from the University of Sfax, analyzed a corpus of 81,869 words taken from the Military Medicine journal at interactive and interactional levels. This study reveals that the distribution of metadiscursive features is motivated by generic conventions, with a preference for interactive over interactional markers in brief reports. However, the opposite is true for case reports, thus revealing that academic writers of case reports show more engagement with the readership than their peers in brief reports. The last chapters of this section focus on English as a Foreign Language (EFL) learning environments. Chapter 5, by Sofía MartínLaguna, from Universitat Jaume I, addresses learners’ development of interpersonal metadiscourse knowledge in multilingual classrooms. The study is set in a bilingual region of Spain, where Spanish and Catalan co-exist as co-official languages and English is learnt as a third language (L3), thus examining to what extent multilingual learners’ use of hedges and attitude markers is related across languages over time. As for the

4

L. Bellés-Calvera and B. Bellés-Fortuño

subjects, 313 bilingual high school learners in Spanish and Catalan learning English as their third language participated in this study writing opinion essays in all of these languages. From a quantitative perspective, the findings show that there are stronger correlations over time in both metadiscourse markers analyzed. The author concludes that these results are supported by qualitative analyses of learners’ essays, which also show transfer at phrase and discourse levels. This section comes to an end with Chapter 6, which delves into the production of metadiscursive elements in EFL digital academic platforms. More specifically, Dongyun Zhang and Diyun Sheng, from Shanghai Jincai North Secondary School and Shanghai Normal University, present a contrastive analysis of metadiscourse by native and EFL lecturers in Chinese university MOOCs, a tool that has gained importance within the Chinese government in the last years, particularly to meet educational needs. As in previous chapters of this book, Hyland’s (2005) interpersonal model has been adopted and serves as the basis for the subsequent analysis of a self-built 100,933-word corpus. The authors claim that oral communication through MOOCs leads to a low frequency of metadiscursive features, which are mainly generated to enhance intelligibility, reliability, and interactivity. The rate of these elements seems to be related to lecturers’ proficiency level in English since native speakers tend to use a higher number in contrast to Chinese EFL lecturers. All in all, what emerges from these results is that Chinese EFL lecturers are more concerned with reliability issues. Part II includes four chapters on non-academic genres. María José Luzón (Chapter 7), from Universidad de Zaragoza opens this section. She discusses the role of Twitter as a powerful tool for the dissemination of scientific knowledge and thus engaging and reaching a larger audience. Her study departs from the need to understand how semiotic resources can be combined in this genre. To achieve this purpose, Hyland’s (2005) model of metadiscourse is expanded with semiotic elements (e.g., pictures, emojis, animations) as they complement written discourse. In fact, they play a significant role in the guidance, engagement, and persuasion of readers through the text. A total number of 150 tweets posted by three scientific organizations make up the corpus examined here. All of them deal with a variety of scientific issues (e.g., climate

1 Introduction

5

change or wildlife, the conservation of nature and life diversity). Both visual and verbal metadiscursive strategies are combined with the aim of linking to content on other sites, influencing readers’ comprehension of and attitude toward content, and prompting the readers to take specific actions. The choice of metadiscourse resources is also determined by the purpose of these tweets. In Chapter 8, Maria Cristina Urloi and Miguel F. Ruiz-Garrido, from Universitat Jaume I, inspect the persuasive communication strategies used in management statements of European renewable energy companies through the identification of interactional metadiscursive elements. Based on Hyland’s model (2005), they perform a comparative analysis of interpersonal features with a focus on the role of stances (Biber, 2006). In this sense, they can explore how these devices contribute to promoting and creating a positive corporate identity as well as to engaging readers by means of pronouns (Fortanet, 2004; Kamio, 2001; Wieczorek, 2009). Quantitative and qualitative analyses report the relevance of metadiscourse in both management statements. In fact, executives and chairmen appeared to employ a similar quantity of interpersonal metadiscursive elements, yet with slightly different tendencies of persuasive linguistic strategies. A different metadiscursive dimension is offered in Chapter 9 with a cross-linguistic study on travel blog posts. Giuliana Diani, from the University of Modena and Reggio Emilia, describes the most common elements spotted in these written texts produced by American and Italian travelers describing Italian destinations. In this regard, emphasis is placed on self-mentions and engagement markers, from text and corpus perspectives, drawing attention to the similarities and dissimilarities across languages and cultures in the construction of the blogger’s stance and in the way bloggers engage with their readers. Finally, new communication opportunities come with the popularization of online spoken discourse in Chapter 10. Syamimi Turiman and Siti Aeisha Joharry, from Universiti Teknologi MARA, investigate the use of stance and engagement as interactional elements (Hyland, 2005) in a collection of podcasts from a local Malaysian independent business radio station. To this end, a corpus linguistics approach was necessary. When it comes to frequency, it was found that “you know” and “I think”

6

L. Bellés-Calvera and B. Bellés-Fortuño

were the most frequent two-word combinations. The former generally functions to signal shared experience. The latter, “I think”, is used mainly as a hedge to sound less assertive when expressing opinions and providing factual information. The authors argue that both discourse organization and audience engagement are conditioned by the use of stance and engagement markers, hence becoming an integral aspect of conversations in podcasts. In sum, all the contributions provide an updated overview in the analysis of academic and non-academic genres in a digital era. Not only has the aim of research focused on digital genres but also in hybrid and traditional genres that are lesser known, thereby enriching a field of growing interest with a variety of cross-linguistic, multilingual, and disciplinary angles and perspectives.

References Abdi, R. (2002). Interpersonal metadiscourse: An indicator of interaction and identity. Discourse Studies, 4 (2), 139–145. Ädel, A. (2006). Metadiscourse in L1 and L2 English. John Benjamins. Biber, D. (2006). The expression of stance in university registers. John Benjamins. Chaudron, C., & Richards, J. C. (1986). The effect of discourse markers on the comprehension of lectures. Applied Linguistics, 7 (2), 113–127. Crawford-Camiciottoli, B. (2004). Interactive discourse structuring in L2 guest lectures: Some insights from a comparative corpus-based study. English for Specific Purposes, 3, 39–54. Dafouz, E., & Sanchez-Garcia, D. (2013). ‘Does everybody understand?’ Teacher questions across disciplines in English-mediated university lectures: An exploratory study. Language Value, 5, 129–151. Dafouz-Milne, E. (2008). The pragmatic role of textual and interpersonal metadiscourse markers in the construction and attainment of persuasion: A cross-linguistic study of newspaper discourse. Journal of Pragmatics, 40 (1), 95–113. Duruk, E. (2017). Analysis of metadiscourse markers in academic written discourse produced by Turkish researchers. Journal of Language and Linguistic Studies, 13(1), 1–9.

1 Introduction

7

Flowerdew, J. (2015). Revisiting metadiscourse: Conceptual and methodological issues concerning signalling nouns. Ibérica, Revista de la Asociación Europea de Lenguas Para Fines Específicos, 29, 15–34. Fortanet, I. (2004). The use of ‘we’ in university lectures: Reference and function. English for Specific Purposes, 23(1), 45–66. Hyland, K. (2005). Metadiscourse: Exploring interaction in writing. Continuum. Hyland, K. (2013). Disciplinary discourses: Social interactions in academic writing. University of Michigan Press. Hyland, K. (2017). Metadiscourse: What is it and where is it going? Journal of Pragmatics, 113, 16–29. Hyland, K. (2019). Metadiscourse: Exploring interaction in writing (2nd ed.). Continuum. Kamio, A. (2001). English generic we, you, and they: An analysis in terms of territory of information. Journal of Pragmatics, 33(7), 1111–1124. Lorés-Sanz, R., Mur-Dueñas, P., & Lafuente-Millán, E. (Eds.). (2010). Constructing interpersonality: Multiple perspectives on written Academic genres. Cambridge Scholars Publishing. Rigol-Verdejo, R., & Sancho-Guinda, C. (2015). Questioning questions: Remodelling oral presentations in the ESP Classroom. Procedia, 173, 132–137. Wieczorek, A. E. (2009). This is to say you’re either in or out: Some remarks on clusivity. Critical Approaches to Discourse Analysis across Disciplines, 3(2), 118–129.

Part I Metadiscourse and Academic Genres

2 So What Do We Have Here? An Engineering Lecturer’s Metadiscursive Use of Rhetorical Questions in L1 and English-Medium Instruction Sarah Khan and Marta Aguilar-Pérez

Introduction With the growth of EMI (English-medium instruction) in European higher education, more and more university lecturers are called upon to teach their subjects in English. In some institutions, parallel subjects are available in both L1 and English as a more inclusive strategy for both local and international students to choose from, depending on their proficiency in the respective languages on offer. Another reason why EMI, together with ESP courses, can be regarded as inclusive strategies is that they both have the potential to improve students’ disciplinary communication skills and their intercultural competence at different S. Khan (B) Universitat de Vic, Barcelona, Spain e-mail: [email protected] M. Aguilar-Pérez Universitat Politècnica de Catalunya, Barcelona, Spain e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 B. Bellès-Fortuño et al. (eds.), New Trends on Metadiscourse, https://doi.org/10.1007/978-3-031-36690-1_2

11

12

S. Khan and M. Aguilar-Pérez

levels (Aguilar, 2018; Aguilar-Pérez, 2021; Arnó-Macià & Aguilar-Pérez, 2019) and hence offer an Internationalization at Home (IaH) experience to those local students who cannot afford a study abroad experience. These potentially international contexts provide a rich source of data for comparative language studies, particularly, as is the case in this study, when the same lecturer teaches the same course in both languages. Such comparisons, on the one hand, serve teachers engaged in professional development, providing personalized feedback and making language and pedagogical differences between their L1 and EMI lecturing much more salient. On the other hand, these comparisons also allow ESP teachers to update knowledge on their students’ needs, as ESP courses in settings where the presence of EMI is increasing are also perceived to be good preparation toward successfully following an EMI course (Arnó-Macià et al., 2020). As lecturers teaching the same content in L1 and EMI are far and few between, little research has been reported on them, despite their potential for improving our understanding of teaching and learning in the EMI classroom. Given that for most students understanding a lecture in English is a demanding task, content lecturers who teach through English should make conscientious efforts to make knowledge accessible (Doiz & Lasagabaster, 2022), a task fraught with difficulties. Among the few studies carried out, metadiscourse appears as a feature of lecture clarity. The following are examples of some differences found in lecturers’ discourse in EMI when compared with L1: lower speech rate, more repetition and greater formality (Thøgersen & Airey, 2011), less stylistic richness in metadiscourse devices, less explicit signaling to mark topic shift and fewer conclusion markers (Dafouz & Núñez, 2010), fewer rhetorical questions and comprehension checks (Maíz-Arévalo, 2017), less interaction triggered by lecturer questions (Sánchez-García, 2018), lower speech rate but more discourse structuring and redundancy (Ädel, 2010; Costa & Mariotti, 2017; Lee & Subtirelu, 2015; Molino, 2018; Suviniitty, 2012; Zare & Tavakoli, 2016), or lower speech rate and less metadiscourse for clarifying, endophoric marking and reviewing (but more for managing terminology, making asides and introducing topics) (Aguilar & Khan, 2022).

2 So What Do We Have Here? An Engineering …

13

Although rhetorical questions (RhQs) have been identified and compared in some of these studies, they have received little mention and, when they have, RhQs have been examined either as one out of many other metadiscursive items or as one of the different types of questions lecturers pose in class. Understood in this study as questions posed by the lecturer that are not expected to be answered by the audience, RhQs in academic lecturing merit special attention mostly due to the explicitness and clarity that their metadiscursive nature is thought to bring to the listener, helping non-native students retain lecture content (Aguilar & Arnó, 2002; Chaudron & Richards, 1986; Flowerdew & Miller, 1997) and allowing teachers to signal and organize the structure (Ädel, 2010). Because of the lack of focal studies on RhQs in lecturing, and particularly in EMI lecturing, little is known about their specific pedagogical function, that is, their role in enhancing comprehension of disciplinary content when the latter is interspersed with RhQs. This may be even more relevant in EMI contexts where English adds an extra layer of complexity for both lecturers and students (Doiz & Lasagabaster, 2022). Thus, the purpose of this study was to foreground RhQs by placing them at the center of our research, firstly comparing the RhQs an engineering lecturer used in delivering the same content in L1 and EMI and secondly examining their communicative function in order to further our understanding of the effects created by the use of RhQs and ultimately of the challenges lecturers face with their spoken delivery when shifting from L1 to EMI.

Background Rhetorical questions have long been known as a common persuasive technique in public discourse. They have been studied in written discourse as well as in public speeches (e.g., TED talks, political speeches, university lectures) or everyday conversations. In research on academic discourse, as mentioned above, RhQs are either regarded as a type of textual metadiscursive device or as a type of teacher question. Within the metadiscourse strand of research, RhQs are studied as self-answered questions that signal topic and topic shift (Ädel, 2010; Hyland, 2005;

14

S. Khan and M. Aguilar-Pérez

Nattinger & DeCarrico, 1992) while within the teacher questions strand, RhQs tend to be defined as questions that are not expected to be answered, often left open for contemplation (Dafouz & García, 2013; Dafouz & Núñez, 2010; Fortanet, 2004; Hu & Duan, 2019; Morell, 2004; Rigol-Verdejo & Sancho-Guinda, 2015; Sánchez-García, 2020). In many cases, they are viewed as content-oriented questions (Querol-Julian, 2008; Suviniitty, 2012; Thompson, 1998), and defined as interrogative forms that “don’t expect an answer”, “have the feel of an assertion” and “can optionally be answered” (Biezma & Rawlins, 2017, p. 302). Doiz and Lasagabaster (2022) and Zhang and Lo (2021) have very recently studied the scaffolding role of metadiscourse in the construction of knowledge and in making its transmission accessible to students within EMI. Drawing on Zhang and Lo’s study (2021), Doiz and Lasagabaster (2022) study the role of textual and interactive metadiscourse (Hyland, 2005) in EMI from an English-as-a-Lingua-Franca perspective (i.e., where native English is not the reference language model), finding that transition markers are the most commonly used metadiscursive items, followed by code glosses and reminders, while frame markers are the least used, in both the Spanish and Chinese contexts. Given the paramount role of frame markers, reminders and code glosses in transmitting knowledge, Zhang and Lo (2021) and Doiz and Lasagabaster (2022) point to the need to draw attention to EMI lecturers and the insufficient use they make of these items. Yet, RhQs are not the focus of their study. By and large, metadiscourse research identifies the role of RhQs in structuring academic discourse (viz. signaling topic shift, marking a transition) and to a lesser extent acknowledges their role in engaging with the audience, thus contemplating the two typical metadiscursive functions, textual/discourse-organizing and interactional/audience-oriented (Ädel, 2010; Aguilar, 2004; Hyland, 2005). Research on metadiscourse in EMI has found lectures to be characterized by discourse-organizing metadiscursive items (Doiz & Lasagabaster, 2022; Molino, 2018; Zare & Tavakoli, 2016). Dafouz and Núñez (2010) found that Spanish lecturers used more explicit metadiscourse and wider stylistic choices when teaching in L1. Molino (2018), for example, analyzed metadiscourse

2 So What Do We Have Here? An Engineering …

15

using Ädel’s (2010) reflexive model to classify personal and impersonal markers retrieved from her corpus. She found the discourse-organizing function of these markers, including an example of a RhQ, more prevalent than audience-oriented metadiscourse. In the same vein, Zare and Tavalokli (2016), who compare lectures with classroom discussions, encounter more discourse organizing metadiscourse in lectures whereas more audience interaction in discussions. In contrast to these studies, Costa and Mariotti (2017), comparing English- and Italian-medium instruction at Master’s level, describe RhQs, one of several discursive features they analyze, as marking lecture interactivity and found that rhetorical questions had slightly fewer occurrences in EMI than in L1. All in all, whether RhQs are considered textual or interactional, they have a relatively inconspicuous place among other metadiscourse markers. Worth mentioning among studies on the interactional function of RhQs is Bamford (2005). In this earlier qualitative study, Bamford views academic lectures by native English speakers as highly interactional where “one of the ways in which the lecturer enhances the interactive conversational nature of the lecture is by asking questions which he then proceeds to answer himself ” (Bamford, 2005, p. 126). She analyzes the function of RhQs using a conversation analysis framework, considering the question-and-answer sequences of adjacency pairs, and claiming that lecturers conform to conversational norms. Within studies centering on teacher questions, the metadiscursive role of RhQs is often acknowledged, though usually attention is placed on the interactive role of display and referential questions and comprehension checks, RhQs being relegated to a monologic-based technique and an opportunity to interact with students that lecturers miss. Hu and Duan (2019), for example, found that teacher questions eliciting a verbal response (even by the teachers themselves) in both Chinese-medium instruction (CMI) and EMI tended to be cognitively simple questions eliciting lower-order thinking skills (e.g., remembering, understanding) in spite of the potential for questions to promote higher-order thinking (analysis, evaluation and creativity), basing complexity on an adaptation of Bloom’s 6-level taxonomy. They found no significant difference in syntactic or cognitive complexity of the questions asked between CMI and EMI. Dafouz and Sánchez-Garcia’s (2013) small-scale study of three

16

S. Khan and M. Aguilar-Pérez

Spanish EMI lectures on Business, Physics, and Engineering courses analyzed four types of teacher questions (display questions, referential questions, confirmation checks and self-answered questions), finding similarities (rather than differences) in the distributions of question types. Neither of these studies draw attention to RhQs, focusing rather on the quality of the questions that elicit a verbal response from the audience in terms of their potential for learning (eliciting verbal interaction and co-constructing knowledge). RhQs, however, can also be regarded as a tool for constructing meaning, a synonymous reformulation of topicalizers and transition markers. In lecture comprehension, macroorganizing phrases that signal topic shift (e.g., let’s start topic number 4, let’s take a peek at these slides), which could be reformulated into RhQs, have been found to create a greater impact than micro monosyllabic markers (e.g., so, ok, right ) because they help non-native students retain lecture content, whereas micro markers “do not add enough content to make the subsequent information more salient or meaningful” (Chaudron & Richards, 1986, p. 123). Thus, because of their length and lexical complexity, questions in general and RhQs in particular (e.g., What’s this first instruction doing? ) act in the same way as other metadiscursive textual items that render lectures more comprehensible and appealing (Morell, 2004). Although their interrogative form is relatively easy to identify, since RhQs act as an assertion or an imperative (e.g., let’s look at x versus What is x? ), their functions are far more difficult to decipher. After identifying three major functions of questions found in civil engineering students’ oral presentations in an ESP setting (rhetorical, display and referential), Rigol-Verdejo and Sancho-Guinda (2015) further subdivide RhQs into three categories, evaluative (eliciting a neutral or confrontational response), repository (research questions ‘Can we eliminate nuclear wastes?’, ‘Not yet’) and mention (used to clarify content as glosses or build intimacy in an aside, ‘Governmental corruption (How long are we going to stand delinquents in office?) is today a big concern’), although few examples of these are provided in their paper. Based on students’ oral presentations, this classification does not provide enough explanatory power about the functions of the RhQs for our study on lecturing, in

2 So What Do We Have Here? An Engineering …

17

spite of shedding light on their functions of capturing the listeners’ attention or establishing some sort of dialogue through irony and humour (Magnuczné-Godó, 2011). A considerable number of studies analyzing teacher questions put forward somewhat similar classifications. Worth mentioning are two studies by Sánchez-García (2010, 2018), who draws upon DaltonPuffer’s taxonomy of questions for a CLIL primary and secondary education setting. In her 2018 study, Sánchez-García studied teacher questions, RhQs being one of the types examined, in EMI and L1 (Spanish). She analyzed the cognitive load of questions using DaltonPuffer’s (2007) classification based on secondary education students, identifying four types of questions, (referential, display, fact-closed and fact-open), to which she added three types dealing with miscomprehension (comprehension check, confirmation checks and clarification requests). Dalton-Puffer classifies them according to their complexity, from greater to lower cognitive load: metacognitive, reason, explanation, description, fact-open and fact-closed . In a previous study, Sánchez-García (2010) examined lecturer questions in six lectures from three disciplines (Business, Physics and Engineering) in L1 Spanish and EMI. She adapted Dalton-Puffer’s taxonomy, including rhetorical questions within the repertoire (display questions, referential questions, rhetorical questions, retrospective questions, self-answered questions, personal addresses). Interestingly, she distinguished between the lecturers’ selfanswered questions which structure or develop the ongoing discourse, and rhetorical questions, where a question is left open and unanswered to provide students with food for thought, and concluded that questions are a pivotal tool for making teachers’ lessons more interactive. Along these lines, Morell’s (2004) study identified four types of lecturer questions: referential, display, rhetorical and indirect questions, finding that the more interactive lecturers used more display questions and few RhQs. Khan (2018), in a comparison of lecturing strategies of 4 EMI lecturers on an International Business program, also found that RhQs were not among the most frequent strategies employed. In this study, not all the lecturers used RhQs, but they were characteristic of one particular lecturer, suggesting that they may be related to a personal lecturing style rather than the discipline being taught. However, as these studies on

18

S. Khan and M. Aguilar-Pérez

teacher questions do not focus on RhQs alone, we could say that the bulk of research has focused on questions that elicit a response from students and render lecturing more interactive and dialogic, somehow overshadowing other important functions of RhQs questions—sometimes viewed as ‘failed’ questions, or display questions that went unanswered because lecturers failed to pause and thus missed the opportunity of creating interactive episodes (Navaz, 2020). In this article, therefore, we aim to place RhQs at the center of our research to study their metadiscursive functions. This study relates to the findings from a previous one (Aguilar & Khan, 2022) where the same lecturer teaching in L1 and EMI was analyzed in terms of the metadiscourse he used. In this previous study, a total of 21 different kinds of metadiscourse were identified in the lectures according to their discourse function, using Ädel’s (2010) classification for spoken metadiscourse. The lecturer was found to employ a substantial amount of metadiscourse striking a balance between structuring his discourse (metatextual) and engaging his students (audience interaction). Comparing the L1 and EMI classes, there were few differences in the frequency and type of metadiscourse, suggesting that the move from L1 to EMI did not hinder the lecturer’s dynamics in this sense. One metadiscourse item, managing the comprehension channel (e.g., Any questions regarding this? Does that make sense? ) was particularly frequent, leading us to conclude that despite the monologic nature of the lectures, the lecturer was consistent in engaging his audience and aware that they may have difficulty following him in both L1 and EMI. It was hypothesized that, among other resources, the use of metadiscourse could account for the students’ high rating of the lecturer. Against this backdrop, rhetorical questions, a striking feature of the lecturer’s discourse, were identified as an outstanding lecturing practice, although Spanish academic culture may also have to be factored in (Maíz-Arévalo, 2017; Sánchez-Garcia, 2020), given that “the Spanish academic style traditionally favors the use of rhetorical questions as a way to organize discourse and to keep the audience’s attention” (Maíz-Arévalo, 2017, p. 25). Assuming the inherently metadiscursive role of RhQs, our aim therefore was to find out if the lecturer transferred them from L1 to EMI, in the same manner as he did with his metadiscourse (Aguilar & Khan, 2022), to examine their

2 So What Do We Have Here? An Engineering …

19

function and to reflect on their value for EMI teacher training. With this in mind, we put forward the following research questions: RQ1: Are there differences in the use of RhQs between L1 and EMI? If so, which kinds of differences? RQ2: What metadiscursive function(s) do the RhQs have?

Method The data analyzed in this study was gathered in an engineering school at a Spanish university (Aguilar-Pérez & Arnó-Macià, 2020). An engineering lecturer was video recorded giving the same two lectures in Advanced Electronics in Catalan (L1) and in English (EMI) on a Master’s in Industrial Technology. The lecturer was 35 years old, had a C1 level of English (according to the CEFR), had been teaching the subject for the third year running in both languages and was highly rated by both the L1 and EMI students, despite his course being considered challenging. Students’ English proficiency was between B2 (upper intermediate level) and C1 (advanced user) in the EMI course. A total of 4.5 hours of recordings and a corpus of 27,694 words was produced from transcribing the 4 lectures (2 in L1 and 2 in EMI) using Atlas.ti software. Lectures averaged between 5600 and 7700 words and lasted between 64 and 77 minutes. They were predominantly monologic with little student interaction. Apart from the coding of RhQs and simple quantitative analysis with Atlas.ti software, and in order to extend our understanding of the quantitative results, we also interviewed the lecturer using stimulated recall by showing him excerpts from the videotaped lectures in which he posed a series of RhQs. Our aim was to further our understanding of his lecturing style and his underlying motivations for using them, if any, so we did not mention we were studying his use of questions until the end of the recall. Before showing him the video, we asked him if he could describe his lecturing method, that he was aware of, and then if he found it hard to teach for 90 minutes in one go. After showing him a couple of video excerpts with the same concepts being explained in L1 and English, we asked him if

20

S. Khan and M. Aguilar-Pérez

he thought they were representative of his way of teaching and showed him the transcription of the excerpts, asking him if he could detect any characteristic way of transmitting knowledge. This brief online interview was recorded and then his answers were transcribed.

Theoretical Framework and Coding As the lecture transcripts in L1 and English were replete with RhQs, we set out to identify and code them. We first defined RhQs as those questions that were directly or indirectly answered by the lecturer himself. They were distinguished from other types of questions produced by the lecturer, such as display or referential questions or comprehension checks, which aim to elicit a response from the audience, often followed by a longer pause. In contrast, RhQs were characterized by the absence of a student response, lack of eye contact on behalf of the lecturer (as observed in the video recordings) and the lecturer self-answering them, often immediately. Therefore, RhQs were usually not followed by a long pause, which helped us to discard those questions the lecturer was actually expecting students to answer. We also included non-standard forms of RhQs (Why did not I see that? ) from the EMI lectures (see also Molino, 2018), as they were still self-answered and intelligible. The short sequence below from a lecture in English, illustrates how the lecturer embedded RhQs (in bold) in his discourse: #1 and what about (3.) the 0 bit and the negative bit? (.) so: the negative bit remains unchanged as well as the 0 bit (.) therefore (.) we have exactly the same as before (2.) okay? (5.) any questions so far? (WB) what’s within B8? and we will finish this block (.) in B8 we have a MOVW to a memory position (WB) to which position position 8C (.) okay? which is position 8C? which variable are we storing in 8C? we actually are storing variable C (.) so we are taking this result this 49 stored within our w register and moving this result towards our address 8: 8C okay? (.)

2 So What Do We Have Here? An Engineering …

21

(See Appendix 1 for transcription conventions). With the RhQs identified in the lecture transcripts, we then read them in context and carried out a two-level analysis to understand their metadiscursive function and their cognitive complexity. In the first level of analysis, we created two metadiscursive categories either denoting a topic shift, which we named macro RhQs or marking the lecturer’s navigation within a given topic, micro RhQs. In Ädel’s taxonomy (2010) of spoken metadiscourse, the RhQs identified in our corpus usually fall within the Introducing Topic and/or Adding to Topic subcategories, both of them regarded as Discourse Organizing devices within the broad Metatextual metadiscourse category. In our data, some RhQs certainly perform a textual role, yet their role in Interacting with the Audience (the second broad category in Ädel’s taxonomy) was noticeable, mostly because they are questions that are implicitly catering for students’ queries, questions that the lecturer anticipates the students may raise and that he uses to build up knowledge. The macro RhQs were all open questions of which the following are representative samples from different parts of an EMI lecture: #2 EMI So what more do we have here? And then what else? Next instruction what is it doing? What else can we see? Macro RhQs, thus, divided the lecturer’s discourse into large portions according to topic. They tended to be lexically and syntactically simpler than the micro RhQs. On the other hand, micro RhQs were predominantly open questions occurring either in isolation or deploying a strategic function to keep students engaged with the rising-falling intonation created by sequences or strings of questions and answers (micro RhQ + Idea 1 + micro RhQ + Idea2 + micro RhQ + Idea3). They were on average lexically and syntactically more complex. Some examples of micro RhQs are illustrated below:

22

S. Khan and M. Aguilar-Pérez

#3 EMI Then what are we doing here if our arithmetic logic unit is performing this operation? What are we doing if we are dealing with unsigned arithmetic? Which is the indicator that is telling us guys this is not working something wrong happened? Closed questions were also found among the micro RhQs (albeit in lower numbers) of which some examples are provided below: #4 EMI Can you see how it is making the loop? Does that mean that we are restricted to addressing just these 256 bits? Do we have a digit carry and a carry? METACOGNITIVE: veieu de quina manera el compilador ha implementat això? [Can you see how the compiler is implementing this?] → followed by metacognitive assessment or opinion (L1) REASON: why do we move this cd value to position eight a? → followed by reason (EMI) EXPLANATION: what do they mean? → followed by an explanation (EMI) DESCRIPTION: com ho fa? [How does it do it?] → followed by a description (L1) FACT: a bit value of # what? → followed by a factual answer, a number (EMI)

Inter-rater reliability was achieved when researchers examined and coded the RhQs in context separately and then discussed the codes together, accepting or disregarding codes. Once all the RhQs were coded and quantified in a spreadsheet, the data was then normalized (codes per 1000 words) to allow comparisons between languages.

Results In this section the results of our quantitative and qualitative analysis of RhQs will be presented, starting with the analysis of the lecture transcripts for differences between L1 and EMI and the metadiscursive

2 So What Do We Have Here? An Engineering …

23

functions of RhQs and finishing with the lecturer’s reflections on his style of lecturing through stimulated recall.

Differences in Use of Rhetorical Questions in EMI and L1 To begin with, the lecturer’s speech rate, in words per minute, across both languages of instruction was slow; in particular his rate was 28.55% slower in the EMI than L1 lectures, a result aligning with previous comparative research on EMI lecturing (Arkin & Osam, 2015; Costa & Mariotti, 2017; Crawford-Camiciottoli, 2004; Thøgersen & Airey, 2011). Turning to the data on the presence of RhQs, a considerable number was found in the corpus: a total of 209 (excluding repetitions), 146 in L1 and 63 in EMI. Comparing RhQs in EMI and L1 per 1000 words (Table 2.1), we still found a proportionally higher number in L1 (M = 9.48) compared to EMI (M = 5.13), resonating with Maíz-Arévalo (2017, p. 19), who found more RhQs in L1 (15.5%) compared to EMI (5.5%) in a first-year undergraduate course, as well as with Costa and Mariotti (2017). To illustrate this difference in context, the following excerpts represent the same point in the L1 and EMI lectures where the lecturer is explaining the first instruction in the two languages. As can be seen, he uses four RhQs in L1 whereas in EMI he only uses one. #5 L1 What is the first instruction doing? well it loads value 0 onto the BSR (.) Macro RhQ Table 2.1 Summary of quantitative differences in RhQs between L1 and EMI Frequency of RhQs (per 1000 words) Mean number of words forming RhQs Mean number of RhQs with personal pronouns

L1

EMI

9.48 5.77 2.86

5.13 7.48 1.14

24

S. Khan and M. Aguilar-Pérez

what does this mean? that in fact it is addressing the first memory bank (.) ok? (.) Micro RhQ that is what it is doing (.) it is addressing the first bank in the data memory (.) very well and after what does it do? (.) Micro RhQ this instruction, what, what do you think it means? Micro RhQ (71 words, in original L1) [our translation]

#6 EMI so (.) the first instruction actually what is it doing? Macro RhQ it is just (.) these instructions (.) do you remember these instructions from the last day movlb what does it mean? (3.)(Disregarded RhQ, as followed by long pause)) move literal to bsr register (.) so basically here what we are doing is just selecting the memory bank (.) so (.) we will be dealing or storing variables throwing things within the first portion of memory (.) so the first two hundred sixty-six bites of our data memory (.) (73 words) In the EMI excerpt the metadiscursive micro marker so appears several times, for example pre-announcing the macro RhQ in line 1, and marking the lecturer’s move from his role as lecturer/speaker to the role of student questioner (Bamford, 2005). We found the metadiscursive use of so very common throughout the EMI lectures (M = 17.6) compared to the use of the equivalent marker in Catalan, doncs (M = 5.4). In the excerpt above, so also precedes explanations in lines 6, 7 and 8, marking shifts in the focus of the discourse (Bamford, 2005). Further differences between the EMI and L1 lectures (Table 2.1) include the average number of words forming the RhQs, which was slightly higher in EMI (M = 7.48) than in L1 (M = 5.77). In other words, when the lecturer used RhQs far more frequently in L1, they

2 So What Do We Have Here? An Engineering …

25

tended to be a little more succinct and to the point. This effect may also have been caused partially by morphological differences in question formation between Catalan and English, as can be seen below in equivalent RhQs from the two languages, where the same meaning is conveyed in fewer words in Catalan than in English. #7 L1 què hem de fer? quins arguments rep? què estem fent?

What do we have to do? Which arguments does it receive? What are we doing? (our translation)

In terms of engaging the audience in his discourse by using personal pronouns, we or you, within the RhQs, curiously, there were about three times more in L1 (M = 2.86) compared to EMI (M = 1.14). Therefore, the lecturer’s discourse may have been perceived as more inclusive and less formal in his L1 whereas in EMI his use of RhQs was more impersonal, substantiating Aguilar-Pérez & Arnó-Macià (2020) analysis of metadiscourse. #8 L1 que és el que guardem aquí? que hauríem d’escriure amb codi C? Què més tenim aquí?

What are we saving here? What should we write in C code? What else do we have here? (our translation)

EMI How is it coded then? How is it stored? What is it doing?

Finally, it can be seen that RhQs in general perform the two textual and engaging functions that metadiscourse has been ascribed to (e.g. Ädel, 2010; Aguilar, 2004; Hyland, 2005); thus, RhQs perform a clear textual, discourse-organizing function of introducing new topics and helping the lecturer to gradually develop and construct knowledge in a stepwise manner, while at the same time the overall effect of having a lecture replete with so many questions—even if not directly addressed to students or eliciting a verbal response—is that of engaging in an implicit dialogue with the audience.

26

S. Khan and M. Aguilar-Pérez

The following step consisted in obtaining a fine-grained examination of the types of macro and micro RhQs. Of the 209 RhQs found in the corpus, 30 were identified as macro RhQs, which divide the lecturer’s discourse into large portions according to topic, and 179 were micro RhQs. Figure 2.1 presents the proportion of macro RhQs to micro RhQs within and across each language. As expected, in both languages there are far fewer macro RhQs compared to micro RhQs restricted by their function for denoting a topic shift. Comparing L1 and EMI, we can see that in EMI the lecturer uses fewer of both types of RhQs. In fact, the shift in language corresponds to a drop in macro RhQs to approximately 40% and to just under 60% in micro RhQs in EMI. In terms of macro RhQs we could conclude from these results that the lecturer denotes topic shift a lot less in EMI. However, in our study of the same lecturer’s metadiscourse (Aguilar & Khan, 2022), although few quantitative differences were found between L1 and EMI for 21 metadiscourse items, we found the opposite result, with more topic shift in EMI (M = 5.3) compared to L1 (M = 1.7). Closer inspection of the topic shift item, to understand these seemingly paradoxical results, revealed that the lecturer was marking topic shift in both languages but substituting RhQs as topic shifters in EMI for a linguistically simpler 9 8

7.85

7 6 4.64

5 4 3 2 1

1.62 0.41

0 Macro

Micro

Macro

L1

Micro EMI

Rhetorical Questions

Fig. 2.1 Metadiscourse functions: Macro RhQs (topic shift) and micro RhQs (shift within topic) per 1000 words

2 So What Do We Have Here? An Engineering …

27

alternative form of metadiscourse like so let’s or let’s, a form he used repeatedly, as seen in the examples below: #9 EMI so let’s write down first the machine code Let’s move on so: let’s talk about the status register which is integrated within our microcontroller let’s talk about these bits so let’s make a table

Functions of Micro-Rhetorical Questions In the next analysis we take a closer look at the micro RhQs the lecturer employs, to understand more specifically what the lecturer is indicating to his students about the information he is transmitting or how he is priming them for the information to come. In Table 2.2 we see examples of the different functions identified (metacognitive, reason, explanation, description, fact), all five of which were present in both languages. The examples included are the shortest question-and-answer sequences found, for reasons of space, but there were often much longer sequences where the answer to the RhQ was not provided immediately but after a long stretch of discourse (see Bamford, 2005; Flowerdew & Miller, 1997). In Fig. 2.2 we compare the average number of micro RhQs which functioned as metacognitive or preceded a reason, explanation, description or fact. As can be seen, most micro RhQs in the four lectures analyzed referred to facts and explanations, the cognitively simpler question types, in both languages of instruction, whereas RhQs which were metacognitive or giving reasons or descriptions were less frequent.

28

S. Khan and M. Aguilar-Pérez

Table 2.2 Micro RhQs and the lecturer’s answers to them in L1 and EMI Micro RhQs

L1

EMI

metacognitive

això té sentit o no té sentit? perquè el bit més significatiu d’un nombre en binari sempre és el bit de signe d’acord? (.) [Does that make sense or doesn’t it? Because the most significant bit of a binary number is always the sign bit ok?] és a dir carregar un literal una constant al registre BSR d’acord? Per què? doncs perquè possiblement el compilador ha decidit que: amb aquestes instruccions(.) de fet ja ho veieu (.) treballaríem amb mode bank (.) [I mean load a literal a constant on the BSR register ok Why? Well because possibly the compiler has decided that with these instruction (.) actually you can see (.) we would work in bank mode] de vegades sí que vindrà alguna dada de la pròpia instrucció (.) quin tipus de dades? (.) les constants [.sometimes some data comes from the same instruction (.) What kind of data? the constants]

it’s fif- fifty-four? yeah? okay (.) why did not I see that? (.) sixteen plus two (2.) + o + yeah yeah yeah yeah it’s okay yeah

reason

explanation

it will be negative why? because here we have a one okay?

what does it mean? it means if our arithmetic logic unit gives us a negative result this bit would be set to one

(continued)

29

2 So What Do We Have Here? An Engineering …

Table 2.2 (continued) Micro RhQs

L1

EMI

description

com sabem a qui representa aquest nombre? (.) què hauríem de fer? complement A1 i sumem 1 (.) vale? [How do we know what the number represents? (.) What should we do? Complement A1 and sum 1 (.) ok] llavors aquest nombre per nosaltres quin nombre és? doncs aquest és el nombre 202 [So this number what is it for us? So this is number 202?]

okay (3.) so for the if part (2.) how do my compiler implement this if part? (.) so: first of all (.) the if clause needs to be evaluated…

fact

where is the sign bit? the sign bit is gonna be (.) right here

Note Micro RhQs are in bold and italics, the lecturer’s answers in italics and translations in square brackets 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0

4.61

2.21

1.95 1.46

0.73 0.13 0.08 metacognitive

0.32 0.16

0.19 reason

explanation

description

Discourse Function L1

EMI

Fig. 2.2 Micro RhQs in L1 and EMI per 1000 words

fact

30

S. Khan and M. Aguilar-Pérez

Lecturer’s Stimulated Recall As to the findings of the stimulated recall exercise, the lecturer highlighted that his method of transmitting knowledge was motivated both by his drive to motivate students and to push them to think. He reportedly did this by “constructing bases”, explaining little by little the complex content. As a student himself, the lecturer found it hard to study something if he did not know why it had to be studied, hence his insistence on helping students understand the reason behind everything. When the lecturer’s memory was stimulated by showing him first the video and then the transcription of the video excerpts, it turned out that he was not conscious of posing so many rhetorical questions, not even after hearing them on the video and reading the transcripts. Eventually acknowledging the rich use of RhQs he made, he again said that his frequent use of RhQs arose from his need to motivate students and help them understand, as these questions allowed him to push them to think. Anticipating the students’ problems, the questions enabled him to: #10 (...) make them click, so this continuous interaction with questions is because I want to make them think and answer the questions that I’m asking. (...) That’s how I construct knowledge. I know what is difficult for them, so I guide them so they don’t get lost (...) Apart from confirming the interactional role of RhQs, which in this particular context is impregnated with this urge to be didactic and motivate students, the lecturer also acknowledges using these questions related to his teaching practice as well, his way of cognitively presenting prior knowledge on which to build new knowledge: #11 I know that some of them don’t dare ask in class so with the questions I’m also leading them to previous lectures and courses, previous knowledge they need to build on the present explanation (...)

2 So What Do We Have Here? An Engineering …

31

Therefore, he partly engages with the students out of empathy when he puts himself in the student’s shoes, when he thinks something is particularly complex and he needs to break it down for them, leading them back to the basic concept and then building it up again: #12 I often think of questions that they may ask themselves. Why? Because I’ve asked myself the very same questions. When there’s something not very intuitive, something that’s hard to understand, then possibly I throw out that question. Why is this like this? What for? Shouldn’t it be like this? Why do we find this? Why is it really like this? So it’s to motivate them, to emphasize something they themselves might be asking. Finally, when we made salient the inclusive use of ‘we’ in his questions, he nodded saying that “of course, we construct knowledge, during those 90 minutes I’m teaching and they’re learning, we’re all studying and doing the same thing ”. In brief, the stimulated recall actually confirmed that the personal metadiscourse in his RhQs rendered his lessons somehow more dialogic (Fortanet, 2004; Zare & Tavakoli, 2016), while paradoxically remaining essentially monologic, and that it was through this teacherinitiated and teacher-answered interaction that he managed to lay out concepts step by step so that students understood and acquired knowledge in its full meaning at those junctures where students would stumble and ponder what, how or why a certain action was performed.

Discussion This study aimed to analyze the use of RhQs in engineering lecturing and look into their metadiscursive functions in an attempt to shed light on the so far rather obliterated role RhQs may play in lecturing in any language. In terms of the first research question (Are there differences in the use of RhQs between L1 and EMI? If so, which kind of differences? ), we found that the EMI lecturer, highly rated by his students and colleagues, uses fewer RhQs in English than in his L1. Additionally, it was also found that in EMI the drop in macro RhQs amounts to approximately 40% and

32

S. Khan and M. Aguilar-Pérez

to under 60% in micro RhQs. This may be due to his confidence in elaborating on his explanations more explicitly when speaking in his mother tongue, even though his use of RhQs is salient across both languages of instruction, mirroring his personal, deep-seated teaching style. While he is reportedly unaware of his teaching practices, which he seems to have internalized after his teaching experience, he is however very articulate when accounting for his underlying motivations: his empathy and his urge to push students to think and understand, all of them pedagogical reasons. Distinguishing macro from micro RhQs has also proved useful to dissect RhQs because it has yielded information about the two broad metadiscursive functions of organizing discourse (signaling topic shift) and engaging with the audience. The role of micro RhQs thus stands out as particularly important in scaffolding knowledge stepwise while macro RhQs guide students and facilitate information processing as the lecturer introduces a topic with a RhQ, and he then develops the topic using micro RhQs pointing out key facts and giving explanations and reasons. When faced with his videotaped lessons, the lecturer himself attributes this greater use of RhQs to the language of instruction, suggesting that language does play a role, as RhQs are more linguistically and cognitively demanding. However, we should not forget that his EMI students were very satisfied with his lessons, hinting at the idea that, even if used in lower quantities, his good practices are somehow retained when shifting language, since his use of RhQs is conducive to the EMI students’ perception of “clear and organised explanations” (Aguilar-Pérez & Arnó-Macià, 2020). As to the second question of what functions the RhQs have, using Dalton-Puffer’s taxonomy from the ‘answer perspective’ has allowed us to pin down the abundant amount of micro RhQs, all of them used as an instructional practice to transmit complex knowledge, answering questions that students may or should have posed. It is worth noticing how the use of RhQs enables the lecturer to deal with complexity and make explicit his students’ anticipated thought processes and reasoning in carrying out the steps he is illustrating (what, how and why he is doing it), thus giving voice to his students’ doubts. Without abandoning the monologue, he is able to leave the apparently egocentric standpoint inherent in any monologue by taking students to a less unidirectional

2 So What Do We Have Here? An Engineering …

33

channel, a more dicentric orientation (Sazdovska, 2009) that allows him to engage with students while maintaining their attention with questions pronounced in rising intonation. The paradoxical dialogic monologue pervading his lectures, created thanks to RhQs, emerges as a very effective teaching practice because of the multiple benefits obtained. The metadiscursive RhQs tie together important functions of discourse: they organize the lecturer’s discourse, signaling every macro or micro transition or topic shift (textual function) while empathizing with students’ difficulties and doubts in places where he anticipates students may stumble over (interactional function). The underlying didactic outcome of breaking down complex knowledge is therefore mediated by these RhQs, which, like stepping stones, gradually pave the ground for the cognitive processing work necessary to deconstruct and subsequently construct knowledge. In this way, RhQs can be regarded as efficient resources to make information explicit and accessible to students, so necessary in lecturing discourse (Flowerdew & Miller, 1997, p. 38) and most importantly, to show a student-oriented attitude on the part of the lecturer. This resource seems to be quite easily transferable from L1 to EMI, given that, even if used less frequently in EMI, the benefits created by RhQs are not wasted. It could therefore be stated that the use of RhQs seems to derive from the lecturer’s motive—good teaching, irrespective of language of instruction—rather than from any attempt to scaffold or integrate language learning. In spite of the limitations of the study—a small sample size and lack of student feedback on the teacher’s use of RhQs—the framework of the study used for the analysis of RhQs in lecturing seems to merit further attention. The validity of our findings should also be confirmed across different disciplines and cultures (Maíz-Arévalo, 2017). However, considering the unconscious adoption of the lecturer’s use of RhQs, the implications for teacher training are clear. Making the potentially metadiscursive and cognitive benefits of RhQs salient to lecturers and putting them in the spotlight in teacher training can therefore be a very efficient technique toward improving lecturing quality.

34

S. Khan and M. Aguilar-Pérez

Appendix 1: Transcription Conventions The following conventions were included in transcriptions: Code

Meaning

(.)

Natural pause between units of speech, (2.) - 2 second pause, (3.) – 3 second pause, etc. Lengthening of word/syllable Lecturer writes on board Interrupted utterance Rising intonation, as in a question Indecipherable speech

: WB … ? (xx)

References Ädel, A. (2010). Just to give you kind of a map of where we are going: A taxonomy of metadiscourse in spoken and written academic English. Nordic Journal of English Studies, 9 (2), 69–97. Aguilar, M. (2004). The peer seminar, a spoken research process genre. Journal of English for Academic Purposes, 3, 55–72. Aguilar, M. (2018). Integrating intercultural competence in ESP and EMI: From theory to practice. ESP Today, 6 , 25–43. Aguilar, M., & Arnó-Macià, E. (2002). Metadiscourse in lecture comprehension: Does it really help foreign language learners? Atlantis, 14, 7–21. Aguilar, M., & Khan, S. (2022). Metadiscourse use when shifting from L1 to EMI lecturing: Implications for teacher training. Journal of Innovation in Teaching and Language Learning. https://doi.org/10.1080/17501229.2022. 2057503 Aguilar-Pérez, M. (2021). EMI lecturers’ and students’ perceptions: Can EMI contribute to enhancing intercultural competence? In M. L. Carrió-Pastor & B. Bellés-Fortuño (Eds.), Teaching language and content in multicultural and multilingual classrooms (pp. 65–96). Palgrave. Aguilar-Pérez, M., & Arnó-Macià, E. (2020). ‘He’s a good lecturer in any language’: Shifting from L1 to English and implications for EMI training. In M. Mar Sánchez-Pérez (Ed.), Teacher Training for English-Medium Instruction in Higher Education (pp. 153–178). IGI.

2 So What Do We Have Here? An Engineering …

35

Arnó-Macià, E., & Aguilar-Pérez, M. (2019). Developing Language and Intercultural Skills through an IaH Programme at University. Studia UBB Philologia, LXIV , 17–34. Arnó-Macià, E., Aguilar-Pérez, M., & Tatzl, D. (2020). Engineering students’ perceptions of ESP courses as a gateway to academic and professional communication. Journal of English for Specific Purposes, 58, 58–74. Arkın, E., & Osam, N. (2015). English-medium higher education: A case study in a Turkish university context. In S. Dimova, A. Hultgren, & C. Jensen (Eds.), English-Medium Instruction in European Higher Education. English in Europe (pp. 177–200). de Gruyter. Bamford, J. (2005). Interactivity in academic lectures: The role of questions and answers. In J. Bamford & M. Bondi (Eds.), Dialogue within Discourse Communities: Metadiscursive perspectives on academic genres (pp. 123–145). de Gruyter. Biezma, M., & Rawlins, K. (2017). Rhetorical questions: Severing asking from questioning. In D. Burgdorf, J. Collard, S. Maspong, & B. Stefánsdóttir (Eds), Proceedings of SALT, 27 , 302–322. https://doi.org/10.3765/salt.v27i0. 4155 Chaudron, C., & Richards, J. C. (1986). The effect of discourse markers on the comprehension of lectures. Applied Linguistics, 7 (2), 113–127. Costa, F., & Mariotti, C. (2017). Differences in content presentation and learning outcomes in English-medium instruction (EMI) vs. Italian-medium instruction (IMI) contexts. In J. Valkle & R. Wilkinson (Eds.), Integrating language and content in higher education (pp. 187–204). Peter Lang. Crawford-Camiciottoli, B. (2004). Interactive discourse structuring in L2 guest lectures: Some insights from a comparative corpus-based study. English for Specific Purposes, 3, 39–54. Dafouz, E., & Núñez, B. (2010). Metadiscursive devices in university lectures: A contrastive analysis of L1 and L2 teacher performance. In C. DaltonPuffer, T. Nikula, & U. Smit (Eds.), Language use and language learning in CLIL classrooms (pp. 213–232). John Benjamins. Dafouz, E., & Sánchez-Garcia, D. (2013). ‘Does everybody understand?’ Teacher questions across disciplines in English-mediated university lectures: An exploratory study. Language Value, 5, 129–151. Dalton-Puffer, C. (2007). Discourse in content and language integrated learning (CLIL) classrooms. John Benjamins. Doiz, A., & Lasagabaster, D. (2022). Looking into English-medium instruction teachers’ metadiscourse: An ELF perspective. System, 105, 102730. https:// doi.org/10.1016/j.system.2022.102730

36

S. Khan and M. Aguilar-Pérez

Flowerdew, J., & Miller, L. (1997). The teaching of academic listening comprehension and the question of authenticity. Journal of English for Specific Purposes, 16 , 27–46. https://doi.org/10.1016/S0889-4906(96)00030-0 Fortanet, I. (2004). The use of ‘we’ in university lectures. Reference and function. English for Specific Purposes, 23, 45–66. Hu, G., & Duan, Y. (2019). Questioning and responding in the classroom: A cross-disciplinary study of the effects of instructional mediums in academic subjects at a Chinese university. International Journal of Bilingual Education and Bilingualism, 22, 303–321. https://doi.org/10.1080/13670050.2018. 1493084 Hyland, K. (2005). Metadiscourse. Continuum. Khan, S. (2018). Lecturing strategies of non-native EMI lecturers on an International Business programme. Bulletin VALS-ASLA, 107 , 65–82. Lee, J. J., & Subtirelu, N. C. (2015). Metadiscourse in the classroom: A comparative analysis of EAP lessons and university lectures. English for Specific Purposes, 37 , 52–62. Magnuczné-Godó, A. (2011). Are you with me? A metadiscursive analysis of interactive strategies in college students’ course presentations. International Journal of English Studies, 6 , 55–78. Maíz-Arévalo, C. (2017). Questions in English as a medium of instruction versus non-English as a medium of instruction lectures. GiST-Education and Learning Research Journal, 14, 6–31. Molino, A. (2018). What I’m Speaking is almost English...’: A corpus-based study of metadiscourse in English-medium lectures at an Italian University. Educational Sciences: Theory and Practice, 18, 935–956. Morell, T. (2004). Interactive lecture discourse for university EFL students. English for Specific Purposes, 23(3), 325–338. Nattinger, J. R., & DeCarrico, J. S. (1992). Lexical phrases and language teaching. Oxford University Press. Navaz, A. M. M. (2020). Questions in english medium Instruction undergraduate lectures in a Sri Lankan university: Why are they important? International Journal of Learning, Teaching and Educational Research, 19 (12), 208–229. Querol-Julian, M. (2008). The role of questions in English academic lectures. Working Papers in Linguistics, 14, 100–111. Rigol-Verdejo, R., & Sancho-Guinda, C. (2015). Questioning questions: Remodelling oral presentations in the ESP Classroom. Procedia, 173, 132–137.

2 So What Do We Have Here? An Engineering …

37

Sánchez-García, D. (2010). Classroom interaction in university questions: The case of questions in three disciplines [Master thesis]. Universidad Complutense de Madrid. http://portal.ucm.es/web/masteres-filologia/mas ter-en-linguistica-inglesa Sánchez-García, D. (2018). Teacher questioning: Exploring student interaction and cognitive engagement in Spanish and EMI university lectures. Monográfico, III , 103–120. Sánchez-Garcia, D. (2020). Mapping lecturer questions and their pedagogical goals in Spanish- and English-medium instruction. Journal of Immersion and Content-Based Language Education, 8, 28–52. https://doi.org/10.1075/jicb. 18016.san Sazdovska, J. (2009). The intentionality model of presentations [PhD dissertation]. ELTE-SEAS. Suviniitty, J. (2012). Lectures in English as a lingua franca-interactional features [PhD dissertation]. University of Helsinki. Thompson, S. (1998). Why ask questions in monologue? Language choice at work in scientific and linguistic talk. In S. Hunston (Ed.), Language at Work (pp. 137–150). Multilingual Matters. Thøgersen, J., & Airey, J. (2011). Lecturing undergraduate science in Danish and in English: A comparison of speaking rate and rhetorical style. English for Specific Purposes, 30, 209–221. https://doi.org/10.1016/j.esp. 2011.01.002 Zare, J., & Tavakoli, M. (2016). The use of personal metadiscourse over the monologic and dialogic modes of academic speech. Discourse Processes, 54, 163–175. Zhang, L., & Lo, Y. Y. (2021). EMI teachers’ use of interactive metadiscourse in lecture organization and knowledge construction. In D. Lasagabaster & A. Doiz (Eds.), Language use in English-medium instruction at university: International perspectives on teacher practice (pp. 56–79). Routledge.

3 Distribution Patterns of Stance Features in English and Russian Conference Presentations G. Gatiyatullina, M. Solnyskina, R. Kupriyanov, and E. Gafiyatova

Introduction Academic discourse is known as “a privileged form of argument in the modern world, offering a model of rationality and detached reasoning” (Hyland, 2008, p. 2). The latter does not exclude the need to establish contact of the researcher with the audience as well as to generate interest in the discourse or theme being discussed. To express attitude, provide evidence, clarify ideas, and guide receivers’ perception of a discourse, presenters, and writers employ numerous devices which we refer to as metadiscourse markers (henceforth MDMs) (Hyland, 2005).

G. Gatiyatullina (B) · M. Solnyskina · R. Kupriyanov · E. Gafiyatova Kazan Federal University, Kazan, Russia e-mail: [email protected] M. Solnyskina e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 B. Bellès-Fortuño et al. (eds.), New Trends on Metadiscourse, https://doi.org/10.1007/978-3-031-36690-1_3

39

40

G. Gatiyatullina et al.

In this study, which is a part of a research project aimed at describing Russian academic discourse features, we examine how Russian and American speakers create rapport and facilitate dialogues with the audience. More specifically, we investigate frequency of interactional metadiscourse markers in medical academic conference oral presentations (henceforth CPs) and possible (dis)similarities in the use and frequency of MDMs in CPs of American and Russian presenters. We also explore genre variations and contrast our findings on stance features in American academic conference presentations with those in three-minute thesis presentations. We address two research questions: RQ1: What are (dis)similarities in the use and frequency of metadiscourse markers in medical conference presentations of American and Russian presenters? RQ2: What are genre variations of stance features in academic conference presentations and thesis presentations?

Literature Review Academic Conference Presentation as a Genre Academic conference presentations as a genre have received less attention than written discourse (Bellozzi, 2014; Carter-Thomas & RowleyJolivet, 2003; Charles & Ventola, 2002; Chen, 2011; Yang, 2014) not only due to its dialogic nature (Kaur & Mohamad Ali, 2017) but mostly because collecting spoken corpus is an exceptionally tedious and quite often unrewarding occupation which involves audio or video recording, transcribing, and annotating. We view a CP as a communicative event taking place in a certain socio-cultural context and performing specific communicative functions (Swales, 1990). As Hyland (2008, p. 3) noted: “… academics don’t just produce texts that plausibly represent an external reality. they use language to acknowledge, construct and negotiate social

3 Distribution Patterns of Stance Features in English …

41

relations”. Although presenters have to adapt their high-density informational content to temporal, technological, and audience constraints, their purpose is not only informative but also rhetorical. And the relationship with physically (or online) present audience is important (Carter-Thomas & Rowley-Jolivet, 2003). CPs are usually well-prepared, and sometimes rehearsed, live events in real-time. The area of the current research, i.e., patterns and markers of interaction between an addresser (a writer or a speaker) and addressees (readers or listeners), has an over 30-year history and was once defined by Hyland (2008, p. 5) as “a heavily populated area of research”. Although the object itself has been referred to with a number of terms (see Hyland, 2008), its focus is always on pragmatic functions of natural discourse and the ways interaction between an author (speaker) and recipients is conducted. Researchers emphasize that in academic discourse the task of an addresser is of a dual character: he (she) has to carry the message and ensure that the message reaches the addressees. The latter may imply that the author adjusts his message thus aligning himself with the audience, and addressees are expected to be involved in the dialogue and respond to the discourse. It is also predominantly accepted that this dialogue between authors/presenters and readers/listeners establishes, on the one hand, significance and ingenuity of the research conducted, utilizing some rhetorical conventions, meeting audience expectations and, on the other hand, possible objections of the audience as well as a certain level of background and professional (special) knowledge. In any case, as Hyland (2008, p. 5) puts it: “All this is done within the broad constraints of disciplinary discourses”, thus emphasizing that metadiscourse is culture dependent.

Metadiscourse Metadiscourse is defined as “writing about writing” (Williams, 1981, p. 40), “communication about communication” (Kopple, 1985, p. 83), and “discourse about discourse” (Hyland, 1998, p. 437). The function of metadiscourse “is essentially evaluative and engaging, expressing solidarity, anticipating objections and responding to an imagined dialogue

42

G. Gatiyatullina et al.

with others” (Hyland, 2005, p. 49). The idea behind it is that metadiscourse reveals the way the text is constructed by the two: writer and reader, speaker and listener (see Hyland, 2005, p. 49). Two different research traditions have been developed to define metadiscourse itself, i.e., integrative (interactive) or broad definition and reflexive or narrow (Ädel, 2006, pp. 167–179; Ebrahimi, 2015). The differences lie in what is viewed fundamental to the category: representatives of integrative approach consider textual interaction to be the most important, while proponents of the reflexive definition consider it to be reflexivity (Ädel & Mauranen, 2010). In addition to the definitional traditions, there are two practices of retrieving metadiscourse markers: thin or quantitative and thick or qualitative (Ädel & Mauranen, 2010). The first enables to retrieve all occurrences of a pre-defined list of metadiscourse markers (see e.g., in TextInspector), while the “thick” practice premises that words used in some texts as metadiscourse markers can be highly ambiguous and pre-defined lists cannot always be reliable (Vassileva, 1998). Hyland’s model of MDMs comprises two levels of metadiscourse: interactive and interactional. The first guides the reader through the text (Thompson, 2001, p. 58), by organizing discourse in accordance with the writer’s anticipation of the reader’s knowledge and the assessment of what the reader can recover from the text. Interactive resources include such categories as code glosses, transitional markers, frame markers, endophoric markers, and evidential markers. Interactional resources involve the reader collaboratively in the development of the text (Thompson, 2001, p. 58). Interactional metadiscourse involves the reader in the argument and employs linguistic resources to “comment on and evaluate material” (Hyland, 2005, p. 44). Interactional markers comprise hedges, boosters, attitude markers, engagement markers and self-mentions (Hyland & Zou, 2021). Hyland (2008) groups interactional macro-functions and their realizations into stance and engagement. While stance refers to the “textual ‘voice’” of an author, engagement fulfills an alignment function and addresses the ways interlocutors are explicitly recognized by the authors (Hyland, 2008, p. 5). Stance markers include hedges, boosters, attitude

3 Distribution Patterns of Stance Features in English …

43

markers, and self-mention. Hedges are defined as devices which withhold complete commitment to a proposition, allowing information to be presented as an opinion rather than fact (Hyland, 1998) (possible, might, perhaps). Boosters are “words such as clearly, obviously and demonstrate, which allow writers to close down alternatives, head off conflicting views and express their certainty in what they say and solidarity with readers”. Attitude markers reveal the author’s “affective attitude to propositions, conveying surprise, agreement, importance, frustration” and comprise the following groups: attitude verbs (agree, prefer ), sentence adverbs (unfortunately, hopefully), and adjectives (appropriate, logical, remarkable). Self-Mention refers to the degree of explicit author presence in the text (I, me, mine, exclusive we, our, ours) (Hyland, 1998, pp. 108–200; 2005, pp. 52–53). As it was mentioned earlier, numerous investigations have been conducted to examine English metadiscourse (see the references above), while Russian metadiscourse resources used in academic spoken discourse as well as (dis)similarities of MDMs usage in English and Russian academic spoken genres is a grossly under-researched area. To the best of our knowledge, few studies conducted in the area were mostly focused on developing taxonomy of Russian MDMs (Namsaraev, 1997; Viktorova, 2014, 2016). Khoutyz (2015) conducted a comparative study of Russian and English MDMs in research articles (RAs) in linguistics and communication theory. The study indicated that English RAs’ authors use significantly more reader-inclusive strategies establishing interpersonal connections with readers, while Russian authors endow readers with a much less active role (Khoutyz, 2015).

Methods In the present study, we employ a corpus-driven approach and focus exclusively on lexical patterns that mark stance. The corpus gives us information about the frequency of MDMs, the ways they are employed, it also demonstrates a comparative cross-section of English and Russian preferences in academic communication. The research is framed by

44

G. Gatiyatullina et al.

Hyland’s approach to academic interaction and the model of stance which includes four types of MDMs, i.e., boosters, hedges, attitude markers, and self-mention (see Table 3.2). The latter are used “to involve potential readers in the text, to make the writer’s views explicit and allow them to respond to the unfolding texts” (Hyland, 2005; Hyland & Tse, 2004).

Analysis The analysis was conducted in 5 stages and included the following:

Stage A. Data Collection On stage A, we compiled a corpus of conference presentations (CoConPres) delivered at English and Russian cell therapy conferences (https://www.youtube.com/channel/UC9hkRyvnE_zEWWtSfqJ8 EJQ/videos), held in 2015 and 2016 (www.youtube.com on NIH [gen ome.gov], Cell and Gene Therapy Conference and UniverTV, FutureBiotech, and RusOncoWeb video channels). We specifically selected presentations delivered by native experienced holders of medical degrees in cell therapy with at least 15 years of experience in the area thus providing consistency of the data and domain under study. We also recorded only prepared speeches no shorter than 20–25 minutes. Before recording live presentations, we talked presenters through aims and procedures of the study, asked them to sign and obtained Recording Consent forms. We also informed participants that we would anonymize the transcripts before anyone receives access to their contributions. Video and audio recordings were later converted to text files and transcribed according to TEI rules (https://tei-c.org/guidelines/). The total size of the corpus is 43667 tokens: 24097 tokens in English subcorpus and 19570 tokens in Russian subcorpus. It comprises 2110 English and 1127 Russian MDSs. The details of the corpus are presented in Table 3.1.

3 Distribution Patterns of Stance Features in English …

45

Table 3.1 Size of CoConPres (min, tokens) Pres 1 Pres 2 Pres 3 Pres 4 Pres 5 Pres 6 Total

English

Russian

E E E E E E E

R R R R R R 2

22M 3348 tokens 29M 3895 tokens 35M 5142 tokens 30M 3825 tokens 24M 3272 tokens 31M 4615 tokens 2 hours 51 min 24097 tokens

31M 3873 tokens 26M 2431 tokens 33M 3017 tokens 28M 3281 tokens 29M 3843 tokens 27M 3125 tokens hours 54 min 19570 tokens

Table 3.2 Size of CoConPres moves by language (tokens) Moves Introduction Literature review Results Conclusion Total

Size of Eng subcorpus

Average EngCP

706 6535

117 2.91% 390 2.02% 65 2.02% 1089 27.12% 11956 62.04% 1992 62.04%

2.93% 2.2%

Size of Rus subcorpus

Average RusCP

14965 62.10% 2494 62.10% 6383 33.12% 1062 33.07% 1891 7.85% 472 11.75% 841 4.36% 140 4.36% 24097 100% 4016 100% 19570 100% 3211 100%

Stage B. Structural Models of CPs Although as Thompson (2003) put it “speakers very rarely use structuring metadiscourse markers to signal the transition from one section of the presentation to the other” we follow a traditional structural model for scientific conference presentation which includes Introduction, Methods, Results, Conclusion. On Stage B, we analyzed structure patterns of the presentations to define four parts or “moves”, i.e., Introduction, Literature Review, Analysis, Results and Conclusion, and the incidence of MDMs in each of them (see Table 3.2). With regard to the length, the conference presentations are heterogeneous within English samples being longer in Results and Conclusion parts while Russian CPs are much longer in Literature Review parts (see average metrics in Table 3.2). As we can see, Russian CPs are significantly shorter (in tokens) which cannot be explained by systemic differences only, i.e., English analytism vs Russian synthetism. Although a typical Russian sentence in an academic discourse is about 1.5 shorter than an English sentence

46

G. Gatiyatullina et al.

(see Solovyev et al., 2018), it cannot be the reason of striking differences in the lengths of average Literature Review (E27.12% vs R 62.04%) and Results (E62.10% vs R33.07%). Russian Analysis and Results are also significantly shorter than Literature Review which may signify that Russian presenters tend to refer to more studies as theoretical and methodological framework of their research than English presenters. We view the differences as culture dependent which does not diminish our aim to find linguistic causes of the findings.

Stage C. Compilation On Stage C, we compiled a pre-determined MDMs lists which comprise 240 MDMs in English and 114 MDMs in Russian (see Table 3.6 in Appendix) using Hyland (2005) as a source for MDMs in English and Vinogradov (1947) and Viktorova (2016) as sources for MDMs in Russian. To extract MDMs from CoConPres, we searched English and Russian subcorpora for the markers and their frequencies with the help of AntConc (Anthony, 2018). We also classified them into four types: hedges, boosters, attitude markers and self-mention. Below we provide one example of each MDM and refer readers to complete CoConPresRu and CoConPresEn uploaded on Text Analytics lab website (https:// kpfu.ru/philology-culture/struktura-instituta/otdelenie-russkoj-i-zarube zhnoj-filologii-imeni/kafedra-inostrannih-yazikov/nil-39intellektualnyetehnologii-upravleniya/issledovaniya). For copyright purposes, we also jumbled sentences in each move of CPs.

Hedges CoConPres comprises texts with three types of hedges, i.e., downtoners (nemnogo ( a little bit), postarayus ( I will try), odna iz pervyh rabot ( one of the first works)); rounders which are associated with approximators (primerno ( approximately), pochti ( almost), poryadka ( about)), and plausibility hedges (might, may, veroyatno ( probably), navernoye ( maybe), vozmozhno ( possibly)).

3 Distribution Patterns of Stance Features in English …

47

E.g. [E31 ] there’s a tiny little bit of nuclear beta-catenin but most of it is absent and we analyzed it that in much more detail in that paper. [R44] kogda mozhno iz prakticheski lyuboy somaticheskoy kletki, to yest’ insulinorezistentnoy kletki patsiyent poluchayet kletki, napominayushchiye embrional’nyye stvolovyye kletki, t ye eto predupotentnyye stvolovyye kletki, kotoryye prevrashchayutsya v fakticheski lyubiye tkani kletki. when it is possible, from almost any somatic cell, that is, differentiated patient cells to receive cells resembling embryonic stem cells, that is, these are pre-potent stem cells that turn into virtually any cell tissue.

Boosters We discriminated three types of boosters in CoConPres: intensity markers (extremely, absolutely, must; ochen (very), imenno (exactly), dolzhen (must)); extremity markers (the highest, the greatest ; vnaibolshiy (the largest), naimenshiy (the least ); certainty markers (of course, no doubt, evidently, define, prove, show; konechno (of course), imenno (exactly), na samom dele (in fact), deystvitelno (indeed)). Intensity markers are relatively few in conference presentations under study, once used they are profoundly accompanied by hedges: E.g. [E12] so now your reporter is telling you how much your therapeutic gene is being expressed now obviously you’re going to want your reporter not to interfere with your therapeutic gene.

The most frequent intensity marker in English subcorpus is “only”, used 5–13 times in every CP: E.g. [14] I think the mark analysis would be more important where you carefully select you only the cells that have all the markers you want for the cells you transplant this was only like a biological read out here.

Interestingly, the speakers tend not to use extremity markers in their presentations. The research findings indicate that they are few and are used in presenting methods implemented by previous researchers.

48

G. Gatiyatullina et al.

However, instead of using superlatives the speakers tend to use the word “very”. E.g. [E17] yeah the ultimate goal probably would be to get differentiation of your whole culture to a degree that you have a very high percentage of cardiomyocytes.

Certainty markers are relatively frequent. E.g. [E25] bodies you get the formation of cardiac fossa ah meaning the cells actually start beating so they form cardiac full side that actually have electrical activity muscle that actually beat. [R47] I tut bezuslovno, biotekhnologi vidyat vozmozhnoye pole dlya sozdaniya biofarmatsevticheskikh sredstv, kotoryye by aktivizirovali krovoobrashcheniye Bezuslovno, osnovnym faktorom, kotoryy predopredelyayet razvitiye krovenosnoy seti vo vzroslom organizme v embrionalnom plotnom periode pokhozhe, no neskol’ko inache. And here, of course, biotechnologists see a possible field for the creation of biopharmaceuticals that would activate blood circulation. Undoubtedly , the main factor that determines the development of the circulatory network in an adult organism in the embryonic dense period is similar, but somewhat different.

Another tendency we observe in CPs is speakers’ tendency to resort to clefts to boost the ideas presented. E.g. [E28] experiments what we did we cultured the cells as embryo bodies for 12 days and from day 0 2 day 12 we digested with chondroitin sabc we supplied new chondroitin is ABC enzyme every two days to make sure the whatever controlling sulfates are expressed it’ll get digested and eliminated continuously macroscopically well actually microscopically but not on a single cell level there was not much of a difference so minus and plus.

Most of the boosters are used in Results or while juxtapositioning the expected and obtained results. Another finding is the use of the word “pretty” as an intensity marker. The studies show individual tendencies of certain speakers to use “pretty” in Literature Review and Results.

3 Distribution Patterns of Stance Features in English …

49

[E32] if you look at the correlation you can see that correlation is really pretty darn nice so what you see in vivo right what you see in vivo predicts what you see.

Attitude markers convey the following meanings: assessment (acuity, novelty, validity, quality, interestingness); effektivniy (effective), nedostoverniy (unreliable); significance (major, key, significant ); (osnovnoy (major), glavniy (main), samiy perviy (the very first), emotion (strange, surprisingly, look forward to, gladly, kindly, unfortunately, hopefully; udivilo (surprised), ogromniy (enormous), krasivo (beautiful)). E.g.: [E34] and unfortunately if you’re a physician you know that patients tend to have more than one disease. [E35] I want to thank you for your attention the work has been done at N University like we said people in the lab [name + surname] did a good amount of this work. Well, first of all, it’s not a fairy tale. These are not some fantastic stories. All this is, all this is happening before our eyes, and the first recorded case of gene therapy was 25 years ago, naturally, logically, for what? To correct a hereditary genetic disease

Self-mention refers to the degree of explicit author presence in the text (I, me, mine, exclusive we, our, ours) (Hyland, 1998, pp. 108–200; 2005, pp. 52–53). E.g.: [E37] okay so we made a reporter this is a promoter this is our gene of interest here’s our iris and here’s our reporter so the expression of this is linked to this okay you can see clearly now that if we use just a reporter we can see the lung tumor if we use this thing we can see the two lung tumors if we use our gene therapy our therapeutic gene alone we don’t see anything so that’s a perfect control okay. [R52] Nu, tak poluchilos, chto nasha organizatsiya stala vladeltsem patenta na plazmidnuyu konstruktsiyu. No, chto patent dolzhen rabotat, chto yemu lezhat? On dolzhen rabotat poetomu my nachali rabotat s patentom. I vot priblizitelnyye sroki pri tom, chto my malenkaya kompaniya byli. Deneg u nas ne bylo i seychas net, kstati. Well, it so happened that our organization became the owner of a patent for a plasmid construct. And a patent should work, shouldn’t it be utilized? It should work, so we started working on

50

G. Gatiyatullina et al.

the patent. And here are the approximate dates given that we were a small company. We didn’t have any money, and we still don’t.

Stage D. Comparison Following the “thick” approach (Ädel & Mauranen, 2010), after retrieving possible MDM candidates, we operated by excluding irrelevant ones and processed only those with metadiscursive senses. We also normalized the frequencies of each group of MDMs (boosters, hedges, attitude markers, self-mention) to 1000 tokens thus allowing further cross-language comparison (see Table 3.3). The findings point to preferences in the ways speakers of different languages use DMDs in their presentations, the ways they construct their professional discourse and dialogues with the audience. We performed a comparative analysis to reveal cross-linguistic differences in MDMs incidence. The chi-square test (chi-squared = 266.42, df = 3, p-value