A Linguistic Analysis of Peer-review Critique in Four Modes of Computer-mediated Communication

(1)

A Linguistic Analysis of Peer-review Critique in Four Modes of Computer- mediated Communication

Irina Frisk

Faculty of Human Sciences

Thesis for Doctoral degree in English Mid Sweden University

Sundsvall, February 12, 2016

(2)

Akademisk avhandling som med tillstånd av Mittuniversitetet i Sundsvall framläggs till offentlig granskning för avläggande av filosofie doktorsexamen, fredagen den 12 februari 2016, klockan 14.15 i N109 (Fälldinsalen), Mittuniversitetet, campus Sundsvall. Seminariet kommer att hållas på engelska.

A Linguistic Analysis of Peer-review Critique in Four Modes of Computer-mediated Communication

Printed by Mid Sweden University, Sundsvall ISSN: 1652-893X

ISBN: 978-91-88025-48-7 Faculty of Human Sciences

Mid Sweden University, 85170 Sundsvall Phone: +46 (0)10 142 80 00

Mid Sweden University Doctoral Thesis 236, 2016

(3)

To my parents,

Liudmila and Alexander

(4)

(5)

Abstract

Irina Frisk. A Linguistic Analysis of Peer-review Critique in Four Modes of Computer- mediated Communication. Mid-Sweden University.

The present work is a quantitative and qualitative analysis of pragmatic strategies for delivering critique, and types of politeness, used by undergraduate L2 students of English at different stages of peer-review discussion. The material examined consists of four corpora of authentic conversations between students, the main purpose of which was to give feedback on each other’s contributions during an English A-level course, at Mid-Sweden University. The conversations explored were carried out electronically, and represent four different online environments, or modes of computer-mediated communication (CMC). The material from the two asynchronous modes of CMC is comprised of L2 students’ written discussion board messages and spoken posts recorded using online software. The two synchronous environments under investigation are text-based and voice-based chat. Taking Brown and Levinson’s (1987) framework of politeness as a point of departure, the present study uses a combination of corpus and conversation analytical methods. The basic unit of analysis has been defined as the shortest message of peer-review critique that constitutes a thematic unit: these have been examined in terms of their content and politeness features associated with them, and analyzed in terms of the pragmatic strategy and type of politeness adopted. The types of pragmatic strategies or message organization patterns at different stages, i.e. initial versus subsequent feedback, of the peer-review discussion have also been analyzed. The results of the study show that the pragmatic strategies aimed at praise and agreement prevail in the corpus data produced by predominantly native speakers of Swedish. Even though the pragmatic strategies used for disagreement and negative evaluation are rich in propositional content, their occurrences and distribution vary across the four modes of CMC examined. These results seem to have wider implications in the context of online L2 learning activities, providing insights about the language of peer- review critique in a Swedish academic setting.

Keywords: computer-mediated communication (CMC), Conversation Analysis (CA), conversation management, discussion boards, feedback category, mode of CMC, peer- review discussion, politeness theory, pragmatic strategy, speech act of critique, text-based chat, type of politeness, voice-based chat, VoiceThread

(6)

iii

List of tables and figures

Tables

Table 2.1. Example strategies for positive/negative politeness (adapted from Brown and Levinson 1987:102−131) ... 7 Table 2.2. Advantages of audio discussion (ranked) (adapted from Hew and Cheung 2012a:107) ... 19 Table 3.1. Corpora of peer-review discussion ... 29 Table 3.2. Transcription conventions (based on Du Bois et al. (1993), adapted from Jenks (2011:115)) ... 31 Table 3.3. Pragmatic strategies……….34 Table 4.1. Situational variables in the Discussion Boards of McCALL (adapted from Maricic 2005:77) ... 49 Table 4.2. Raw figures and percentages of the units of critique in the Discussion Boards of McCALL by type of pragmatic strategy ... 59 Table 4.3. Raw figures and percentages of the units of critique for the top ten pragmatic strategies by type of politeness ... 63 Table 4.4. Raw figures and percentages of the peer-review messages in the Discussion Boards of McCALL by organization pattern and type of feedback ... 65 Table 5.1. Situational variables in the Text-based Chat Corpus (adapted from Maricic 2005:77) ... 80 Table 5.2. Raw figures and percentages of the units of critique in the Text-based Chat Corpus by type of pragmatic strategy ... 86 Table 5.3. Raw figures and percentages of the units of critique in the Text-based Chat Corpus by pragmatic strategy and type of politeness ... 87 Table 5.4. Raw figures and percentages of the units of critique in the Text-based Chat Corpus by pragmatic strategy and type of feedback ... 88 Table 6.1. Situational variables in the VoiceThread Discussion Corpus (adapted from Maricic 2005:77) ... 101 Table 6.2. Raw figures and percentages of the units of critique in the VoiceThread

Discussion Corpus by type of pragmatic strategy ... 109 Table 6.3. Raw figures and percentages of the units of critique in the VoiceThread

Discussion Corpus by pragmatic strategy and type of politeness ... 110 Table 6.4. Raw figures and percentages of the peer-review posts in the VoiceThread Discussion Corpus by organization pattern and type of feedback ... 112 Table 7.1. Situational variables in the Voice-based Chat Corpus (adapted from Maricic 2005:77) ... 127 Table 7.2. Raw figures and percentages of the units of critique in the Voice-based Chat Corpus by type of pragmatic strategy ... 132 Table 7.3. Raw figures and percentages of the units of critique in the Voice-based Chat Corpus by pragmatic strategy and type of politeness ... 133 Table 7.4. Raw figures and percentages of the units of critique in the Voice-based Chat Corpus by pragmatic strategy and type of feedback ... 135

(10)

vii

Table 8.1. Density of peer-review content in four modes of CMC ... 146 Table 8.2. Activity type, topic of exchanges and assignment instructions for four contexts of peer-review critique ... 150

(11)

viii Figures

Figure 3.1. Taxonomy of key terms……….………...27 Figure 4.1. Frequencies of the top ten pragmatic strategies in the Discussion Boards of McCALL by semester and team………...62 Figure 4.2. Percentages of peer-review messages in the Discussion Boards of McCALL by

organization pattern………...65

Figure 6.1. Percentages of peer-review posts in the VoiceThread Discussion Corpus by organization pattern………..…..112 Figure 8.1. Pragmatic strategies: Discussion Boards of McCALL (asynchronous written

mode of CMC)……….…148

Figure 8.2. Pragmatic strategies: Text-based Chat Corpus (synchronous written mode of

CMC)………...148

Figure 8.3. Pragmatic strategies: VoiceThread Discussion Corpus (asynchronous spoken

mode of CMC)……….149

Figure 8.4. Pragmatic strategies: Voice-based Chat Corpus (synchronous spoken mode of

CMC)………...149

(12)

ix

Acknowledgments

During the years of my doctoral studies, funded by the department of Humanities at Mid- Sweden University, I have received support and encouragement from the department.

Moreover, the extension of my doctoral position financed by the department made it possible for me to finalize the transcriptions of the material used in this thesis. My fellow doctoral students and colleagues at the department have provided a constructive learning environment. I appreciate docent Hedda Friberg-Harnesk’s help in always handling the administrative issues related to the doctoral program, and Dr. Rachel Allan’s encouraging comments on my chapter drafts. Dr. Martin Shaw’s support when I was teaching courses at the department has been invaluable. His guidance has given me new insights and tools for approaching the material analyzed in this thesis from a didactic perspective.

I would especially like to thank my supervisors. My secondary supervisor docent Mats Deutschmann (Umeå University) introduced me to computer-mediated communication as a valuable resource for studying authentic interaction. His passion and enthusiasm for language learning and teaching have been an inspiration! Docent Terry Walker stepped in as my main supervisor and helped me structure my work when I was wandering around in the corridors of linguistics, taking on random ideas and trying to get started with my own writing. Her methodical approach to work in general, in regards to both her own area of research and daily preparation for classes, was educational. With attention to detail and patience, she constantly raised the bar enough to challenge me and keep me motivated.

I am very grateful to Professor Merja Kytö for letting me join the series of higher seminars in English linguistics at Uppsala University. I would like to thank the seminar participants for providing insightful comments and suggestions for improvement on individual thesis chapters and, most importantly, at my final seminar. Furthermore, I truly appreciate the feedback I received from the participants at a MODIS seminar, arranged by the department of Modern Languages at Uppsala University, which helped me to fine-tune my arguments.

I have been given an amazing opportunity to observe and study naturally-occurring language thanks to the kind permission of the students taking undergraduate courses in English at Mid-Sweden University. Thank you all for making my teaching experience so versatile and rewarding! Last, but by no means least, I would like to thank my family and friends for bearing with me.

October 12, 2015 Sundsvall

(13)

1

Chapter 1 Introduction

1.1 Introductory remarks

The current investigation is a description and analysis of the language of peer-review critique, as represented in four different online environments, or modes of computer- mediated communication (CMC). The material examined is comprised of (a)synchronous written and spoken authentic conversations between undergraduate students taking an online English course at Mid-Sweden University, Sweden. Peer-to-peer talk still constitutes an underexplored area in linguistic research due to restricted access to authentic classroom interaction (cf. Ädel 2011; Hewett 2000). However, due to the increasing popularity of online courses, in which exchanges between students can be easily recorded and archived, naturally-occurring conversations between peers have become more readily available for research (cf. Skogs 2015; McBride and Fägersten 2010). The main aim and contribution of this study is to provide a deeper understanding of the pragmatics of the genre of peer- review discussion, i.e. student-produced feedback on each other’s work, in the context of online L2 learning activities.

In language didactics, the genre of peer-reviewing, peer assistance (Storch 2001, 2000) or peer response (Tang and Tithecott 1999), has been previously studied with regard to its potential to increase L2 learners’ motivation and engagement in the learning process (cf. Swain et al. 2002). The focus in the present study, however, is on the linguistic aspects of peer-review critique, particularly, on L2 learners’ choice of pragmatic strategies and means of maintaining politeness when providing feedback on a peer’s work. As an object of linguistic inquiry, peer-review critique has been approached both in terms of the speech act of criticism (Nguyen 2008) and the language of rapport (Ädel 2011). Moreover, within the scope of (im)politeness theory, peer-review critique has been strongly associated with academic debate, disagreement and conflict (cf. Rees-Miller 2000).

Depending on the participants’ communicative goal as well as the type of environment involved, the language of peer-review critique reveals varying degrees of (in)directness. Four examples of peer-to-peer communication are given in (1a)–(1d), all of which are instances of academic peer-review discussions and come from the material analyzed in this thesis (for details, see Chapter 3). The lexical and grammatical items associated with (in)directness of critique that constitute one of the main concerns of the current investigation (see Section 1.2) are highlighted in bold.

(1a) I think you did a great job with subject-verb agreement (Text-based Chat Corpus, SS2012: group7, l. 69)¹ [synchronous written CMC]

(1b) I’m totally on your page that main reason I also studied I .. well

1 The reference after each corpus example provides information about the name of the corpus, the semester (SS for Spring Semester and FS for Fall Semester, which correspond to the Swedish abbreviations VT and HT in the McCALL data) and year when the material was collected (see Chapters 4–7).

(14)

2

studied at distance even if I hadn’t done it by now you definitely would’ve erm erm erm what do you say .. persuaded me to do it because it sounds like it’s amazing

(VoiceThread Discussion Corpus, FS2010: V9, C3) [asynchronous spoken CMC]

(1c) at first .. I .. thought it was a little bit long .. but when I looked at it again I think the structure may fool you a little bit because it's a lot of space .. but then I realized that you should have 1.5 spaces between the rows am I right

(Voice-based Chat Corpus, SS2011: group1) [synchronous spoken CMC]

(1d) Amar didn’t understand what you meant by "demand" in 1c, and neither did I. Hopefully you can enlighten us. :)

(McCALL: Message VT06S.D.227) [asynchronous written CMC]

Clearly, the examples above differ in terms of content, politeness and context. The messages in (1a) and (1b) contain an appreciation of a peer’s work and an element of agreement with a peer’s opinion, respectively. As to the authors’ communicative styles, these examples are characterized by casual vocabulary and fairly straightforward syntax, which does not leave room for misinterpretation. By contrast, the message in (1c), expressing a negative evaluation of a peer’s work, is very indirect. The author seems to emphasize her uncertainty regarding the length of the essay she is commenting on instead of openly criticizing her peer. Interestingly, the indirectness of the peer-review message in (1d) is maintained by partially attributing the disagreement to a third party. Context-wise, the examples in (1a)–(1d) represent four different online environments (see the comment in square brackets after each corpus reference), and therefore can be expected to follow the conventions typical of a given mode of CMC. Finally, the messages in the examples above address different topics, which might have important implications for the nature of peer- review critique contained in them (for details, see Chapter 3).

In Sweden, the academic genre of peer-review discussion has undergone a number of significant changes during the past decade. This is a very common format for defending one’s Bachelor’s and Master’s theses at Swedish universities (cf. Swedish National Agency for Higher Education). With the introduction of online courses, the peer-review, or

‘opposition’, seminar has gradually evolved into a discussion assignment allowing course instructors both to monitor students’ participation and to evaluate their progress in different stages of the learning process, rather than examining the final product, i.e. an essay draft, alone (for details, see the Course Plan: English BA (A) on the Mid-Sweden University website and Chapter 3). In the absence of physical co-presence, asynchronous text and voice-based tools, such as discussion boards and VoiceThread (see Section 3.2; Chapters 4 and 6), as well as synchronous text and voice-based chat, offered by e.g. Skype (see Section

(15)

3

3.2; Chapters 5 and 7), have become increasingly popular as a means of in-course communication (see Lyddon and Sydorenko 2008). Needless to say, this shift from the real- time face-to-face examination to the virtual classroom situation has had implications for the genre of peer-review discussion (cf. Ädel 2011), for example in terms of alternative ways of expressing positive and negative evaluation, such as emoticons and CMC abbreviations.

Therefore, in order to account for the variety of lexical and grammatical means for delivering critique used by L2 students of English, one needs to consider different types of online learning environments.

As mentioned above, the present analysis of the genre of peer-reviewing involves two written and two spoken modes of CMC, both synchronous and asynchronous. The two written modes include authentic peer-review discussions between students of English at Mid-Sweden University, taking place in the Discussion Boards of the virtual learning platform WebCT/BlackBoard (asynchronous), and in the text chat window of the online conferencing software Skype (synchronous). The two spoken modes of CMC are represented by transcribed authentic conversations between students enrolled on an online English course at Mid-Sweden University, recorded by using the asynchronous net-based conferencing tool VoiceThread and the synchronous ‘conference call’ feature in Skype (for details see Section 3.2 and Chapters 4–7).

1.2 Terminology

In order to clarify the terms mentioned in the previous section, I will here briefly define them and explain how they will be used throughout this thesis. In CMC research, a mode is

“a distinct channel of communication enabled by technology and therefore includes the range of channels used in synchronous and asynchronous computer-mediated communication, such as voice chat, text chat, email, etc.” (Beers Fägersten et al. 2010:148).

Henceforth, in the present investigation, the four distinct online learning environments represented by different forms of technology, asynchronous or synchronous, and channels of communication, written versus spoken, will be referred to as modes (for details, see Section 3.2).

As the language of peer-review critique is the primary concern of this study, two frameworks of linguistic politeness constitute a point of departure for the present investigation, namely, Brown and Levinson’s (1987) face-saving model and Locher and Watts’ (2005) relational work. The term pragmatic strategy will be used to refer to the L2 learners’ choice of lexical and grammatical means for delivering critique. This term should not be confused with politeness strategies in Brown and Levinson’s (1987) framework, which view the speaker’s and the hearer’s communicative behavior as verbal actions aimed at maximizing or minimizing social distance in conversation. In this thesis, politeness strategies will be discussed in terms of types of politeness, based on Brown and Levinson’s taxonomy of positive and negative politeness strategies (see Section 2.2.1). The term pragmatic strategy, on the other hand, will be applied descriptively in order to account for the occurrence of certain lexical and grammatical items (the highlighted parts in (1a)–(1d) in Section 1.1) within the structure of authentic peer-review exchanges between the

(16)

4

participants in the material examined. Academic peer-review exchanges will thus be analyzed in terms of different pragmatic strategies based on their content, such as agreement, disagreement, positive and/or negative evaluation. Throughout the thesis, italics will be used for pragmatic strategy labels, e.g. complex disagree or positive evaluation (for details, see Section 3.4.2).

Furthermore, Austin’s (1962) and Searle’s (1969) theory of speech acts provides an important theoretical background for the present work (see Section 2.2.4). From a broader perspective, the term speech act denotes different types of verbal actions, such as “making statements, giving commands, asking questions, making promises, and so on” (Searle 1969:16). However, for a string of words uttered in a conversation to count as a speech act, they have to occur in “certain contexts, under certain conditions and with certain intentions” (Searle 1969:25). The instances of peer-to-peer talk examined in this thesis thus present an intriguing opportunity for studying the speech act of critique in authentic contexts of language use. Throughout this work, the term speech act of (peer-review) critique will be viewed as consisting of one of two equally important manifestations, i.e.

agreement and positive evaluation on the one hand, and disagreement and negative evaluation, on the other. In other words, expressions of appreciation and agreement, which have traditionally been analyzed as mitigating elements within the structure of the speech act of critique, in pragmatics research (see Section 2.2.2), will be treated as separate examples of peer-review critique.

1.3 Aims and scope

Herring (2002) and Soukup (2000) suggest that each mode of CMC should be examined in its own right in order to identify the patterns of language use typical of a certain online environment. The main aim of the present investigation is thus to provide a description and analysis of peer-review critique representative of four distinct modes of CMC, using the following research questions as a point of departure:

 What pragmatic strategies are characteristic of the peer-review critique in the (a)synchronous written and spoken modes of CMC?

 What types of politeness do the students favor?

 What are the types of pragmatic strategies and message organization patterns (see Sections 4.3 and 6.3) at different stages, i.e. initial versus subsequent feedback, of the peer-review discussion (see Section 3.4.2)?

1.4 Thesis outline

This work is organized into nine chapters. As the purpose of the current chapter was to provide an introduction to the thesis, the genre of the peer-review discussion, the modes of CMC and the theoretical approach undertaken in this study have been briefly introduced.

Moreover, the aim of this thesis, including the three research questions, has been explained.

Chapter 2 positions the current study in the field of language pragmatics, recent CMC research and the pedagogical framework underlying the design of online courses in English at Mid-Sweden University. Chapter 3 introduces the material examined, the four corpora of

(17)

5

peer-review discussion, and a step-by-step description of the methodological procedures implemented throughout this thesis.

Chapters 4–7, the results chapters, each focusing on a particular mode of CMC, are structured as follows. First, the context of the peer-review discussion is accounted for, and the analytical categories used throughout this thesis are illustrated with examples of peer- review critique from the respective corpus. Second, the results of the corpus study with regard to the three research questions specified in Section 1.3 are given. Finally, each data chapter ends with a qualitative discussion of the nature of peer-review critique associated with the mode of CMC under investigation. The focus in Chapter 4 lies on an asynchronous written mode of CMC, comprised of the Discussion Boards part of the McCALL corpus (see Deutschmann et al. 2009 and Section 3.2). Chapter 5 deals with a synchronous written mode of CMC, i.e. the Text-based Chat Corpus (see Section 3.2). Chapters 6 and 7 present an analysis of peer-review critique in two spoken environments, the asynchronous VoiceThread Discussion Corpus, and the synchronous Voice-based Chat Corpus (see Section 3.2). Chapter 8 brings up external factors that might have influenced students’

choice of pragmatic strategies. Finally, the concluding chapter, Chapter 9, briefly discusses the implications of the results of this study and suggests possible areas of future research within the field.

(18)

6

Chapter 2 Previous Research

2.1 Introduction

The overarching aim of this thesis is to provide a pragmatic analysis of peer-review critique in the context of (a)synchronous written and spoken CMC; therefore, the current chapter presents an overview of previous accounts of linguistic politeness in general (see Section 2.2.1) and the speech act of critique in particular (see Section 2.2.2). As the present study examines authentic conversations between peers, who are students on the same language course (see Chapter 3), empirical approaches to politeness are of great importance when it comes to interpreting the results of the data chapters (see Section 2.2.3). The conversation analytical methods involved in the analysis of the qualitative results of the synchronous material (see Chapter 3) will be addressed in Sections 2.2.4 and 2.2.5. The computer- mediated nature of peer-reviewing emphasizes the role of the channel, as well as form, of communication; therefore, the findings of previous CMC research with respect to these two variables will be outlined in Section 2.3. Finally, as the genre of peer-reviewing constitutes one of the recurrent practices in the context of interactive e-learning (see Berge 2002), it is important to consider the theoretical framework underlying the pedagogical setup of the online English courses at Mid-Sweden University investigated (see Section 2.4).

2.2 Linguistic politeness

2.2.1 Politeness theory

The framework of linguistic politeness, first developed by Brown and Levinson in 1978 and revised in 1987 (all subsequent references are to the latest edition), has been adopted as a point of departure for the present analysis of pragmatic strategies for peer-reviewing, because it relies on a number of well-defined concepts. A further model used in this thesis is Watts’ (1989) relational approach to politeness, which allows one to interpret the actual contexts of language use through the lens of the social norms and values of a given society (cf. Locher and Watts 2005; Watts 1992). As the material examined in this thesis comprises naturally-occurring conversations between students of English, Locher and Watts’ (2005) perspective on politeness is very relevant to the present discussion of peer-review critique.

Brown and Levinson’s (1987) and Locher and Watts’ (2005) approaches to politeness have been implemented by a number of researchers (cf. Rees-Miller 2011; 2000; Spencer-Oatey 2008; Holtgraves 1997). Therefore, this section will provide an introduction to linguistic politeness by defining and illustrating the central concepts within the field.

Developments in politeness theory since the mid-twentieth century have resulted in the emergence of new categories for describing and analyzing the mechanisms of human communicative behavior. Goffman (1967 [1955]) introduces the notion of face, which is developed further by Brown and Levinson (1987:61), who define it as “something that is emotionally invested, and that can be lost, maintained, or enhanced, and must be constantly attended to in interaction”. Brown and Levinson (1987:61) view face as “the public self- image that every member wants to claim for himself, consisting of two related aspects”,

(19)

7

negative face and positive face. By negative face they mean “the want of every ‘competent adult member’ [of society] that his actions be unimpeded by others”, whereas positive face relates to “the want of every member that his wants be desirable to at least some others”

(Brown and Levinson 1987:62). The two sides of face are further associated with two types of politeness: negative politeness, which addresses the hearer’s negative face wants and serves to maintain the distance between interlocutors; and positive politeness, which addresses the hearer’s positive face wants and helps to reduce the distance in interaction.

Brown and Levinson (1987:65) use the notion of speech act (see Section 2.2.2) to indicate

“what is intended to be done by verbal or non-verbal communication” and argue that some speech acts can be interpreted as directly damaging to the face wants of interlocutors, such as offers, promises and compliments on the one hand (threatening the addressee’s negative- face wants) and criticisms, expressions of disapproval, contradictions or disagreements, challenges, etc. (threatening the addressee’s positive-face wants) on the other (1987:65−66). Brown and Levinson thus outline two sets of strategies for positive and negative politeness; those that are relevant to the discourse of critique are listed in Table 2.1 below.

Table 2.1. Example strategies for positive/negative politeness (adapted from Brown and Levinson 1987:102−131)

Positive politeness Negative politeness

Use in-group identity markers Be conventionally indirect

Seek agreement Question, hedge

Avoid disagreement Joke

Be optimistic Be pessimistic

Offer, promise Apologize

Include both S (Speaker) and H (Hearer)

in the activity

Impersonalize S and H: Avoid pronouns ‘I’

and ‘you’

As seen from Table 2.1, the strategies associated with positive politeness are used to reinforce personal and interactional proximity, while the negative politeness strategies serve to maintain a certain distance between the speaker and the hearer in a conversation.

However, to what extent this distance should be maintained will directly depend on the nature of a Face-Threatening Act (FTA). Brown and Levinson suggest that the seriousness of an FTA can be calculated from the following equation:

W = D (S,H) + P (S,H) + R,

W stands for ‘weightiness’, or seriousness of a given FTA;

D – the social distance of S and H;

P – the relative power of S and H;

R – the absolute ranking of impositions in the particular culture

(taken from Brown and Levinson 1987:74−75).

(20)

8

In Brown and Levinson’s model, the face-threatening potential of the speech act of critique (R) depends on the two other factors, namely the social distance and power relations between the interlocutors. The more offensive an FTA is considered to be, the more mitigating devices (e.g. hedges, apologies, etc.) it requires. Applying this view to the contexts of peer-review critique examined in this thesis, one could assume that the relative power (P) factor should be a constant regardless of the mode of communication, i.e.

(a)synchronous written or spoken, since the participants are students on the same course (as opposed to students criticizing teachers or vice versa). However, the social distance (D) factor is likely to vary in the four sets of data analyzed, depending on the dynamics within peer-review groups (cf. Deutschmann and Lundmark 2008), the form, asynchronous or synchronous, and the channel of communication, written or spoken (for details, see Chapter 3 and the data chapters 4–8).

Brown and Levinson’s theory of politeness has been considered insufficient in some respects, and there have been several attempts to ‘refine’ their model. In approaching politeness from the point of view of linguistic pragmatics, Leech (1983:16) introduces a Politeness Principle (PP), equal in status to Grice’s (1975) Cooperative Principle, and consisting of a large number of maxims: Maxim of Tact, Maxim of Generosity, etc.

Culpeper (1996:349) argues that Brown and Levinson’s model fails to account for the logic behind impolite behavior and proposes a theory of impoliteness, which is “parallel but opposite to Brown and Levinson’s theory of politeness”. Fraser questions Brown and Levinson’s (1987:5) claim that “politeness has to be communicated” in the form of

“additional message […]: ‘I’m being polite here because I’m sensitive to your face needs’”

(2005:66). Instead, he sees politeness as the norm that governs one’s social behavior, and thus a matter of social rather than language competence. Like Culpeper (1996), Fraser criticizes Brown and Levinson (1987) for not dealing with the notions of

“impoliteness/rudeness/aggravation”, and plays with the idea of a “continuum with politeness at one end, impoliteness at the other, and degrees of (im)politeness along the way” (2005:70). One of the major weaknesses with Brown and Levinson’s (1987) model, Fraser (1990:235) suggests, is that there is not enough “empirical evidence that their claims about politeness strategies correlate with naturally occurring conversations”. Considering the authentic nature of the material analyzed in this thesis, it thus becomes of interest to see whether the two-dimensional taxonomy of positive and negative politeness (Brown and Levinson 1987) holds for the instances of naturally-occurring data.

The critics of Brown and Levinson’s (1987) theory of politeness argue that it does not provide a sufficient account of strategies that fall outside of individuals’ polite behavior, “dealing only with the mitigation of face-threatening acts” (Locher and Watts 2005:10). Locher and Watts’ relational work represents one of the latest extensions of Brown and Levinson’s (1987) framework, within which politeness is defined as a

“discursive concept arising out of interactants’ perceptions and judgments of their own and others’ verbal behavior” (2005:10). They argue that in order to fully understand polite realizations of a communicative event, it should be studied within a given context and across “the entire continuum of verbal behavior from direct, impolite, rude or aggressive interaction through to polite interaction, encompassing both appropriate and inappropriate

(21)

9

forms of social behavior” (Locher 2004:51, cited in Locher and Watts 2005:11). Based on Goffman (1967), Locher and Watts suggest that face is constructed through interaction (rather than acquired during one’s upbringing), “which implies that any individual may be attributed a potentially infinite number of faces” (2005:12). Moreover, they discuss the notions of “first order politeness” (politeness1) and “second order politeness” (politeness2), introduced in Watts et al. (1992), highlighting the fact that politeness1 deals with the actual contexts of language use, while politeness2 corresponds to the social norms of a given language (Locher and Watts 2005:15). When viewed out of context, some utterances may be perceived as more polite that others. However, within a certain context, the same utterances would be interpreted as appropriate rather than polite. Locher and Watts refer to this “unmarked, socially appropriate behavior as ‘politic behavior’”, thereby arguing that

“no linguistic expression can be taken to be inherently polite” (2005:16).

Rees-Miller (2011) applies Locher and Watt’s (2005) data-driven (politeness1) approach to her study on the gendered use of compliments in a contemporary university setting in the US. She considers the nature of environment as one of the key variables for an adequate interpretation of different instances of this speech act. Rees-Miller makes a distinction between unstructured settings, when “students interact with no fixed purpose”, and goal-oriented activities including different types of “organized student activities”

(2011:2679). The discursive approach allows Rees-Miller to determine the function of compliments in different contexts, drawing a line between the compliments that are expected within a given setting, or unmarked and politic, and those that are “‘marked’

either positively […] or negatively” (Rees-Miller 2011:2674). To illustrate the latter, she points out that “the absence of a compliment when an athlete had made a good play would be negatively marked as a passive-aggressive insult that excludes the athlete from solidarity with spectators or other team members” (Rees-Miller 2011:2685). In other words, Rees- Miller’s results provide further support for Locher and Watts’ (2005:26) claim that

“compliments as such are not inherently politeness markers”. Therefore, her study is highly relevant to the discussion in this thesis (see Chapters 4–7) because the speech act of compliment is closely related to the positive evaluation strategy in the context of students’

peer-review exchanges.

Studying disagreement in an academic setting in the United States, Rees-Miller (2000) examines Brown and Levinson’s (1987) strategies for positive and negative politeness from the point of view of their linguistic realization. In her taxonomy of disagreement, three types of disagreement are distinguished, softened disagreement, disagreement not softened or strengthened and aggravated disagreement, in which each type of disagreement is further linked with a specific type of linguistic marker. Rees-Miller (2000:1095) suggests that within the category of softened disagreement positive politeness is associated with the following linguistic markers: positive comments, expressions of humor, inclusive first person, and partial agreement, whereas negative politeness is expressed through questions, hedged performatives (I think, I don’t know), downtoners (maybe, sort of) and verbs of uncertainty (seems). Her findings reveal that in a synchronous face-to-face academic debate softened disagreement is the prevailing type of disagreement.

(22)

10

In examining rapport management strategies, Spencer-Oatey (2008) takes the speech act-oriented approach to politeness theory as a point of departure. She builds her discussion around the following issues:

 the selection of speech act components;

 the degree of directness-indirectness;

 the type and amount of upgraders/downgraders

(adapted from Spencer-Oatey 2008:22).

By speech act components Spencer-Oatey is referring to the semantic structure of a speech act, realized by a head act and a number of optional acts. To illustrate this, she provides an example (given below) by analyzing the speech act of request.

The speech act of request can thus be analyzed as consisting of the head act and a number of optional components, such as different types of mitigating moves. The additional components make the act of request more indirect. Spencer-Oatey’s model for examining the semantic structure of a speech act will be evaluated from the point of view of the approach adopted in this thesis and further discussed in the data chapters (see Chapters 4–

8).

Semantic structure of the speech act of request (adapted from Spencer-Oatey 2008:22) Do you mind if I ask you a big favour? Mitigating supportive move (preparator) I know you don’t like lending your car, Mitigating supportive move (disarmer) but I was wondering if I could possibly

borrow it

Head act just for an hour or so on Tuesday afternoon,

if you’re not using it then. Mitigating supportive move (imposition downgrader)

I need to take my mother to the hospital Mitigating supportive move (grounder) and it’s difficult getting there by bus.

A fundamentally different approach to the study of the speech act of disagreement within politeness theory has been proposed by Holtgraves (1997). Adopting Brown and Levinson’s (1987) distinction between the two types of politeness, Holtgraves (1997) argues that since disagreement presents a threat to the hearer’s positive face by interfering with his/her wish to be part of a group, all types of disagreements need to be redressed with positive politeness strategies. By drawing a line between hypothetical and authentic contexts of language use, Holtgraves (1997:230-233) provides a new interpretation for hedges, traditionally understood as a mechanism of negative politeness, and expands the original list of positive politeness strategies (Brown and Levinson 1987:102) with four new strategies, namely, express agreement, personalize opinion, express distaste with one’s position and self-deprecation. Holtgraves defines positive politeness in terms of “deviations

(23)

11

from maximum Gricean (Grice, 1975) efficiency […] either emphasizing the degree of agreement between the interactants or […] downplaying the extent and significance of their disagreement” (1997:236). Within this framework the contexts of disagreement are viewed from a broader, more descriptive, perspective. Moreover, Holtgraves (1997:225) claims that

“[a]greements are explicit, syntactically simple, and occupy an entire turn; disagreements are more syntactically complex, and often implicit” (for details see Section 2.2.4).

The empirical results of Rees-Miller’s (2000) and Holtgraves’ (1997) studies are highly relevant from the point of view of analysis of the pragmatic strategies for delivering critique carried out in this thesis and will be referred to in Chapters 4–7.

2.2.2 The speech act of critique

As a speech act constitutes a basic unit of analysis in politeness research (cf. Fraser 1990), it is important to review how the speech act of critique has been defined in the previous studies. Some researchers (cf. Spencer-Oatey 2008; House and Kasper 1981) view speech acts as “sequences which include not only head acts, but also supporting moves that strengthen or weaken the head act” (Sykes 2005:403). For example, Nguyen (2008:771) views the speech act of critique as comprised of the core and optional elements. In other words, in case of criticism, the ‘core’ element would be realized by the negative evaluation only, while different types of linguistic devices aimed at lessening the face-threatening potential of the negative evaluation would be considered optional. Nguyen (2008:770) defines “criticizing as an illocutionary act whose illocutionary point is to give negative evaluation of the hearer’s (H) actions, choice, words, and products for which he or she may be held responsible”. This approach is consistent with Spencer-Oatey’s (2008) claim that a certain speech act can be viewed as consisting of one head act and a number of supporting moves (see Section 2.2.1).

Nguyen (2008) studies the development of L2 learners’ pragmatic competence, based on their ability to effectively mitigate the face-threatening potential of the speech act of criticizing. Following House and Kasper (1981), she applies distributional criteria to the structure of the speech act of criticism by distinguishing between its external and internal modifiers (Nguyen 2008:771–772). Among Nguyen’s (2008:771) external modifiers, there are sweeteners (analyzed as the pragmatic strategy of positive criticism in this thesis, with one or more speech acts within its structure (see Chapters 5–8 in the present study)) and grounders (which are referred to in this thesis as a ‘motivation’ or ‘explanation’ in cases of complex agree and complex disagree). On the other hand, within the category of internal modifiers, Nguyen (2008:771–772) makes a distinction between syntactic (i.e. the use of past tense with present time reference and modals) and lexical/phrasal (i.e. hedges, downtoners, etc.) devices.

The aim of Nguyen’s (2008:768) study is to provide quantitative and qualitative analyses of different stages in L2 learners’ acquisition of the speech act of criticism by comparing three groups of participants – native speakers of Vietnamese with different levels of proficiency in the English language – beginners, intermediate and advanced. The data analyzed consists of recorded face-to-face conversations between peers, the goal of which was to give feedback on each other’s argumentative essay, commenting on its

(24)

12

structure, quality of argumentation, grammar and vocabulary (Nguyen 2008:774). As the L2 learners were mainly exposed to Australian English, the speech acts of criticism generated by them were compared to those produced by a reference group of native speakers of Australian English. Nguyen’s (2008:777) findings reveal that all learners appear to use fewer mitigating elements than the native speakers, partly due to their insufficient proficiency in the target language, and partly because they seem to rely too much on their L1 pragmatics, in which the use of modifiers in the context of peer feedback is not as conventionalized as it is in the target language (Nguyen 2008:783). Furthermore, there are substantial differences in terms of the structural complexity of modifiers employed by the three groups; namely, the use of internal modifiers increases with the learners’ level of language proficiency (Nguyen 2008:787). Even though the instances of peer-to-peer talk analyzed in the present study mainly involve Swedish students of English (see Section 3.2), Nguyen’s (2008) observations concerning the relationship between L2 learners’ level of language proficiency and their preference towards modifiers of a certain type present a point of reference for the analysis of peer-review critique pursued in this thesis.

2.2.3 Empirical evidence: disagreement and humor

According to the approach implemented throughout this thesis, the speech act of critique includes both positive and negative evaluation (see Chapter 3). However, most research within (im)politeness theory has focused on the discourse of disagreement (Bousfield 2007;

Maricic 2005; Rees-Miller 2000; Holtgraves 1997) or conflict (Graham 2007). This section reviews a number of empirical studies examining the role of humor in cases of disagreement.

Within Brown and Levinson’s (1987) politeness theory, humor represents a positive politeness strategy, “conveying or emphasizing intimacy or commonality” between participants in conversation (Holtgraves 1997:224). However, depending on the type of communicative medium and context involved, humor can be realized differently and display a number of functions ranging from friendly banter to hostile forms of teasing.

Examining the discourse strategies involved in conflict management in online asynchronous text-based discussions, Maricic (2005:110) reports that humor can either be

“successfully exploited to release intra-group tensions and decrease conflict”, or “it may also be exploited in subtle, ironic ways to achieve quite the opposite, i.e. damage the recipient’s public face”. As written language constitutes the only medium of communication available to participants in Maricic’s data, the focus in her discussion lies on emoticons, the CMC abbreviation <g> for ‘grin’ and lexico-grammatical means of conveying a humorous message, such as the use of special in-group jargon by participants (2005:118, 147). The results of Maricic’s (2005:119) study show that humor has the potential to lessen the face-threatening nature of the speech act of disagreement, thereby providing empirical evidence for Brown and Levinson’s (1987) taxonomy of politeness strategies.

Norrick and Spitz (2007) investigate the mitigating effect and functions of humor in conflict sequences in spontaneous face-to-face conversation. They conduct a qualitative

(25)

13

analysis of different scenarios of how participants employ and respond to humor and present a list of factors influencing the potential of a humorous tone to resolve conflict, which are as follows:

 the seriousness of conflict;

 the social power relationships between the participants;

 the kind of humor;

 the reactions of the participants;

 who initiates the humor

(adapted from Norrick and Spitz (2007:1661).

It is important to note that Norrick and Spitz’s (2007:1661) factors influencing the

“effectiveness of humor” closely correspond to the ‘weightiness of an FTA’ formula in Brown and Levinson’s (1987:74–75) theory of politeness, including the ‘social distance’

(D) and ‘power’ (P) of the speaker and the hearer as well as the ‘ranking of the imposition’

(R) components (see Section 2.2.1). Therefore, it can be assumed that Norrick and Spitz’s (2007:1661) model outlines a set of variables affecting the mitigating potential of humor for, in Brown and Levinson’s (1987) terms, a certain type of imposition, namely the speech act of disagreement.

Furthermore, Norrick and Spitz differentiate between humor and laughter by defining the latter as “the natural reaction to humor” (2007:1679). One of the results reported in their study is that “laughter alone – even in the absence of any clear orientation to humor – can ameliorate conflict” (Norrick and Spitz 2007:1682). Also, laughter is one of the main concerns in Habib’s (2008) analysis of the role of teasing in disagreement among peers. Unlike Straehle (1993), who viewed teasing as an aggressive form of humor, the evidence in Habib’s data suggests that “[t]he laughter [in cases of peer disagreement]

reveal[s] their strong connection and reflect[s] the coherence of the conversation, which is another indication of their existence as a cohesive group” (Habib 2008:1142). In other words, Habib’s (2008:1143) conclusion (see also Ädel 2011; Vaughan 2008; Miller 1995) is that humor works as a cohesive device binding together parts of conversation, evoking in-group jokes and positive associations, which is, in turn, indicative of interlocutors’

“pragmatic competence” (Fraser 2010:15) and “face management” skills (Brown and Levinson 1987). Naturally, the approaches to humor and laughter within the contexts of disagreement and conflict, as outlined above, are very relevant to the present thesis because they view humor as a “mechanism” (Brown and Levinson 1987) or “resource” (Norrick and Spitz 2007) downplaying or resolving tension in conversation.

2.2.4 Conversation Analysis

Conversation Analysis (CA) represents a central approach “within the scope of empirical pragmatics” (Örnberg Berglund 2009:25) dealing with “the systematic analysis of the talk produced in everyday situations of human interaction” (Hutchby and Wooffitt 2008:11). As a discipline, it started in the early 1960s when Sacks, Schegloff and Jefferson formulated the basic assumptions of CA research, one of which is “to view utterances as actions which

(26)

14

are situated within specific contexts” (Hutchby and Wooffitt 2008:18). According to Hutchby and Wooffitt (2008), at first sight, this approach is similar to Austin (1962) and Searle’s (1969) theory of speech acts; however, there is a fundamental difference between the two, namely, the object of study. They argue that speech act theory focuses on the hypothetical scenarios of language use, whereas conversation analysts are concerned with the instances of authentic language, or “actual utterances in actual contexts” (Hutchby and Wooffitt 2008:18). In other words, CA takes a data-driven approach to the study of social interaction. Another important assumption within CA is the idea that “social contexts are not static but are constantly being formed by the participants through their use of language and the ways in which turn-taking, openings and closures, sequencing of acts and so on are locally managed” (Walsh 2011:84). This means that an ordinary two-party conversation can be analyzed on different levels, which, taken together, provide a deeper understanding of the ways human interaction is organized.

Even though the main application of CA is in the area of spoken language, this approach proves highly useful when dealing with synchronous computer-mediated communication (SCMC). Cameron (2001:9) draws on the similarities between a face-to- face and an online conversation by highlighting the informal nature of both, “[t]he term

‘chat room’ makes an explicit parallel with a certain kind of informal conversation, namely chat”. Based on Heritage (1997), Walsh (2011:85) argues that CA presents an essential toolkit for a description and analysis of a classroom discourse since it views turns in conversation as related to each other. This ‘relatedness’ of turns in a conversation is reflected in the CA term adjacency pairs (Hutchby and Wooffitt 2008), which is frequently applied to the study of conversational coherence (e.g. Örnberg-Berglund 2009; Herring 1999).

Another important CA concept, relevant to the discussion of the qualitative results of the data chapters dealing with text-based and voice-based CMC (see Chapters 5 and 7), is that of preference, or a certain conventionalized organization of turns in interaction, “[f]or example, the preferred response to an assessment is an agreement and the nonpreferred response is a disagreement” (Holtgraves 1992:148, adapted from Pomerantz 1984, 1978).

The traditional CA description of turns as preferred or non-preferred relies on the assumption that the structure of the analyzed conversation is independent from the communicative intentions of the participants. Holtgraves (1992), however, adds a new dimension to the CA term of preference organization by placing it in the context of Brown and Levinson’s (1987) politeness framework (see Section 2.2.1). He argues that if “no face concerns existed, talk probably would be structured differently; probably it would be fast and explicit, unlike human communication as we know it, which is sometimes slow and implicit” (Holtgraves 1992:156).

Also, Walsh (2011:85) highlights the importance of a larger social context by stating that the methods of CA can be used to “uncover something of the detail of interactions by looking at the ways in which contexts are co-created in relation to the goal-oriented activity in which [the participants] are engaged”. As the main focus of this thesis lies on the pragmatic properties of peer-review critique, Walsh’s (2011) emphasis on the idea of “co- creation” of contexts becomes central and will be considered in relation to the findings of

(27)

15

this investigation in Chapter 8, which provides a comparison of pragmatic strategies for peer-reviewing used in four different modes of CMC.

2.2.5 Conversation management: backchannels

The ‘interactive’ nature of talk has been one of the most central objects of study in linguistics (Lambertz 2011; Gardner 2001; Goffman 1976; Sacks et al. 1974). There are different approaches to the purpose and function of short ‘listener’-produced utterances in conversation, commonly referred to as backchannels (Holmes 1995; Yngve 1970) or response tokens and assessments (Gardner 2001; Goodwin 1986). In language pragmatics, the term backchannel emphasizes the relationship between the speaker and the hearer with regard to the conversational floor, “‘main’ channel being the person who is holding the floor – the speaker – and the ‘back’ channel being the addressed recipient of the talk – the listener – who gives information without claiming the floor” (Lambertz 2011:12; White 1989). In CA research, on the other hand, it is the listener’s activities that are the focus of attention, and the functional label assigned to the listener’s responses reflects the extent to which he or she is contributing to the conversation at hand (Gardner 2001:2; Gardner 1994). Within this approach, backchannels constitute a far broader class stretching beyond response tokens (which are traditionally interpreted as backchannels in language pragmatics), and comprised of a number of categories, such as assessments, brief questions and collaborative completions. The list below thus contains the function-based types of backchannels (Gardner 2001:3).

 Response tokens:

Continuers, which function to hand the floor back to the immediately prior speaker (e.g. Mm hm, Uh huh);

Acknowledgements, which claim agreement or understanding of the prior turn (e.g. Mm, Yeah);

Newsmarkers, and newsmarker-like objects, which mark the prior speaker’s turn as newsworthy in some way, (e.g. Really?, the change-of-state token Oh, the ‘idea-connector’ Right);

Change-of-activity tokens, which mark a transition to a new activity or a new topic in the talk (e.g. Okay, Alright).

 Assessments, which evaluate the talk of the prior speakers (e.g. Great, How intriguing, What a load of rubbish);

 Brief questions for clarifications or other types of repair, which seek to clarify mishearings or misunderstandings (e.g. Who?, Which book do you mean?, or the very generalized Huh?);

 Collaborative completions, whereby one speaker finishes a prior speaker’s utterance (e.g. A: So he’s moved into … B: commercial interests);

 Many non-verbal vocalizations and kinesic actions (e.g. sighs, laughter, nods and head shakes)

(adapted from Gardner 2001:2–3).

(28)

16

The term backchannel will be used throughout this thesis to refer to hearer-produced utterances indicating his or her “engaged listenership” (Lamberz 2011:11) in conversation, corresponding to the sub-class of response tokens in Gardner’s (2001) taxonomy. However, Gardner’s (2001) function-based approach presents a useful point of departure for the analysis of conversation management devices employed by the students in my material, and will be revisited in Chapters 5 and 7.

2.3 CMC

r

esearch

The term computer-mediated communication (CMC), used to refer to “[i]nterpersonal interaction (traditionally text-based) via the use of computers or other digital media”, constitutes one of the rapidly developing fields of research (Beers Fägersten et al.

2010:162). Previous studies report that different modes of CMC are associated with different social and educational functions in online collaborative environments (Beers Fägersten et al. 2010; McBride and Beers Fägersten 2008; Xie 2008). Online courses in English at Mid-Sweden University are based on the socio-constructivist pedagogical model (see Section 2.4), emphasizing the student’s individual contribution to the learning process and a collaborative construction of knowledge (cf. Deutschmann 2013; Lyddon and Sydorenko 2010; Comeaux 2005; Davies and Graff 2005). In the context of “interactive e- learning” (Berge 2002), the peer-review discussion constitutes one of the core practices as it promotes “learner-centered collaboration” (Lyddon and Sydorenko 2010). Moreover, it reinforces the communicative aspect of learning and thereby helps “minimize […] [t]he risk of students feeling isolated [, which is] very common in distance learning” (McBride and Fägersten 2010:44; see also Ädel 2011; Huang 2002; Spitzer 1998). Naturally, for the peer- review exchanges to become a stimulating L2 learning experience, the emerging tools for online interaction need to be evaluated in terms of their potential to substitute the physical space of the traditional face-to-face classroom (cf. Skogs 2015). This section outlines a number of relevant studies, which either focus on discourse strategies for expressing (dis)agreement, or aim to evaluate the role of a certain computer-mediated environment in the context of online L2 learning.

2.3.1 Asynchronous text-based and voice-based CMC

As the oldest publicly available type of CMC (cf. FOLDOC, The Free On-Line Dictionary of Computing), asynchronous discussion boards constitute a rich source of material for linguistic research. Within the scope of politeness theory (Brown and Levinson 1987), a number of empirical studies involving corpora of asynchronous written discussions have been carried out. Baym’s (1996) work aims to analyze the structure and use of (dis)agreements in an online asynchronous written environment, represented by a Usenet discussion group on the topic of soap operas. Baym’s (1996:31–32) results suggest that the instances of agreement and disagreement in the data are characterized by a complex structure, i.e. the speech act of (dis)agreement is often followed by an elaboration. Most significant, the members of the Usenet threads examined express agreement by quoting parts of the others’ messages; and even in cases of disagreement, they still tend to

A Linguistic Analysis of Peer-review Critique in Four Modes of Computer-mediated Communication

A Linguistic Analysis of Peer-review Critique in Four Modes of Computer- mediated Communication

Irina Frisk

A Linguistic Analysis of Peer-review Critique in Four Modes of Computer-mediated Communication

To my parents,

Liudmila and Alexander

Abstract

Table of contents

List of tables and figures

Acknowledgments

Chapter 1 Introduction

1.1 Introductory remarks

1.2 Terminology

1.3 Aims and scope

1.4 Thesis outline

Chapter 2 Previous Research

2.1 Introduction

2.2 Linguistic politeness

2.3 CMC

esearch